Blackwell Publishing IncMalden, USA CORGCorporate Governance: An International Review0964-8410Blackwell Publishing Ltd. 2006 2006144••••ORIGINAL ARTICLES TRANSPARENCY AND DISCLOSURE SCORES AND THEIR DETERMINANTS IN THE ISTANBUL STOCK EXCHANGE COPRORATE GOVERNANCE *Address for correspondence: Faculty of Management, Sabanci University, 34952 Orhanli-Tuzla, Istanbul, Tur- key. Tel: +90 (216) 483-9678; Fax: +90 (216) 483-9699; E-mail: [email protected]Transparency and Disclosure Scores and their Determinants in the Istanbul Stock Exchange Mine Aksu* and Arman Kosedag Recent financial reporting and auditing scandals on both sides of the Atlantic have led to a global realisation of the importance of sound corporate governance (CG) practices in alleviating the agency problems in the corporate form of business and for efficient allocation of capital in international markets. Transparency and disclosure (T&D) practices followed by firms are an important component and a leading indicator of CG quality. Transparent and full- disclosure of information is especially vital for Turkey where external capital is necessary to sustain the high growth rate and the biggest agency problem centres on asymmetric information and expropriation by majority shareholders. We collaborate with Standard & Poor’s (S&P) and base our survey on their scoring methodology, a customised version of the 98 desirable T&D attributes they used in several other countries, and their classification of the attributes into three categories: ownership structure and investor relations, financial transparency and information disclosure, and board and management structures and processes. We evaluate the T&D practices of the 52 largest and most liquid firms in the Istanbul Stock Exchange (ISE), based on their English and local language annual reports and websites. Our rankings provide a first time, objective assessment of the corporate disclosure practices of ISE firms and uncover that they are, at best, moderate and vary with respect to the three sub-categories of T&D. We also consider a simple model that sequentially links agency problems to CG/T&D mechanisms in place, which in turn impact firm-level and economy-wide financial perfor- mance. Concentrating on the causal side of the model – the determinants of T&D scores – we provide out-of-sample evidence that firm size, financial performance and market-to-book equity best explain the variation in T&D scores in the ISE. While our results provide consid- erable support for prior findings in developed markets, they also shed light on how specific agency problems faced by ISE firms impact their T&D scores. Keywords: Corporate governance, transparency and disclosure index, Istanbul Stock Exchange, agency conflicts Introduction he recent corporate governance and finan- cial reporting scandals on both sides of the Atlantic have led to a global realisation that sound corporate governance (CG) practices including transparency and full disclosure T (T&D) are important for long-term viability of companies as well as for efficient allocation of capital in the international financial markets. Starting with the Sarbanes-Oxley Act in the US, many countries have experienced an unprece- dented amount of governmental and institu- tional intervention and are now in the process
Transcript
Blackwell Publishing IncMalden USA
CORGCorporate Governance An International Review0964-8410Blackwell Publishing Ltd 20062006144bullbullbullbullORIGINAL ARTICLES
TRANSPARENCY AND DISCLOSURE SCORES AND THEIR DETERMINANTS IN THE ISTANBUL STOCK EXCHANGE
COPRORATE GOVERNANCE
Address for correspondenceFaculty of ManagementSabanci University 34952Orhanli-Tuzla Istanbul Tur-key Tel
+
90 (216) 483-9678Fax
+
90 (216) 483-9699 E-mailmaksusabanciunivedu
Transparency and Disclosure Scores and their Determinants in the Istanbul Stock Exchange
Mine Aksu and Arman Kosedag
Recent financial reporting and auditing scandals on both sides of the Atlantic have ledto a global realisation of the importance of sound corporate governance (CG) practices inalleviating the agency problems in the corporate form of business and for efficient allocationof capital in international markets Transparency and disclosure (TampD) practices followed byfirms are an important component and a leading indicator of CG quality Transparent and full-disclosure of information is especially vital for Turkey where external capital is necessaryto sustain the high growth rate and the biggest agency problem centres on asymmetricinformation and expropriation by majority shareholders
We collaborate with Standard amp Poorrsquos (SampP) and base our survey on their scoringmethodology a customised version of the 98 desirable TampD attributes they used in severalother countries and their classification of the attributes into three categories ownershipstructure and investor relations financial transparency and information disclosure and boardand management structures and processes We evaluate the TampD practices of the 52 largestand most liquid firms in the Istanbul Stock Exchange (ISE) based on their English and locallanguage annual reports and websites Our rankings provide a first time objective assessmentof the corporate disclosure practices of ISE firms and uncover that they are at best moderateand vary with respect to the three sub-categories of TampD
We also consider a simple model that sequentially links agency problems to CGTampDmechanisms in place which in turn impact firm-level and economy-wide financial perfor-mance Concentrating on the causal side of the model ndash
the determinants of TampD scores
ndash weprovide out-of-sample evidence that firm size financial performance and market-to-bookequity best explain the variation in TampD scores in the ISE While our results provide consid-erable support for prior findings in developed markets they also shed light on how specificagency problems faced by ISE firms impact their TampD scores
Keywords Corporate governance transparency and disclosure index Istanbul StockExchange agency conflicts
Introduction
he recent corporate governance and finan-cial reporting scandals on both sides of
the Atlantic have led to a global realisation thatsound corporate governance (CG) practicesincluding transparency and full disclosure
T
(TampD) are important for long-term viability ofcompanies as well as for efficient allocation ofcapital in the international financial marketsStarting with the Sarbanes-Oxley Act in the USmany countries have experienced an unprece-dented amount of governmental and institu-tional intervention and are now in the process
of restructuring their current laws regulationsand enforcement capabilities within the frame-work of best corporate governance practices
1
Furthermore academicians investment anddata-base companies have started to constructand evaluate ethical socially responsible andstrong CG indices
2
and companies themselvesare paying hefty sums of money to have theirCG and TampD practices rated by rating agenciessuch as Standard amp Poorrsquos (SampP) and Institu-tional Shareholder Services (ISS) to signal theirquality in their quest for external capital Par-allel to these developments there has been aproliferation of country-level (La Porta
et al
1997 Shleifer and Wolfenzon 2000 Gugler
et al
2004) and firm-level (see for exampleAshbaugh
et al
2004 Brown and Caylor 2004)evidence that better CG practices are associatedwith lower cost of capital higher firm valuesand thus easier access to external capital
Full disclosure and transparency of financialinformation are vital components of the CGframework (OECD 1999) and are regarded asimportant indicators of CG quality
3
Beeks andBrown (2005) find that firms with higher CGquality make more informative disclosuresSadka (2004) provides theoretical and empiri-cal evidence that the public sharing of finan-cial and analyst reports have enhanced factorproductivity and economic growth in 30countries
These relationships and the expected bene-fits of good CG and TampD practices are es-pecially important for emerging markets likeTurkey because they are in dire need of externalcapital as their economies grow faster than thatof more developed nations As a result it isimportant to investigate whether the historicalscarcity of external capital flow to Turkey isrelated to the low quality of TampD This studywhich provides a first-time objective measure-ment of TampD quality in the Istanbul StockExchange (ISE) in a form suitable for time-series and cross-sectional comparisons includ-ing comparison with other sample countriessurveyed by SampP is a first step in this direction
Turkey is a French origin civil law countrywhere corporate ownership structure can becharacterised by concentrated family owner-ship and low float rates leading to certain typesof agency problems
4
As a result of its highlyconcentrated pyramidal family-based owner-ship structures substantive inter-corporateshareholdings and dual class shares the mainagency conflict centres around the expropria-tion of minority shareholdersrsquo rights by ulti-mate owners with superior information evenin cases where they donrsquot hold a significantamount of the cash-flow rights The ensuinginformation asymmetry between the insidersand outsiders has resulted in low float rates
and dividend payments (for an in-depthdescription of ownership structures of ISEfirms its disadvantages in terms of investmentreturns and development of the capital mar-kets see Yurtoglu 2000 2003) Furthermoredue to the concentration of ownership andpower in influential families and its Frenchlegal origin Turkey has also been slow in thedevelopment and enforcement of CG and TampDprinciples and best practices
The potential for this special kind of agencyconflict and the importance of TampD as a lead-ing CG mechanism in alleviating its draw-backs by reducing the asymmetry betweenthe insiders and outsiders have been animportant motivation of this study Thisagency problem was recently acknowledgedin a task force report of the Institute of Inter-national Finance (4122005 page 1) as thefollowing quote indicates ldquoTurkish compa-nies are often controlled through the use offoundersrsquo shares that carry multiple votingrights andor board nominating rights As aresult the protection of minority shareholderinterests rests primarily on full disclosure andaccurate financial reportingrdquo
This study is the first attempt to create asummary TampD index for the largest and mostliquid firms trading on the ISE We undertookthe project in collaboration with SampP a presti-gious global rating and financial servicescompany that has been carrying out large-scale research since 2001 to score the TampDpractices of a number of large and liquidfirms they follow in the US Latin AmericaEurope Japan Russia and some emergingAsian markets We customised the 98 attrib-utes used in these prior surveys and came upwith a final set of 106 attributes which not onlyreflects the expertise of SampP analysts andresearchers but also the regulatory institu-tional and economic environment in TurkeyWe consider the 106 attributes in three sub-categories again using SampPrsquos classificationscheme (i) Disclosure of ownership structureand investor relations (OwnStr) (ii) Financialtransparency and disclosure in the financialstatements (FinDisc) and (iii) Disclosure of theboard and management structure and pro-cesses (BrdMgmt) Next the annual reportsand the websites (both English and local lan-guage versions) of the 52 largest and most liq-uid firms listed in the ISE were searched forthe inclusion of these information attributesEach firm was then graded and ranked in theabove-mentioned three categories as well as intheir overall TampD quality
5
We find that the transparency and disclo-sure scores of ISE companies are not impres-sive especially in terms of the board andmanagement structure and processes Never-
theless considerably high scores (710) areobtained in the category of financial trans-parency and information disclosure and theaverage overall TampD score (TDS) of Turkishcompanies is 510 a score comparable tomany countries in continental Europe
In this paper we draw on the agency theoryof the firm and model CG and TampD as a sys-tem of checks and balances that has evolvedto mitigate the agency problems often encoun-tered in the corporate form Our expectation isthat if companies have effective mechanismsin place the agency problems ndash such as ex-traction of private benefits by majorityshareholders consumption of perquisites byentrenched management under or overinvest-ment ndash and their associated costs will be miti-gated We base our analysis on this body ofresearch for the second objective of the studyexploring the cross-sectional differences inquality of disclosure in ISE firms We find thatlarger and more profitable firms as well asfirms with higher information asymmetryexhibit higher TDS to partially resolve the con-flict with minority shareholders and regula-tory institutions while free cash flow availableto management and leverage do not haveexplanatory power This is not surprising sincethe conflict between the shareholders andmanagers and creditors is not a major agencyproblem in Turkey
The study has several contributions andis expected to be of interest to regulatorsresearchers managers and market partici-pants The rankings provide a first time objec-tive quantitative and verifiable assessment ofcorporate disclosure practices of listed compa-nies in the ISE encompassing not only financialstatement disclosure but also disclosures ofshareholderrsquos rights and board and manage-ment structuresprocesses As such it isexpected to be a benchmark for future longi-tudinal and cross-country studies Moreoversince we use the same attributes and method-ology of prior SampP studies the findingsallow comparison with the scores previouslyobtained in other markets The rankings basedon a universal benchmark are expected to helpinvestors and other stakeholders in theirdecision making as they allow an objectiveassessment of differences in the level of report-ing quality and hence investment risks acrossfirms markets and sectors The TampD rankingscan also be used as a meaningful objectivemonitoring and enforcement tool by local andinternational regulators Finally this is thefirst study that investigates the relationshipbetween TampD and some agency conflictproxies prevalent in the ISE The resultsprovide out-of-sample evidence that thesefirm-specific characteristics do a good job in
explaining the variation in TDS as in someother countries
The remaining part of the paper is organisedas follows the next section describes the poorculture and regulation of disclosure in Turkeyafter which we discuss our simple model ofthe causes and the consequences of the CGTampD framework and develop our hypothesisWe then discuss prior research and the choiceof our agency conflict proxies used as deter-minants of the TampD scores The sample datarequirements and methods of analysis aredescribed followed by the results The finalsection concludes and presents some futureresearch ideas
The poor regulatory framework and culture of CG and TampD in Turkey
Turkey has historically had a poor culture ofCG and TampD It has always been a civil lawcountry and has had a very late start in theliberalisation of its economy the establishmentof its stock market (1986) and financial report-ing standards based on generally acceptedaccounting principles and the concept of fulland fair disclosure (1994)
6
The first legislation concerning commerciallaw was enacted for the first time in theOttoman Empire in 1850 and was based onthe French Code of Commerce of 1807 Underthe legal reforms of 1926 this was replaced bya system based on Swiss Civil Code and Codeof Obligations the Italian Penal Code andthe German Code of Penal Procedure Neitherthese nor the present Turkish Code of Com-merce dating back to 1957 were based on thegenerally accepted principles of accountingand auditing and hence did not regulate finan-cial reporting They highlight the strength ofthe civil law tradition weak enforcement ofrules and lack of a disclosure philosophy inthe Turkish business culture
This legal and cultural infrastructure seemsto be an important factor in the historic diffi-culty with which Turkish firms have attractedexternal capital Table 1 reconstructed fromtables 2 and 8 in La Porta
et al
(1997) placesTurkey in the French origin legal systemwhich has enjoyed the lowest access to ex-ternal capital markets as well as the weakestrule of law and investor rights protectionamong the three other civil law and the com-mon law traditions The table also depicts thatwhile GDP growth is higher in Turkey than inthe other legal origins access to external cap-ital and measures of investor rights are lowerthan even the average of the French origincountries Hence while the fast growth rate inthe Turkish economy underscores the need to
access external debt and equity capital thetable highlights its difficulty in countrieswhere enforcement of the rule of law and pro-tection of investorsrsquo rights are weak The solu-tion to this dilemma through observance ofgood CG and TampD practices has been a majormotivation for this study
Recent studies have observed that the weakCG and disclosure tradition in Turkey havecontinued until very recently Ararat and Ugur(2003) describe Turkeyrsquos civil law tradition
and its inefficient and inconsistent regulatoryframework and the ensuing paucity of the ruleof law and its enforcement They also elabo-rate on the specific CG problems faced byTurkish firms the concentrated and pyramidalownership structure dominated by familiesownership of many banks by these groupcompanies inconsistent and opaque account-ing and tax regulations and the investor mis-information caused by the absence of inflationand consolidation accounting
7
As a result of
Table 1 Access to external financing and measures of investor protection in Turkey and by legal origin
Panel A
a
Aggregate mean values for the Turkish economy and stock market and comparison with different legal origins
(1977) Table 2 presented for 49 countries and by legal origin
b
Panels C and D are reconstructed from La Porta
et al
(1977) Table 8 presented for 49 countries and by legal origin All values aretaken from the Worldscope database
this infrastructure agency conflicts are con-centrated around weak minority shareholdersrsquoand creditorsrsquo rights inconsistent and opaquedisclosure policies and lack of a culture ofvoluntary disclosure and convergence (in-separability) of ownership and management
8
These create an environment that fosters cor-ruption share dilution asset stripping tunnel-ling insider trading and market manipulation
Given these specific agency problems fos-tered by the ownership and regulatory struc-ture in large public Turkish firms we concen-trate on the disclosure and transparencycomponents of CG which are the mechanismsmost likely to improve the protection of minor-ity shareholdersrsquo and the creditorsrsquo rightsMcKinsey (2002) has carried out a global sur-vey in which institutional investors were askedto identify their CG priorities for companiesand policy makers The survey administeredin 2002 covers 201 responses from institutionalinvestors located in 31 countries Table 2reports the weights assigned by global institu-tional investors to the top seven most impor-tant CG factors for their investment decisionsin Turkey The survey results depict a patternsimilar to the average of all countries sur-veyed timely broad understandable account-ing disclosure and safeguarding of the share-holdersrsquo rights are the most important CGpriorities of institutional investors in theirdecisions to invest in Turkish companies
Unfortunately Turkey has not scored highin the scarce prior disclosure studies eitheras will be described below We believe thatthe aforementioned bleak picture is changingfast as a result of the reforms in the financialreporting environment the recent stability inTurkish politics economic reforms new regu-lation and more effective enforcement and inthis study we hope to quantify the expectedimprovement in disclosure practices of ISEfirms Some of these developments inlcude thefollowing
(a) accelerated privatisation of State Eco-nomic Enterprises
(b) the creation of private pension funds thatare expected to serve as institutionalinvestors and enhance monitoring inpublic firms
(c) monitoring of banks by the Banking Regu-lation and Supervison Agency (BRSA)
(d) the Capital Markets Boardrsquos (CMBrsquos)commitment to improvement of the regu-latory framework
(e) the CMBrsquos Corporate Governance Prin-ciples promulgated on a comply or ex-plain basis and covering areas of dis-closure minority rights board structureand management oversight The extentof compliance and the reasons for non-compliance to the principles would beindicated in the 2004 annual reports pub-lished in 2005
(f) the establishment of the Turkish Account-ing Standards Board and their mandate toconverge the local accounting standardsto the International Financial ReportingStandards (IFRS) by 2005 in line with thedirectives of the European Commission
(g) Of course Turkeyrsquos recent progress inachieving full membership of the EU willprovide the strongest inertia in establish-ing the ldquorule of lawrdquo and improved CGand TampD practices in Turkey
A model of incentives for good CGTampD and our main hypothesis
Our basic model of the relationships betweenthe conflicts of interest and information asym-metry between the stakeholders in the corpo-rate form CGTampD practices as mechanismsof checks and balances put in place to mitigatethem firm performance and the localglobaleconomy is depicted in Figure 1 In this paperwe explore the left-hand side of the model ie
Table 2 CG priorities of institutional investors for their investment decisions in 30 countries
CG factors Turkey () All 30 countries surveyed ()
Accounting disclosure 73 52Shareholder equality 64 47Market regulation and infrastructure 55 43Takeover markets 45 23Property rights 45 46Credit information 36 29Board independence 27 44
Sources McKinsey amp Company (2002) Global Investor Opinion Survey on Corporate Governance andArarat and Ugur (2003)
the quality of TampD practices of ISE firms andthe firm-specific agency-conflict related deter-minants of their TampD scores
Based on the agency theory of the firm andviewing the firm as a nexus of contracts proneto information asymmetry and conflicts ofinterest between its stakeholders (Jensen andMeckling 1976) we envision good CGTampDpractices as mechanisms of checks and bal-ances that have evolved to mitigate theseagency problems
9
Such best practices areexpected to lower the agency costs leading tohigher firm performance In turn the CGTampDpractices themselves are shaped by the legaltraditions and regulations in the firmrsquos countryof domicile as well as by international lawsand standards and the global economy Notonly do these impose new CGTampD rules andbest practices on the firm ndash such as the IFRS orCG principles of the OECD ndash but they alsocreate new opportunities in the form of positiveNPV projects and ease of access to foreigncapital on the one hand and new threats suchas regulatory intervention and litigation dueto perhaps non-compliance or misrepresenta-tion Since undertaking positive NPV projectsdepends on the firmrsquos ability to tap the equityand debt markets and since capital will flowto the highest return investments for a givenrisk level firms have the incentives to maxi-mise the return to their capital suppliers whileminimising their perceived risk by putting inplace optimal CG mechanisms Good TampDmechanisms are thus set in place to protect therights of shareholders creditors and other out-siders from extraction of private benefits byinsiders based on their superior informationThis is expected to minimise the informationalasymmetry increase investor awareness andtrust which in turn should reduce the uncer-tainty of the returns to capital suppliers andlead to lower cost of capital and hence higherfirm value (see for example Ashbaugh
et al
2004 Brown and Caylor 2004 Berglof and
Pajuste 2005)
10
This back and forth flowbetween the economic units and capital sup-pliers can be sustained only in the absence ofappropriation by insiders and other commonagency problems Thus in our model best CGTampD practices are used by firms not only tosignal their quality but also as monitoring toolsto maintain their quality
11
This simple model of incentives for goodCGTampD practices has been well-studied andreviewed extensively (see for example Wattsand Zimmerman 1986 Healy and Palepu2001 Core 2001 Rahman 2002) in literatureA recent example that draws on this literatureis the Albuquerque and Wang (2005) studywhich develops a dynamic general equilib-rium model to acknowledge the implicationsof agency conflicts through weak investorprotection for security prices They find thatcountries with weaker investor protectionhave higher overinvestment lower valuesand higher expected returns return volatilityand interest rates Ashbaugh
et al
(2004)discuss the agency problems created by theseparation of ownership and control andconclude that firms with better governancepresent less agency risk to shareholders lead-ing to lower cost of equity capital Accord-ingly we expect to observe higher quality CGTampD practices in firms that have solved or thatare less susceptible to or that have committedthemselves to reduce their agency conflictsAccordingly we pose the following hypoth-esis presented in the alternative form
H
1
Firms that have lower potential agency con-flicts are expected to have higher TampD scores
Prior research and the choice of the agency conflict proxies
Since this paper adapts an index of disclosureattributes to measure the TampD quality in a
Figure 1 A model of incentives for sound CGTampD practices
FIRM Nexus of contracts between stakeholders
CGTampD mechanisms
FIRM
Accounting Performance
Market Performance
COUNTRY
Local Economy
GLOBAL
Economy
specific country and also explores the deter-minants of the TampD scores obtained it isclosely related to both of these vast strands ofdisclosure literature Such indices have beendeveloped in several countries in the past 40years mostly carried out by rating firmsfinancial analysts academicians and databasecompanies They are indeed far too many tolist here Some examples are the disclosurequality evaluations published by the Asso-ciation for Investment Management andResearch (AIMR) over the decade 1980ndash90analystsrsquo ratings of firmsrsquo disclosure policiesdisclosure quality evaluations published bythe
Financial Post
and the country-based indexof accounting quality and transparency devel-oped by the Center for International FinancialAnalysis and Research (the CIFAR Index) in1993 and 1995 This latter index represents theaverage number of 90 items included in theannual reports of a sample of companiescovering 42 countries It has been used bymany academicians in their cross-countryresearch to uncover the relationship betweenaccounting transparency legal origin cultureand economic performance (see for exampleLa Porta
et al
1998 Hope 2003) To ourknowledge the CIFAR Index is the earliestdisclosure study that includes Turkey in itssample The average score of the 10 Turkishcompanies is measured as 586 the secondlowest after Portugal (537) among the 25code-law countries and after India (54) amongthe 17 common-law countries included in theCIFAR Index Of course this survey was car-ried out in 1995 and Turkey has come a longway since then
The more recent Patel
et al
(2002) studywhich uses the TampD scores of SampP measuredat the end of 2000 covers 354 firms in 19 emerg-ing markets The study includes only 12Turkish firms covering only 64 percent of themarket capitalisation of Turkish companiesincluded in the SampPInternational FinanceCorporation (IFC) Index As of the end of 2000the average TampD score of these 12 firms is 34the third lowest after the five Latin Americancountries and the Philippines ndash both groupsexhibiting an average score of 29
More closely related to this study are therecent surveys conducted by SampP to measurea broader-based firm-specific TampD quality inmany regions of the world often in collabora-tion with academicians in respective countriesIn 2002 SampP Governance Services publishedits first TampD study that includes companies inthe following SampPIFC indices Asia LatinAmerica Asia Pacific 100 TOPIX 150 (Japan)Then came the surveys in Russia in 20022003 and as a result of continued interestamong investors in 2004 Furthermore in
2003 SampP released its study of the SampP Europe350 companies
12
Our main objective in thisstudy is to quantify the improvement in dis-closure practices of Turkish firms in the recentyears by using a much larger sample (covering100 percent of SampPIFC Index market capital-isation of Turkish firms) and a more compre-hensive set of attributes encompassing notonly financial statement disclosure but alsodisclosure of other aspects of CG ie owner-ship board and management structures andprocesses and covering the disclosure inannual reports and company web sitesAccordingly we collaborated with SampP tocarry out the Turkish survey Our and theSampPrsquos roles in this study are explained atthe beginning of the next section SampP and theCorporate Governance Forum of Turkey(CGFT) of Sabanci University announced theresults to the public in a press release in May2002 and SampP officially published the resultsin their newsletter Ratings Direct dated 662005
13
This research is also closely related to alarge body of accounting and finance litera-ture that investigates the causes and conse-quences of voluntary disclosure In this paperwe limit our investigation to the agencytheory-based determinants of discretionarydisclosure reviewed in seminal studies suchas Verrecchia (1983 2001) Healy and Palepu(2001) and Core (2001) The theory predictsthat even though self-serving managers haveincentives to disclose their insider informa-tion they will withhold private informationwhen disclosure is too costly and that there isa strong country effect in what companiesdisclose (for empirical evidence on country-based differences see for example Berglofand Pajuste 2005 Hope
et al
2003) Priorresearch has shown that there is also consider-able within-country variation in disclosuredriven by firm characteristics
We draw on agency research studies onfirm-specific determinants of disclosure andwe consider the specific agency problemsfaced by Turkish firms in our choice of therelevant agency conflict proxies There existsan extensive body of prior research on deter-minants of disclosure that has accumulatedover the past 40 years a review of which isbeyond the scope of this paper Here we willjust mention a few important or recentpapers
14
The variables we explore undertwo categories of agency conflicts are sizemeasured as market value of equity (MVE)market-to-book value of equity ratio (MTBmeasured as MVEBVE) firm performancemeasured as earnings before interest andtaxestotal assets (ROA) and net incomeownerrsquos equity (ROE) leverage measured as
total liabilitiestotal assets (TLTA) and freecash flow (FCF)
Proxies for conflict with minority shareholders and regulatory institutions
Size and profitability
Watts and Zimmerman (1986) suggest thatlarger companies are more visible and thusthey are politically more sensitive We expectlarger and more profitable firms to havehigher TDS because they are closely followedby financial intermediaries have more com-prehensive disclosure standards in place tominimise the political costs of noncompliancewith generally accepted accounting prin-ciples (GAAP) and can better afford the costof voluntary disclosure Furthermore profit-able firms are likely to have less to hide fromtheir constituents and thus are expected toemploy less earnings management
15
Manyempirical studies have associated disclosurequantity and quality measured by a disclo-sure index with firm size and performanceand many have investigated the relationshipbetween firm characteristics and agencyproblems In their well-cited empirical workLang and Lundholm (1993) find that analystratings of disclosure are higher for firms thatperform well for larger firms firms witha weaker relation between annual stockreturns and earnings and firms that issuesecurities Doyle
et al
(2005) and Bryan andLilien (2005) document that material weak-nesses in internal control are more likely forsmaller younger financially weaker fastgrowth firms and worse performers Simi-larly using voluntary segment disclosureas a proxy for transparency Sidney andBertrand (2004) find that concentrated boardownership results in low disclosure and thiseffect is stronger when firm performance ispoor Black
et al
(2006) investigates the cross-sectional differences in Korean firmsrsquo CGpractices and again finds that firm size riskand long-term profitability and need forequity capital are positively related to betterCG Hope (2003) observes a positive correla-tion between firm size and the CIFAR indexfor annual report disclosure SimilarlyHossain
et al
(2005) has found that voluntarydisclosure of prospective information isrelated to firm size Veronina
et al
(2005) is arecent study which is similar to the presentone in that it measures the transparencypractices of 102 listed Russian firms in 2001and also investigates the cross-sectional dif-ferences in their transparency scores Usinga checklist of 441 items from InternationalAccounting Standards (IAS) they find that
use of a Big-5 auditor foreign listing sizegovernment shareholdings and indepen-dence of CEO and board chair are associatedwith transparency
Market-to-book ratio
MTB ratio has been used in the literature as aproxy for risk growth potential unreportedintangibles and firm prospects assessed bymarket participants A low ratio is associatedwith low growth potential and high free cashflows under the discretion of insiders Suchfirms have little need for external finance andthus voluntary disclosure (Core 2001) There-fore within the agency theory framework oneshould expect to see a positive relationshipbetween MTB and TampD scores For exampleBerglof and Pajuste (2005) report that moreinformation is publicly available in largerfirms firms with lower leverage higherfinancial performance higher market-to-bookratios and more concentrated ownership Onthe other hand a low MTB can also be con-sidered a sign of undervaluation by themarket The equityrsquos market value might below relative to its book value not because ofthe firmrsquos low growth potential but simplybecause future prospects of the firm are notproperly communicated to the public or thereis a general undervaluation in the market dueto local economic uncertainties For examplethe cash flows from valuable intangible assetssuch as human resources internally generatedpatents and RampD activities which are notrecognised in financial statements in line withGAAP may not have been impounded inprices in an inefficient information setting Insuch cases management may take actions toincrease transparency and put better CGpractices in action to remedy this unwantedperception and its negative effect on firmvalue Furthermore low MTB may be theresult of an increase in book value of equitydue to recently issued capital and severalresearchers have found that firms disclosemore when they have recently issued capital(see for example Frankel
et al
1995) Con-trary to the predictions of agency theory thisargument suggests a negative relationshipbetween MTB and TampD scores
16
Proxies for conflict between the shareholders and managers and creditors
Leverage
Agency theory also suggests a strong linkbetween leverage and disclosure (Jensen andMeckling 1976) In highly levered firms there
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
of restructuring their current laws regulationsand enforcement capabilities within the frame-work of best corporate governance practices
1
Furthermore academicians investment anddata-base companies have started to constructand evaluate ethical socially responsible andstrong CG indices
2
and companies themselvesare paying hefty sums of money to have theirCG and TampD practices rated by rating agenciessuch as Standard amp Poorrsquos (SampP) and Institu-tional Shareholder Services (ISS) to signal theirquality in their quest for external capital Par-allel to these developments there has been aproliferation of country-level (La Porta
et al
1997 Shleifer and Wolfenzon 2000 Gugler
et al
2004) and firm-level (see for exampleAshbaugh
et al
2004 Brown and Caylor 2004)evidence that better CG practices are associatedwith lower cost of capital higher firm valuesand thus easier access to external capital
Full disclosure and transparency of financialinformation are vital components of the CGframework (OECD 1999) and are regarded asimportant indicators of CG quality
3
Beeks andBrown (2005) find that firms with higher CGquality make more informative disclosuresSadka (2004) provides theoretical and empiri-cal evidence that the public sharing of finan-cial and analyst reports have enhanced factorproductivity and economic growth in 30countries
These relationships and the expected bene-fits of good CG and TampD practices are es-pecially important for emerging markets likeTurkey because they are in dire need of externalcapital as their economies grow faster than thatof more developed nations As a result it isimportant to investigate whether the historicalscarcity of external capital flow to Turkey isrelated to the low quality of TampD This studywhich provides a first-time objective measure-ment of TampD quality in the Istanbul StockExchange (ISE) in a form suitable for time-series and cross-sectional comparisons includ-ing comparison with other sample countriessurveyed by SampP is a first step in this direction
Turkey is a French origin civil law countrywhere corporate ownership structure can becharacterised by concentrated family owner-ship and low float rates leading to certain typesof agency problems
4
As a result of its highlyconcentrated pyramidal family-based owner-ship structures substantive inter-corporateshareholdings and dual class shares the mainagency conflict centres around the expropria-tion of minority shareholdersrsquo rights by ulti-mate owners with superior information evenin cases where they donrsquot hold a significantamount of the cash-flow rights The ensuinginformation asymmetry between the insidersand outsiders has resulted in low float rates
and dividend payments (for an in-depthdescription of ownership structures of ISEfirms its disadvantages in terms of investmentreturns and development of the capital mar-kets see Yurtoglu 2000 2003) Furthermoredue to the concentration of ownership andpower in influential families and its Frenchlegal origin Turkey has also been slow in thedevelopment and enforcement of CG and TampDprinciples and best practices
The potential for this special kind of agencyconflict and the importance of TampD as a lead-ing CG mechanism in alleviating its draw-backs by reducing the asymmetry betweenthe insiders and outsiders have been animportant motivation of this study Thisagency problem was recently acknowledgedin a task force report of the Institute of Inter-national Finance (4122005 page 1) as thefollowing quote indicates ldquoTurkish compa-nies are often controlled through the use offoundersrsquo shares that carry multiple votingrights andor board nominating rights As aresult the protection of minority shareholderinterests rests primarily on full disclosure andaccurate financial reportingrdquo
This study is the first attempt to create asummary TampD index for the largest and mostliquid firms trading on the ISE We undertookthe project in collaboration with SampP a presti-gious global rating and financial servicescompany that has been carrying out large-scale research since 2001 to score the TampDpractices of a number of large and liquidfirms they follow in the US Latin AmericaEurope Japan Russia and some emergingAsian markets We customised the 98 attrib-utes used in these prior surveys and came upwith a final set of 106 attributes which not onlyreflects the expertise of SampP analysts andresearchers but also the regulatory institu-tional and economic environment in TurkeyWe consider the 106 attributes in three sub-categories again using SampPrsquos classificationscheme (i) Disclosure of ownership structureand investor relations (OwnStr) (ii) Financialtransparency and disclosure in the financialstatements (FinDisc) and (iii) Disclosure of theboard and management structure and pro-cesses (BrdMgmt) Next the annual reportsand the websites (both English and local lan-guage versions) of the 52 largest and most liq-uid firms listed in the ISE were searched forthe inclusion of these information attributesEach firm was then graded and ranked in theabove-mentioned three categories as well as intheir overall TampD quality
5
We find that the transparency and disclo-sure scores of ISE companies are not impres-sive especially in terms of the board andmanagement structure and processes Never-
theless considerably high scores (710) areobtained in the category of financial trans-parency and information disclosure and theaverage overall TampD score (TDS) of Turkishcompanies is 510 a score comparable tomany countries in continental Europe
In this paper we draw on the agency theoryof the firm and model CG and TampD as a sys-tem of checks and balances that has evolvedto mitigate the agency problems often encoun-tered in the corporate form Our expectation isthat if companies have effective mechanismsin place the agency problems ndash such as ex-traction of private benefits by majorityshareholders consumption of perquisites byentrenched management under or overinvest-ment ndash and their associated costs will be miti-gated We base our analysis on this body ofresearch for the second objective of the studyexploring the cross-sectional differences inquality of disclosure in ISE firms We find thatlarger and more profitable firms as well asfirms with higher information asymmetryexhibit higher TDS to partially resolve the con-flict with minority shareholders and regula-tory institutions while free cash flow availableto management and leverage do not haveexplanatory power This is not surprising sincethe conflict between the shareholders andmanagers and creditors is not a major agencyproblem in Turkey
The study has several contributions andis expected to be of interest to regulatorsresearchers managers and market partici-pants The rankings provide a first time objec-tive quantitative and verifiable assessment ofcorporate disclosure practices of listed compa-nies in the ISE encompassing not only financialstatement disclosure but also disclosures ofshareholderrsquos rights and board and manage-ment structuresprocesses As such it isexpected to be a benchmark for future longi-tudinal and cross-country studies Moreoversince we use the same attributes and method-ology of prior SampP studies the findingsallow comparison with the scores previouslyobtained in other markets The rankings basedon a universal benchmark are expected to helpinvestors and other stakeholders in theirdecision making as they allow an objectiveassessment of differences in the level of report-ing quality and hence investment risks acrossfirms markets and sectors The TampD rankingscan also be used as a meaningful objectivemonitoring and enforcement tool by local andinternational regulators Finally this is thefirst study that investigates the relationshipbetween TampD and some agency conflictproxies prevalent in the ISE The resultsprovide out-of-sample evidence that thesefirm-specific characteristics do a good job in
explaining the variation in TDS as in someother countries
The remaining part of the paper is organisedas follows the next section describes the poorculture and regulation of disclosure in Turkeyafter which we discuss our simple model ofthe causes and the consequences of the CGTampD framework and develop our hypothesisWe then discuss prior research and the choiceof our agency conflict proxies used as deter-minants of the TampD scores The sample datarequirements and methods of analysis aredescribed followed by the results The finalsection concludes and presents some futureresearch ideas
The poor regulatory framework and culture of CG and TampD in Turkey
Turkey has historically had a poor culture ofCG and TampD It has always been a civil lawcountry and has had a very late start in theliberalisation of its economy the establishmentof its stock market (1986) and financial report-ing standards based on generally acceptedaccounting principles and the concept of fulland fair disclosure (1994)
6
The first legislation concerning commerciallaw was enacted for the first time in theOttoman Empire in 1850 and was based onthe French Code of Commerce of 1807 Underthe legal reforms of 1926 this was replaced bya system based on Swiss Civil Code and Codeof Obligations the Italian Penal Code andthe German Code of Penal Procedure Neitherthese nor the present Turkish Code of Com-merce dating back to 1957 were based on thegenerally accepted principles of accountingand auditing and hence did not regulate finan-cial reporting They highlight the strength ofthe civil law tradition weak enforcement ofrules and lack of a disclosure philosophy inthe Turkish business culture
This legal and cultural infrastructure seemsto be an important factor in the historic diffi-culty with which Turkish firms have attractedexternal capital Table 1 reconstructed fromtables 2 and 8 in La Porta
et al
(1997) placesTurkey in the French origin legal systemwhich has enjoyed the lowest access to ex-ternal capital markets as well as the weakestrule of law and investor rights protectionamong the three other civil law and the com-mon law traditions The table also depicts thatwhile GDP growth is higher in Turkey than inthe other legal origins access to external cap-ital and measures of investor rights are lowerthan even the average of the French origincountries Hence while the fast growth rate inthe Turkish economy underscores the need to
access external debt and equity capital thetable highlights its difficulty in countrieswhere enforcement of the rule of law and pro-tection of investorsrsquo rights are weak The solu-tion to this dilemma through observance ofgood CG and TampD practices has been a majormotivation for this study
Recent studies have observed that the weakCG and disclosure tradition in Turkey havecontinued until very recently Ararat and Ugur(2003) describe Turkeyrsquos civil law tradition
and its inefficient and inconsistent regulatoryframework and the ensuing paucity of the ruleof law and its enforcement They also elabo-rate on the specific CG problems faced byTurkish firms the concentrated and pyramidalownership structure dominated by familiesownership of many banks by these groupcompanies inconsistent and opaque account-ing and tax regulations and the investor mis-information caused by the absence of inflationand consolidation accounting
7
As a result of
Table 1 Access to external financing and measures of investor protection in Turkey and by legal origin
Panel A
a
Aggregate mean values for the Turkish economy and stock market and comparison with different legal origins
(1977) Table 2 presented for 49 countries and by legal origin
b
Panels C and D are reconstructed from La Porta
et al
(1977) Table 8 presented for 49 countries and by legal origin All values aretaken from the Worldscope database
this infrastructure agency conflicts are con-centrated around weak minority shareholdersrsquoand creditorsrsquo rights inconsistent and opaquedisclosure policies and lack of a culture ofvoluntary disclosure and convergence (in-separability) of ownership and management
8
These create an environment that fosters cor-ruption share dilution asset stripping tunnel-ling insider trading and market manipulation
Given these specific agency problems fos-tered by the ownership and regulatory struc-ture in large public Turkish firms we concen-trate on the disclosure and transparencycomponents of CG which are the mechanismsmost likely to improve the protection of minor-ity shareholdersrsquo and the creditorsrsquo rightsMcKinsey (2002) has carried out a global sur-vey in which institutional investors were askedto identify their CG priorities for companiesand policy makers The survey administeredin 2002 covers 201 responses from institutionalinvestors located in 31 countries Table 2reports the weights assigned by global institu-tional investors to the top seven most impor-tant CG factors for their investment decisionsin Turkey The survey results depict a patternsimilar to the average of all countries sur-veyed timely broad understandable account-ing disclosure and safeguarding of the share-holdersrsquo rights are the most important CGpriorities of institutional investors in theirdecisions to invest in Turkish companies
Unfortunately Turkey has not scored highin the scarce prior disclosure studies eitheras will be described below We believe thatthe aforementioned bleak picture is changingfast as a result of the reforms in the financialreporting environment the recent stability inTurkish politics economic reforms new regu-lation and more effective enforcement and inthis study we hope to quantify the expectedimprovement in disclosure practices of ISEfirms Some of these developments inlcude thefollowing
(a) accelerated privatisation of State Eco-nomic Enterprises
(b) the creation of private pension funds thatare expected to serve as institutionalinvestors and enhance monitoring inpublic firms
(c) monitoring of banks by the Banking Regu-lation and Supervison Agency (BRSA)
(d) the Capital Markets Boardrsquos (CMBrsquos)commitment to improvement of the regu-latory framework
(e) the CMBrsquos Corporate Governance Prin-ciples promulgated on a comply or ex-plain basis and covering areas of dis-closure minority rights board structureand management oversight The extentof compliance and the reasons for non-compliance to the principles would beindicated in the 2004 annual reports pub-lished in 2005
(f) the establishment of the Turkish Account-ing Standards Board and their mandate toconverge the local accounting standardsto the International Financial ReportingStandards (IFRS) by 2005 in line with thedirectives of the European Commission
(g) Of course Turkeyrsquos recent progress inachieving full membership of the EU willprovide the strongest inertia in establish-ing the ldquorule of lawrdquo and improved CGand TampD practices in Turkey
A model of incentives for good CGTampD and our main hypothesis
Our basic model of the relationships betweenthe conflicts of interest and information asym-metry between the stakeholders in the corpo-rate form CGTampD practices as mechanismsof checks and balances put in place to mitigatethem firm performance and the localglobaleconomy is depicted in Figure 1 In this paperwe explore the left-hand side of the model ie
Table 2 CG priorities of institutional investors for their investment decisions in 30 countries
CG factors Turkey () All 30 countries surveyed ()
Accounting disclosure 73 52Shareholder equality 64 47Market regulation and infrastructure 55 43Takeover markets 45 23Property rights 45 46Credit information 36 29Board independence 27 44
Sources McKinsey amp Company (2002) Global Investor Opinion Survey on Corporate Governance andArarat and Ugur (2003)
the quality of TampD practices of ISE firms andthe firm-specific agency-conflict related deter-minants of their TampD scores
Based on the agency theory of the firm andviewing the firm as a nexus of contracts proneto information asymmetry and conflicts ofinterest between its stakeholders (Jensen andMeckling 1976) we envision good CGTampDpractices as mechanisms of checks and bal-ances that have evolved to mitigate theseagency problems
9
Such best practices areexpected to lower the agency costs leading tohigher firm performance In turn the CGTampDpractices themselves are shaped by the legaltraditions and regulations in the firmrsquos countryof domicile as well as by international lawsand standards and the global economy Notonly do these impose new CGTampD rules andbest practices on the firm ndash such as the IFRS orCG principles of the OECD ndash but they alsocreate new opportunities in the form of positiveNPV projects and ease of access to foreigncapital on the one hand and new threats suchas regulatory intervention and litigation dueto perhaps non-compliance or misrepresenta-tion Since undertaking positive NPV projectsdepends on the firmrsquos ability to tap the equityand debt markets and since capital will flowto the highest return investments for a givenrisk level firms have the incentives to maxi-mise the return to their capital suppliers whileminimising their perceived risk by putting inplace optimal CG mechanisms Good TampDmechanisms are thus set in place to protect therights of shareholders creditors and other out-siders from extraction of private benefits byinsiders based on their superior informationThis is expected to minimise the informationalasymmetry increase investor awareness andtrust which in turn should reduce the uncer-tainty of the returns to capital suppliers andlead to lower cost of capital and hence higherfirm value (see for example Ashbaugh
et al
2004 Brown and Caylor 2004 Berglof and
Pajuste 2005)
10
This back and forth flowbetween the economic units and capital sup-pliers can be sustained only in the absence ofappropriation by insiders and other commonagency problems Thus in our model best CGTampD practices are used by firms not only tosignal their quality but also as monitoring toolsto maintain their quality
11
This simple model of incentives for goodCGTampD practices has been well-studied andreviewed extensively (see for example Wattsand Zimmerman 1986 Healy and Palepu2001 Core 2001 Rahman 2002) in literatureA recent example that draws on this literatureis the Albuquerque and Wang (2005) studywhich develops a dynamic general equilib-rium model to acknowledge the implicationsof agency conflicts through weak investorprotection for security prices They find thatcountries with weaker investor protectionhave higher overinvestment lower valuesand higher expected returns return volatilityand interest rates Ashbaugh
et al
(2004)discuss the agency problems created by theseparation of ownership and control andconclude that firms with better governancepresent less agency risk to shareholders lead-ing to lower cost of equity capital Accord-ingly we expect to observe higher quality CGTampD practices in firms that have solved or thatare less susceptible to or that have committedthemselves to reduce their agency conflictsAccordingly we pose the following hypoth-esis presented in the alternative form
H
1
Firms that have lower potential agency con-flicts are expected to have higher TampD scores
Prior research and the choice of the agency conflict proxies
Since this paper adapts an index of disclosureattributes to measure the TampD quality in a
Figure 1 A model of incentives for sound CGTampD practices
FIRM Nexus of contracts between stakeholders
CGTampD mechanisms
FIRM
Accounting Performance
Market Performance
COUNTRY
Local Economy
GLOBAL
Economy
specific country and also explores the deter-minants of the TampD scores obtained it isclosely related to both of these vast strands ofdisclosure literature Such indices have beendeveloped in several countries in the past 40years mostly carried out by rating firmsfinancial analysts academicians and databasecompanies They are indeed far too many tolist here Some examples are the disclosurequality evaluations published by the Asso-ciation for Investment Management andResearch (AIMR) over the decade 1980ndash90analystsrsquo ratings of firmsrsquo disclosure policiesdisclosure quality evaluations published bythe
Financial Post
and the country-based indexof accounting quality and transparency devel-oped by the Center for International FinancialAnalysis and Research (the CIFAR Index) in1993 and 1995 This latter index represents theaverage number of 90 items included in theannual reports of a sample of companiescovering 42 countries It has been used bymany academicians in their cross-countryresearch to uncover the relationship betweenaccounting transparency legal origin cultureand economic performance (see for exampleLa Porta
et al
1998 Hope 2003) To ourknowledge the CIFAR Index is the earliestdisclosure study that includes Turkey in itssample The average score of the 10 Turkishcompanies is measured as 586 the secondlowest after Portugal (537) among the 25code-law countries and after India (54) amongthe 17 common-law countries included in theCIFAR Index Of course this survey was car-ried out in 1995 and Turkey has come a longway since then
The more recent Patel
et al
(2002) studywhich uses the TampD scores of SampP measuredat the end of 2000 covers 354 firms in 19 emerg-ing markets The study includes only 12Turkish firms covering only 64 percent of themarket capitalisation of Turkish companiesincluded in the SampPInternational FinanceCorporation (IFC) Index As of the end of 2000the average TampD score of these 12 firms is 34the third lowest after the five Latin Americancountries and the Philippines ndash both groupsexhibiting an average score of 29
More closely related to this study are therecent surveys conducted by SampP to measurea broader-based firm-specific TampD quality inmany regions of the world often in collabora-tion with academicians in respective countriesIn 2002 SampP Governance Services publishedits first TampD study that includes companies inthe following SampPIFC indices Asia LatinAmerica Asia Pacific 100 TOPIX 150 (Japan)Then came the surveys in Russia in 20022003 and as a result of continued interestamong investors in 2004 Furthermore in
2003 SampP released its study of the SampP Europe350 companies
12
Our main objective in thisstudy is to quantify the improvement in dis-closure practices of Turkish firms in the recentyears by using a much larger sample (covering100 percent of SampPIFC Index market capital-isation of Turkish firms) and a more compre-hensive set of attributes encompassing notonly financial statement disclosure but alsodisclosure of other aspects of CG ie owner-ship board and management structures andprocesses and covering the disclosure inannual reports and company web sitesAccordingly we collaborated with SampP tocarry out the Turkish survey Our and theSampPrsquos roles in this study are explained atthe beginning of the next section SampP and theCorporate Governance Forum of Turkey(CGFT) of Sabanci University announced theresults to the public in a press release in May2002 and SampP officially published the resultsin their newsletter Ratings Direct dated 662005
13
This research is also closely related to alarge body of accounting and finance litera-ture that investigates the causes and conse-quences of voluntary disclosure In this paperwe limit our investigation to the agencytheory-based determinants of discretionarydisclosure reviewed in seminal studies suchas Verrecchia (1983 2001) Healy and Palepu(2001) and Core (2001) The theory predictsthat even though self-serving managers haveincentives to disclose their insider informa-tion they will withhold private informationwhen disclosure is too costly and that there isa strong country effect in what companiesdisclose (for empirical evidence on country-based differences see for example Berglofand Pajuste 2005 Hope
et al
2003) Priorresearch has shown that there is also consider-able within-country variation in disclosuredriven by firm characteristics
We draw on agency research studies onfirm-specific determinants of disclosure andwe consider the specific agency problemsfaced by Turkish firms in our choice of therelevant agency conflict proxies There existsan extensive body of prior research on deter-minants of disclosure that has accumulatedover the past 40 years a review of which isbeyond the scope of this paper Here we willjust mention a few important or recentpapers
14
The variables we explore undertwo categories of agency conflicts are sizemeasured as market value of equity (MVE)market-to-book value of equity ratio (MTBmeasured as MVEBVE) firm performancemeasured as earnings before interest andtaxestotal assets (ROA) and net incomeownerrsquos equity (ROE) leverage measured as
total liabilitiestotal assets (TLTA) and freecash flow (FCF)
Proxies for conflict with minority shareholders and regulatory institutions
Size and profitability
Watts and Zimmerman (1986) suggest thatlarger companies are more visible and thusthey are politically more sensitive We expectlarger and more profitable firms to havehigher TDS because they are closely followedby financial intermediaries have more com-prehensive disclosure standards in place tominimise the political costs of noncompliancewith generally accepted accounting prin-ciples (GAAP) and can better afford the costof voluntary disclosure Furthermore profit-able firms are likely to have less to hide fromtheir constituents and thus are expected toemploy less earnings management
15
Manyempirical studies have associated disclosurequantity and quality measured by a disclo-sure index with firm size and performanceand many have investigated the relationshipbetween firm characteristics and agencyproblems In their well-cited empirical workLang and Lundholm (1993) find that analystratings of disclosure are higher for firms thatperform well for larger firms firms witha weaker relation between annual stockreturns and earnings and firms that issuesecurities Doyle
et al
(2005) and Bryan andLilien (2005) document that material weak-nesses in internal control are more likely forsmaller younger financially weaker fastgrowth firms and worse performers Simi-larly using voluntary segment disclosureas a proxy for transparency Sidney andBertrand (2004) find that concentrated boardownership results in low disclosure and thiseffect is stronger when firm performance ispoor Black
et al
(2006) investigates the cross-sectional differences in Korean firmsrsquo CGpractices and again finds that firm size riskand long-term profitability and need forequity capital are positively related to betterCG Hope (2003) observes a positive correla-tion between firm size and the CIFAR indexfor annual report disclosure SimilarlyHossain
et al
(2005) has found that voluntarydisclosure of prospective information isrelated to firm size Veronina
et al
(2005) is arecent study which is similar to the presentone in that it measures the transparencypractices of 102 listed Russian firms in 2001and also investigates the cross-sectional dif-ferences in their transparency scores Usinga checklist of 441 items from InternationalAccounting Standards (IAS) they find that
use of a Big-5 auditor foreign listing sizegovernment shareholdings and indepen-dence of CEO and board chair are associatedwith transparency
Market-to-book ratio
MTB ratio has been used in the literature as aproxy for risk growth potential unreportedintangibles and firm prospects assessed bymarket participants A low ratio is associatedwith low growth potential and high free cashflows under the discretion of insiders Suchfirms have little need for external finance andthus voluntary disclosure (Core 2001) There-fore within the agency theory framework oneshould expect to see a positive relationshipbetween MTB and TampD scores For exampleBerglof and Pajuste (2005) report that moreinformation is publicly available in largerfirms firms with lower leverage higherfinancial performance higher market-to-bookratios and more concentrated ownership Onthe other hand a low MTB can also be con-sidered a sign of undervaluation by themarket The equityrsquos market value might below relative to its book value not because ofthe firmrsquos low growth potential but simplybecause future prospects of the firm are notproperly communicated to the public or thereis a general undervaluation in the market dueto local economic uncertainties For examplethe cash flows from valuable intangible assetssuch as human resources internally generatedpatents and RampD activities which are notrecognised in financial statements in line withGAAP may not have been impounded inprices in an inefficient information setting Insuch cases management may take actions toincrease transparency and put better CGpractices in action to remedy this unwantedperception and its negative effect on firmvalue Furthermore low MTB may be theresult of an increase in book value of equitydue to recently issued capital and severalresearchers have found that firms disclosemore when they have recently issued capital(see for example Frankel
et al
1995) Con-trary to the predictions of agency theory thisargument suggests a negative relationshipbetween MTB and TampD scores
16
Proxies for conflict between the shareholders and managers and creditors
Leverage
Agency theory also suggests a strong linkbetween leverage and disclosure (Jensen andMeckling 1976) In highly levered firms there
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
theless considerably high scores (710) areobtained in the category of financial trans-parency and information disclosure and theaverage overall TampD score (TDS) of Turkishcompanies is 510 a score comparable tomany countries in continental Europe
In this paper we draw on the agency theoryof the firm and model CG and TampD as a sys-tem of checks and balances that has evolvedto mitigate the agency problems often encoun-tered in the corporate form Our expectation isthat if companies have effective mechanismsin place the agency problems ndash such as ex-traction of private benefits by majorityshareholders consumption of perquisites byentrenched management under or overinvest-ment ndash and their associated costs will be miti-gated We base our analysis on this body ofresearch for the second objective of the studyexploring the cross-sectional differences inquality of disclosure in ISE firms We find thatlarger and more profitable firms as well asfirms with higher information asymmetryexhibit higher TDS to partially resolve the con-flict with minority shareholders and regula-tory institutions while free cash flow availableto management and leverage do not haveexplanatory power This is not surprising sincethe conflict between the shareholders andmanagers and creditors is not a major agencyproblem in Turkey
The study has several contributions andis expected to be of interest to regulatorsresearchers managers and market partici-pants The rankings provide a first time objec-tive quantitative and verifiable assessment ofcorporate disclosure practices of listed compa-nies in the ISE encompassing not only financialstatement disclosure but also disclosures ofshareholderrsquos rights and board and manage-ment structuresprocesses As such it isexpected to be a benchmark for future longi-tudinal and cross-country studies Moreoversince we use the same attributes and method-ology of prior SampP studies the findingsallow comparison with the scores previouslyobtained in other markets The rankings basedon a universal benchmark are expected to helpinvestors and other stakeholders in theirdecision making as they allow an objectiveassessment of differences in the level of report-ing quality and hence investment risks acrossfirms markets and sectors The TampD rankingscan also be used as a meaningful objectivemonitoring and enforcement tool by local andinternational regulators Finally this is thefirst study that investigates the relationshipbetween TampD and some agency conflictproxies prevalent in the ISE The resultsprovide out-of-sample evidence that thesefirm-specific characteristics do a good job in
explaining the variation in TDS as in someother countries
The remaining part of the paper is organisedas follows the next section describes the poorculture and regulation of disclosure in Turkeyafter which we discuss our simple model ofthe causes and the consequences of the CGTampD framework and develop our hypothesisWe then discuss prior research and the choiceof our agency conflict proxies used as deter-minants of the TampD scores The sample datarequirements and methods of analysis aredescribed followed by the results The finalsection concludes and presents some futureresearch ideas
The poor regulatory framework and culture of CG and TampD in Turkey
Turkey has historically had a poor culture ofCG and TampD It has always been a civil lawcountry and has had a very late start in theliberalisation of its economy the establishmentof its stock market (1986) and financial report-ing standards based on generally acceptedaccounting principles and the concept of fulland fair disclosure (1994)
6
The first legislation concerning commerciallaw was enacted for the first time in theOttoman Empire in 1850 and was based onthe French Code of Commerce of 1807 Underthe legal reforms of 1926 this was replaced bya system based on Swiss Civil Code and Codeof Obligations the Italian Penal Code andthe German Code of Penal Procedure Neitherthese nor the present Turkish Code of Com-merce dating back to 1957 were based on thegenerally accepted principles of accountingand auditing and hence did not regulate finan-cial reporting They highlight the strength ofthe civil law tradition weak enforcement ofrules and lack of a disclosure philosophy inthe Turkish business culture
This legal and cultural infrastructure seemsto be an important factor in the historic diffi-culty with which Turkish firms have attractedexternal capital Table 1 reconstructed fromtables 2 and 8 in La Porta
et al
(1997) placesTurkey in the French origin legal systemwhich has enjoyed the lowest access to ex-ternal capital markets as well as the weakestrule of law and investor rights protectionamong the three other civil law and the com-mon law traditions The table also depicts thatwhile GDP growth is higher in Turkey than inthe other legal origins access to external cap-ital and measures of investor rights are lowerthan even the average of the French origincountries Hence while the fast growth rate inthe Turkish economy underscores the need to
access external debt and equity capital thetable highlights its difficulty in countrieswhere enforcement of the rule of law and pro-tection of investorsrsquo rights are weak The solu-tion to this dilemma through observance ofgood CG and TampD practices has been a majormotivation for this study
Recent studies have observed that the weakCG and disclosure tradition in Turkey havecontinued until very recently Ararat and Ugur(2003) describe Turkeyrsquos civil law tradition
and its inefficient and inconsistent regulatoryframework and the ensuing paucity of the ruleof law and its enforcement They also elabo-rate on the specific CG problems faced byTurkish firms the concentrated and pyramidalownership structure dominated by familiesownership of many banks by these groupcompanies inconsistent and opaque account-ing and tax regulations and the investor mis-information caused by the absence of inflationand consolidation accounting
7
As a result of
Table 1 Access to external financing and measures of investor protection in Turkey and by legal origin
Panel A
a
Aggregate mean values for the Turkish economy and stock market and comparison with different legal origins
(1977) Table 2 presented for 49 countries and by legal origin
b
Panels C and D are reconstructed from La Porta
et al
(1977) Table 8 presented for 49 countries and by legal origin All values aretaken from the Worldscope database
this infrastructure agency conflicts are con-centrated around weak minority shareholdersrsquoand creditorsrsquo rights inconsistent and opaquedisclosure policies and lack of a culture ofvoluntary disclosure and convergence (in-separability) of ownership and management
8
These create an environment that fosters cor-ruption share dilution asset stripping tunnel-ling insider trading and market manipulation
Given these specific agency problems fos-tered by the ownership and regulatory struc-ture in large public Turkish firms we concen-trate on the disclosure and transparencycomponents of CG which are the mechanismsmost likely to improve the protection of minor-ity shareholdersrsquo and the creditorsrsquo rightsMcKinsey (2002) has carried out a global sur-vey in which institutional investors were askedto identify their CG priorities for companiesand policy makers The survey administeredin 2002 covers 201 responses from institutionalinvestors located in 31 countries Table 2reports the weights assigned by global institu-tional investors to the top seven most impor-tant CG factors for their investment decisionsin Turkey The survey results depict a patternsimilar to the average of all countries sur-veyed timely broad understandable account-ing disclosure and safeguarding of the share-holdersrsquo rights are the most important CGpriorities of institutional investors in theirdecisions to invest in Turkish companies
Unfortunately Turkey has not scored highin the scarce prior disclosure studies eitheras will be described below We believe thatthe aforementioned bleak picture is changingfast as a result of the reforms in the financialreporting environment the recent stability inTurkish politics economic reforms new regu-lation and more effective enforcement and inthis study we hope to quantify the expectedimprovement in disclosure practices of ISEfirms Some of these developments inlcude thefollowing
(a) accelerated privatisation of State Eco-nomic Enterprises
(b) the creation of private pension funds thatare expected to serve as institutionalinvestors and enhance monitoring inpublic firms
(c) monitoring of banks by the Banking Regu-lation and Supervison Agency (BRSA)
(d) the Capital Markets Boardrsquos (CMBrsquos)commitment to improvement of the regu-latory framework
(e) the CMBrsquos Corporate Governance Prin-ciples promulgated on a comply or ex-plain basis and covering areas of dis-closure minority rights board structureand management oversight The extentof compliance and the reasons for non-compliance to the principles would beindicated in the 2004 annual reports pub-lished in 2005
(f) the establishment of the Turkish Account-ing Standards Board and their mandate toconverge the local accounting standardsto the International Financial ReportingStandards (IFRS) by 2005 in line with thedirectives of the European Commission
(g) Of course Turkeyrsquos recent progress inachieving full membership of the EU willprovide the strongest inertia in establish-ing the ldquorule of lawrdquo and improved CGand TampD practices in Turkey
A model of incentives for good CGTampD and our main hypothesis
Our basic model of the relationships betweenthe conflicts of interest and information asym-metry between the stakeholders in the corpo-rate form CGTampD practices as mechanismsof checks and balances put in place to mitigatethem firm performance and the localglobaleconomy is depicted in Figure 1 In this paperwe explore the left-hand side of the model ie
Table 2 CG priorities of institutional investors for their investment decisions in 30 countries
CG factors Turkey () All 30 countries surveyed ()
Accounting disclosure 73 52Shareholder equality 64 47Market regulation and infrastructure 55 43Takeover markets 45 23Property rights 45 46Credit information 36 29Board independence 27 44
Sources McKinsey amp Company (2002) Global Investor Opinion Survey on Corporate Governance andArarat and Ugur (2003)
the quality of TampD practices of ISE firms andthe firm-specific agency-conflict related deter-minants of their TampD scores
Based on the agency theory of the firm andviewing the firm as a nexus of contracts proneto information asymmetry and conflicts ofinterest between its stakeholders (Jensen andMeckling 1976) we envision good CGTampDpractices as mechanisms of checks and bal-ances that have evolved to mitigate theseagency problems
9
Such best practices areexpected to lower the agency costs leading tohigher firm performance In turn the CGTampDpractices themselves are shaped by the legaltraditions and regulations in the firmrsquos countryof domicile as well as by international lawsand standards and the global economy Notonly do these impose new CGTampD rules andbest practices on the firm ndash such as the IFRS orCG principles of the OECD ndash but they alsocreate new opportunities in the form of positiveNPV projects and ease of access to foreigncapital on the one hand and new threats suchas regulatory intervention and litigation dueto perhaps non-compliance or misrepresenta-tion Since undertaking positive NPV projectsdepends on the firmrsquos ability to tap the equityand debt markets and since capital will flowto the highest return investments for a givenrisk level firms have the incentives to maxi-mise the return to their capital suppliers whileminimising their perceived risk by putting inplace optimal CG mechanisms Good TampDmechanisms are thus set in place to protect therights of shareholders creditors and other out-siders from extraction of private benefits byinsiders based on their superior informationThis is expected to minimise the informationalasymmetry increase investor awareness andtrust which in turn should reduce the uncer-tainty of the returns to capital suppliers andlead to lower cost of capital and hence higherfirm value (see for example Ashbaugh
et al
2004 Brown and Caylor 2004 Berglof and
Pajuste 2005)
10
This back and forth flowbetween the economic units and capital sup-pliers can be sustained only in the absence ofappropriation by insiders and other commonagency problems Thus in our model best CGTampD practices are used by firms not only tosignal their quality but also as monitoring toolsto maintain their quality
11
This simple model of incentives for goodCGTampD practices has been well-studied andreviewed extensively (see for example Wattsand Zimmerman 1986 Healy and Palepu2001 Core 2001 Rahman 2002) in literatureA recent example that draws on this literatureis the Albuquerque and Wang (2005) studywhich develops a dynamic general equilib-rium model to acknowledge the implicationsof agency conflicts through weak investorprotection for security prices They find thatcountries with weaker investor protectionhave higher overinvestment lower valuesand higher expected returns return volatilityand interest rates Ashbaugh
et al
(2004)discuss the agency problems created by theseparation of ownership and control andconclude that firms with better governancepresent less agency risk to shareholders lead-ing to lower cost of equity capital Accord-ingly we expect to observe higher quality CGTampD practices in firms that have solved or thatare less susceptible to or that have committedthemselves to reduce their agency conflictsAccordingly we pose the following hypoth-esis presented in the alternative form
H
1
Firms that have lower potential agency con-flicts are expected to have higher TampD scores
Prior research and the choice of the agency conflict proxies
Since this paper adapts an index of disclosureattributes to measure the TampD quality in a
Figure 1 A model of incentives for sound CGTampD practices
FIRM Nexus of contracts between stakeholders
CGTampD mechanisms
FIRM
Accounting Performance
Market Performance
COUNTRY
Local Economy
GLOBAL
Economy
specific country and also explores the deter-minants of the TampD scores obtained it isclosely related to both of these vast strands ofdisclosure literature Such indices have beendeveloped in several countries in the past 40years mostly carried out by rating firmsfinancial analysts academicians and databasecompanies They are indeed far too many tolist here Some examples are the disclosurequality evaluations published by the Asso-ciation for Investment Management andResearch (AIMR) over the decade 1980ndash90analystsrsquo ratings of firmsrsquo disclosure policiesdisclosure quality evaluations published bythe
Financial Post
and the country-based indexof accounting quality and transparency devel-oped by the Center for International FinancialAnalysis and Research (the CIFAR Index) in1993 and 1995 This latter index represents theaverage number of 90 items included in theannual reports of a sample of companiescovering 42 countries It has been used bymany academicians in their cross-countryresearch to uncover the relationship betweenaccounting transparency legal origin cultureand economic performance (see for exampleLa Porta
et al
1998 Hope 2003) To ourknowledge the CIFAR Index is the earliestdisclosure study that includes Turkey in itssample The average score of the 10 Turkishcompanies is measured as 586 the secondlowest after Portugal (537) among the 25code-law countries and after India (54) amongthe 17 common-law countries included in theCIFAR Index Of course this survey was car-ried out in 1995 and Turkey has come a longway since then
The more recent Patel
et al
(2002) studywhich uses the TampD scores of SampP measuredat the end of 2000 covers 354 firms in 19 emerg-ing markets The study includes only 12Turkish firms covering only 64 percent of themarket capitalisation of Turkish companiesincluded in the SampPInternational FinanceCorporation (IFC) Index As of the end of 2000the average TampD score of these 12 firms is 34the third lowest after the five Latin Americancountries and the Philippines ndash both groupsexhibiting an average score of 29
More closely related to this study are therecent surveys conducted by SampP to measurea broader-based firm-specific TampD quality inmany regions of the world often in collabora-tion with academicians in respective countriesIn 2002 SampP Governance Services publishedits first TampD study that includes companies inthe following SampPIFC indices Asia LatinAmerica Asia Pacific 100 TOPIX 150 (Japan)Then came the surveys in Russia in 20022003 and as a result of continued interestamong investors in 2004 Furthermore in
2003 SampP released its study of the SampP Europe350 companies
12
Our main objective in thisstudy is to quantify the improvement in dis-closure practices of Turkish firms in the recentyears by using a much larger sample (covering100 percent of SampPIFC Index market capital-isation of Turkish firms) and a more compre-hensive set of attributes encompassing notonly financial statement disclosure but alsodisclosure of other aspects of CG ie owner-ship board and management structures andprocesses and covering the disclosure inannual reports and company web sitesAccordingly we collaborated with SampP tocarry out the Turkish survey Our and theSampPrsquos roles in this study are explained atthe beginning of the next section SampP and theCorporate Governance Forum of Turkey(CGFT) of Sabanci University announced theresults to the public in a press release in May2002 and SampP officially published the resultsin their newsletter Ratings Direct dated 662005
13
This research is also closely related to alarge body of accounting and finance litera-ture that investigates the causes and conse-quences of voluntary disclosure In this paperwe limit our investigation to the agencytheory-based determinants of discretionarydisclosure reviewed in seminal studies suchas Verrecchia (1983 2001) Healy and Palepu(2001) and Core (2001) The theory predictsthat even though self-serving managers haveincentives to disclose their insider informa-tion they will withhold private informationwhen disclosure is too costly and that there isa strong country effect in what companiesdisclose (for empirical evidence on country-based differences see for example Berglofand Pajuste 2005 Hope
et al
2003) Priorresearch has shown that there is also consider-able within-country variation in disclosuredriven by firm characteristics
We draw on agency research studies onfirm-specific determinants of disclosure andwe consider the specific agency problemsfaced by Turkish firms in our choice of therelevant agency conflict proxies There existsan extensive body of prior research on deter-minants of disclosure that has accumulatedover the past 40 years a review of which isbeyond the scope of this paper Here we willjust mention a few important or recentpapers
14
The variables we explore undertwo categories of agency conflicts are sizemeasured as market value of equity (MVE)market-to-book value of equity ratio (MTBmeasured as MVEBVE) firm performancemeasured as earnings before interest andtaxestotal assets (ROA) and net incomeownerrsquos equity (ROE) leverage measured as
total liabilitiestotal assets (TLTA) and freecash flow (FCF)
Proxies for conflict with minority shareholders and regulatory institutions
Size and profitability
Watts and Zimmerman (1986) suggest thatlarger companies are more visible and thusthey are politically more sensitive We expectlarger and more profitable firms to havehigher TDS because they are closely followedby financial intermediaries have more com-prehensive disclosure standards in place tominimise the political costs of noncompliancewith generally accepted accounting prin-ciples (GAAP) and can better afford the costof voluntary disclosure Furthermore profit-able firms are likely to have less to hide fromtheir constituents and thus are expected toemploy less earnings management
15
Manyempirical studies have associated disclosurequantity and quality measured by a disclo-sure index with firm size and performanceand many have investigated the relationshipbetween firm characteristics and agencyproblems In their well-cited empirical workLang and Lundholm (1993) find that analystratings of disclosure are higher for firms thatperform well for larger firms firms witha weaker relation between annual stockreturns and earnings and firms that issuesecurities Doyle
et al
(2005) and Bryan andLilien (2005) document that material weak-nesses in internal control are more likely forsmaller younger financially weaker fastgrowth firms and worse performers Simi-larly using voluntary segment disclosureas a proxy for transparency Sidney andBertrand (2004) find that concentrated boardownership results in low disclosure and thiseffect is stronger when firm performance ispoor Black
et al
(2006) investigates the cross-sectional differences in Korean firmsrsquo CGpractices and again finds that firm size riskand long-term profitability and need forequity capital are positively related to betterCG Hope (2003) observes a positive correla-tion between firm size and the CIFAR indexfor annual report disclosure SimilarlyHossain
et al
(2005) has found that voluntarydisclosure of prospective information isrelated to firm size Veronina
et al
(2005) is arecent study which is similar to the presentone in that it measures the transparencypractices of 102 listed Russian firms in 2001and also investigates the cross-sectional dif-ferences in their transparency scores Usinga checklist of 441 items from InternationalAccounting Standards (IAS) they find that
use of a Big-5 auditor foreign listing sizegovernment shareholdings and indepen-dence of CEO and board chair are associatedwith transparency
Market-to-book ratio
MTB ratio has been used in the literature as aproxy for risk growth potential unreportedintangibles and firm prospects assessed bymarket participants A low ratio is associatedwith low growth potential and high free cashflows under the discretion of insiders Suchfirms have little need for external finance andthus voluntary disclosure (Core 2001) There-fore within the agency theory framework oneshould expect to see a positive relationshipbetween MTB and TampD scores For exampleBerglof and Pajuste (2005) report that moreinformation is publicly available in largerfirms firms with lower leverage higherfinancial performance higher market-to-bookratios and more concentrated ownership Onthe other hand a low MTB can also be con-sidered a sign of undervaluation by themarket The equityrsquos market value might below relative to its book value not because ofthe firmrsquos low growth potential but simplybecause future prospects of the firm are notproperly communicated to the public or thereis a general undervaluation in the market dueto local economic uncertainties For examplethe cash flows from valuable intangible assetssuch as human resources internally generatedpatents and RampD activities which are notrecognised in financial statements in line withGAAP may not have been impounded inprices in an inefficient information setting Insuch cases management may take actions toincrease transparency and put better CGpractices in action to remedy this unwantedperception and its negative effect on firmvalue Furthermore low MTB may be theresult of an increase in book value of equitydue to recently issued capital and severalresearchers have found that firms disclosemore when they have recently issued capital(see for example Frankel
et al
1995) Con-trary to the predictions of agency theory thisargument suggests a negative relationshipbetween MTB and TampD scores
16
Proxies for conflict between the shareholders and managers and creditors
Leverage
Agency theory also suggests a strong linkbetween leverage and disclosure (Jensen andMeckling 1976) In highly levered firms there
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
access external debt and equity capital thetable highlights its difficulty in countrieswhere enforcement of the rule of law and pro-tection of investorsrsquo rights are weak The solu-tion to this dilemma through observance ofgood CG and TampD practices has been a majormotivation for this study
Recent studies have observed that the weakCG and disclosure tradition in Turkey havecontinued until very recently Ararat and Ugur(2003) describe Turkeyrsquos civil law tradition
and its inefficient and inconsistent regulatoryframework and the ensuing paucity of the ruleof law and its enforcement They also elabo-rate on the specific CG problems faced byTurkish firms the concentrated and pyramidalownership structure dominated by familiesownership of many banks by these groupcompanies inconsistent and opaque account-ing and tax regulations and the investor mis-information caused by the absence of inflationand consolidation accounting
7
As a result of
Table 1 Access to external financing and measures of investor protection in Turkey and by legal origin
Panel A
a
Aggregate mean values for the Turkish economy and stock market and comparison with different legal origins
(1977) Table 2 presented for 49 countries and by legal origin
b
Panels C and D are reconstructed from La Porta
et al
(1977) Table 8 presented for 49 countries and by legal origin All values aretaken from the Worldscope database
this infrastructure agency conflicts are con-centrated around weak minority shareholdersrsquoand creditorsrsquo rights inconsistent and opaquedisclosure policies and lack of a culture ofvoluntary disclosure and convergence (in-separability) of ownership and management
8
These create an environment that fosters cor-ruption share dilution asset stripping tunnel-ling insider trading and market manipulation
Given these specific agency problems fos-tered by the ownership and regulatory struc-ture in large public Turkish firms we concen-trate on the disclosure and transparencycomponents of CG which are the mechanismsmost likely to improve the protection of minor-ity shareholdersrsquo and the creditorsrsquo rightsMcKinsey (2002) has carried out a global sur-vey in which institutional investors were askedto identify their CG priorities for companiesand policy makers The survey administeredin 2002 covers 201 responses from institutionalinvestors located in 31 countries Table 2reports the weights assigned by global institu-tional investors to the top seven most impor-tant CG factors for their investment decisionsin Turkey The survey results depict a patternsimilar to the average of all countries sur-veyed timely broad understandable account-ing disclosure and safeguarding of the share-holdersrsquo rights are the most important CGpriorities of institutional investors in theirdecisions to invest in Turkish companies
Unfortunately Turkey has not scored highin the scarce prior disclosure studies eitheras will be described below We believe thatthe aforementioned bleak picture is changingfast as a result of the reforms in the financialreporting environment the recent stability inTurkish politics economic reforms new regu-lation and more effective enforcement and inthis study we hope to quantify the expectedimprovement in disclosure practices of ISEfirms Some of these developments inlcude thefollowing
(a) accelerated privatisation of State Eco-nomic Enterprises
(b) the creation of private pension funds thatare expected to serve as institutionalinvestors and enhance monitoring inpublic firms
(c) monitoring of banks by the Banking Regu-lation and Supervison Agency (BRSA)
(d) the Capital Markets Boardrsquos (CMBrsquos)commitment to improvement of the regu-latory framework
(e) the CMBrsquos Corporate Governance Prin-ciples promulgated on a comply or ex-plain basis and covering areas of dis-closure minority rights board structureand management oversight The extentof compliance and the reasons for non-compliance to the principles would beindicated in the 2004 annual reports pub-lished in 2005
(f) the establishment of the Turkish Account-ing Standards Board and their mandate toconverge the local accounting standardsto the International Financial ReportingStandards (IFRS) by 2005 in line with thedirectives of the European Commission
(g) Of course Turkeyrsquos recent progress inachieving full membership of the EU willprovide the strongest inertia in establish-ing the ldquorule of lawrdquo and improved CGand TampD practices in Turkey
A model of incentives for good CGTampD and our main hypothesis
Our basic model of the relationships betweenthe conflicts of interest and information asym-metry between the stakeholders in the corpo-rate form CGTampD practices as mechanismsof checks and balances put in place to mitigatethem firm performance and the localglobaleconomy is depicted in Figure 1 In this paperwe explore the left-hand side of the model ie
Table 2 CG priorities of institutional investors for their investment decisions in 30 countries
CG factors Turkey () All 30 countries surveyed ()
Accounting disclosure 73 52Shareholder equality 64 47Market regulation and infrastructure 55 43Takeover markets 45 23Property rights 45 46Credit information 36 29Board independence 27 44
Sources McKinsey amp Company (2002) Global Investor Opinion Survey on Corporate Governance andArarat and Ugur (2003)
the quality of TampD practices of ISE firms andthe firm-specific agency-conflict related deter-minants of their TampD scores
Based on the agency theory of the firm andviewing the firm as a nexus of contracts proneto information asymmetry and conflicts ofinterest between its stakeholders (Jensen andMeckling 1976) we envision good CGTampDpractices as mechanisms of checks and bal-ances that have evolved to mitigate theseagency problems
9
Such best practices areexpected to lower the agency costs leading tohigher firm performance In turn the CGTampDpractices themselves are shaped by the legaltraditions and regulations in the firmrsquos countryof domicile as well as by international lawsand standards and the global economy Notonly do these impose new CGTampD rules andbest practices on the firm ndash such as the IFRS orCG principles of the OECD ndash but they alsocreate new opportunities in the form of positiveNPV projects and ease of access to foreigncapital on the one hand and new threats suchas regulatory intervention and litigation dueto perhaps non-compliance or misrepresenta-tion Since undertaking positive NPV projectsdepends on the firmrsquos ability to tap the equityand debt markets and since capital will flowto the highest return investments for a givenrisk level firms have the incentives to maxi-mise the return to their capital suppliers whileminimising their perceived risk by putting inplace optimal CG mechanisms Good TampDmechanisms are thus set in place to protect therights of shareholders creditors and other out-siders from extraction of private benefits byinsiders based on their superior informationThis is expected to minimise the informationalasymmetry increase investor awareness andtrust which in turn should reduce the uncer-tainty of the returns to capital suppliers andlead to lower cost of capital and hence higherfirm value (see for example Ashbaugh
et al
2004 Brown and Caylor 2004 Berglof and
Pajuste 2005)
10
This back and forth flowbetween the economic units and capital sup-pliers can be sustained only in the absence ofappropriation by insiders and other commonagency problems Thus in our model best CGTampD practices are used by firms not only tosignal their quality but also as monitoring toolsto maintain their quality
11
This simple model of incentives for goodCGTampD practices has been well-studied andreviewed extensively (see for example Wattsand Zimmerman 1986 Healy and Palepu2001 Core 2001 Rahman 2002) in literatureA recent example that draws on this literatureis the Albuquerque and Wang (2005) studywhich develops a dynamic general equilib-rium model to acknowledge the implicationsof agency conflicts through weak investorprotection for security prices They find thatcountries with weaker investor protectionhave higher overinvestment lower valuesand higher expected returns return volatilityand interest rates Ashbaugh
et al
(2004)discuss the agency problems created by theseparation of ownership and control andconclude that firms with better governancepresent less agency risk to shareholders lead-ing to lower cost of equity capital Accord-ingly we expect to observe higher quality CGTampD practices in firms that have solved or thatare less susceptible to or that have committedthemselves to reduce their agency conflictsAccordingly we pose the following hypoth-esis presented in the alternative form
H
1
Firms that have lower potential agency con-flicts are expected to have higher TampD scores
Prior research and the choice of the agency conflict proxies
Since this paper adapts an index of disclosureattributes to measure the TampD quality in a
Figure 1 A model of incentives for sound CGTampD practices
FIRM Nexus of contracts between stakeholders
CGTampD mechanisms
FIRM
Accounting Performance
Market Performance
COUNTRY
Local Economy
GLOBAL
Economy
specific country and also explores the deter-minants of the TampD scores obtained it isclosely related to both of these vast strands ofdisclosure literature Such indices have beendeveloped in several countries in the past 40years mostly carried out by rating firmsfinancial analysts academicians and databasecompanies They are indeed far too many tolist here Some examples are the disclosurequality evaluations published by the Asso-ciation for Investment Management andResearch (AIMR) over the decade 1980ndash90analystsrsquo ratings of firmsrsquo disclosure policiesdisclosure quality evaluations published bythe
Financial Post
and the country-based indexof accounting quality and transparency devel-oped by the Center for International FinancialAnalysis and Research (the CIFAR Index) in1993 and 1995 This latter index represents theaverage number of 90 items included in theannual reports of a sample of companiescovering 42 countries It has been used bymany academicians in their cross-countryresearch to uncover the relationship betweenaccounting transparency legal origin cultureand economic performance (see for exampleLa Porta
et al
1998 Hope 2003) To ourknowledge the CIFAR Index is the earliestdisclosure study that includes Turkey in itssample The average score of the 10 Turkishcompanies is measured as 586 the secondlowest after Portugal (537) among the 25code-law countries and after India (54) amongthe 17 common-law countries included in theCIFAR Index Of course this survey was car-ried out in 1995 and Turkey has come a longway since then
The more recent Patel
et al
(2002) studywhich uses the TampD scores of SampP measuredat the end of 2000 covers 354 firms in 19 emerg-ing markets The study includes only 12Turkish firms covering only 64 percent of themarket capitalisation of Turkish companiesincluded in the SampPInternational FinanceCorporation (IFC) Index As of the end of 2000the average TampD score of these 12 firms is 34the third lowest after the five Latin Americancountries and the Philippines ndash both groupsexhibiting an average score of 29
More closely related to this study are therecent surveys conducted by SampP to measurea broader-based firm-specific TampD quality inmany regions of the world often in collabora-tion with academicians in respective countriesIn 2002 SampP Governance Services publishedits first TampD study that includes companies inthe following SampPIFC indices Asia LatinAmerica Asia Pacific 100 TOPIX 150 (Japan)Then came the surveys in Russia in 20022003 and as a result of continued interestamong investors in 2004 Furthermore in
2003 SampP released its study of the SampP Europe350 companies
12
Our main objective in thisstudy is to quantify the improvement in dis-closure practices of Turkish firms in the recentyears by using a much larger sample (covering100 percent of SampPIFC Index market capital-isation of Turkish firms) and a more compre-hensive set of attributes encompassing notonly financial statement disclosure but alsodisclosure of other aspects of CG ie owner-ship board and management structures andprocesses and covering the disclosure inannual reports and company web sitesAccordingly we collaborated with SampP tocarry out the Turkish survey Our and theSampPrsquos roles in this study are explained atthe beginning of the next section SampP and theCorporate Governance Forum of Turkey(CGFT) of Sabanci University announced theresults to the public in a press release in May2002 and SampP officially published the resultsin their newsletter Ratings Direct dated 662005
13
This research is also closely related to alarge body of accounting and finance litera-ture that investigates the causes and conse-quences of voluntary disclosure In this paperwe limit our investigation to the agencytheory-based determinants of discretionarydisclosure reviewed in seminal studies suchas Verrecchia (1983 2001) Healy and Palepu(2001) and Core (2001) The theory predictsthat even though self-serving managers haveincentives to disclose their insider informa-tion they will withhold private informationwhen disclosure is too costly and that there isa strong country effect in what companiesdisclose (for empirical evidence on country-based differences see for example Berglofand Pajuste 2005 Hope
et al
2003) Priorresearch has shown that there is also consider-able within-country variation in disclosuredriven by firm characteristics
We draw on agency research studies onfirm-specific determinants of disclosure andwe consider the specific agency problemsfaced by Turkish firms in our choice of therelevant agency conflict proxies There existsan extensive body of prior research on deter-minants of disclosure that has accumulatedover the past 40 years a review of which isbeyond the scope of this paper Here we willjust mention a few important or recentpapers
14
The variables we explore undertwo categories of agency conflicts are sizemeasured as market value of equity (MVE)market-to-book value of equity ratio (MTBmeasured as MVEBVE) firm performancemeasured as earnings before interest andtaxestotal assets (ROA) and net incomeownerrsquos equity (ROE) leverage measured as
total liabilitiestotal assets (TLTA) and freecash flow (FCF)
Proxies for conflict with minority shareholders and regulatory institutions
Size and profitability
Watts and Zimmerman (1986) suggest thatlarger companies are more visible and thusthey are politically more sensitive We expectlarger and more profitable firms to havehigher TDS because they are closely followedby financial intermediaries have more com-prehensive disclosure standards in place tominimise the political costs of noncompliancewith generally accepted accounting prin-ciples (GAAP) and can better afford the costof voluntary disclosure Furthermore profit-able firms are likely to have less to hide fromtheir constituents and thus are expected toemploy less earnings management
15
Manyempirical studies have associated disclosurequantity and quality measured by a disclo-sure index with firm size and performanceand many have investigated the relationshipbetween firm characteristics and agencyproblems In their well-cited empirical workLang and Lundholm (1993) find that analystratings of disclosure are higher for firms thatperform well for larger firms firms witha weaker relation between annual stockreturns and earnings and firms that issuesecurities Doyle
et al
(2005) and Bryan andLilien (2005) document that material weak-nesses in internal control are more likely forsmaller younger financially weaker fastgrowth firms and worse performers Simi-larly using voluntary segment disclosureas a proxy for transparency Sidney andBertrand (2004) find that concentrated boardownership results in low disclosure and thiseffect is stronger when firm performance ispoor Black
et al
(2006) investigates the cross-sectional differences in Korean firmsrsquo CGpractices and again finds that firm size riskand long-term profitability and need forequity capital are positively related to betterCG Hope (2003) observes a positive correla-tion between firm size and the CIFAR indexfor annual report disclosure SimilarlyHossain
et al
(2005) has found that voluntarydisclosure of prospective information isrelated to firm size Veronina
et al
(2005) is arecent study which is similar to the presentone in that it measures the transparencypractices of 102 listed Russian firms in 2001and also investigates the cross-sectional dif-ferences in their transparency scores Usinga checklist of 441 items from InternationalAccounting Standards (IAS) they find that
use of a Big-5 auditor foreign listing sizegovernment shareholdings and indepen-dence of CEO and board chair are associatedwith transparency
Market-to-book ratio
MTB ratio has been used in the literature as aproxy for risk growth potential unreportedintangibles and firm prospects assessed bymarket participants A low ratio is associatedwith low growth potential and high free cashflows under the discretion of insiders Suchfirms have little need for external finance andthus voluntary disclosure (Core 2001) There-fore within the agency theory framework oneshould expect to see a positive relationshipbetween MTB and TampD scores For exampleBerglof and Pajuste (2005) report that moreinformation is publicly available in largerfirms firms with lower leverage higherfinancial performance higher market-to-bookratios and more concentrated ownership Onthe other hand a low MTB can also be con-sidered a sign of undervaluation by themarket The equityrsquos market value might below relative to its book value not because ofthe firmrsquos low growth potential but simplybecause future prospects of the firm are notproperly communicated to the public or thereis a general undervaluation in the market dueto local economic uncertainties For examplethe cash flows from valuable intangible assetssuch as human resources internally generatedpatents and RampD activities which are notrecognised in financial statements in line withGAAP may not have been impounded inprices in an inefficient information setting Insuch cases management may take actions toincrease transparency and put better CGpractices in action to remedy this unwantedperception and its negative effect on firmvalue Furthermore low MTB may be theresult of an increase in book value of equitydue to recently issued capital and severalresearchers have found that firms disclosemore when they have recently issued capital(see for example Frankel
et al
1995) Con-trary to the predictions of agency theory thisargument suggests a negative relationshipbetween MTB and TampD scores
16
Proxies for conflict between the shareholders and managers and creditors
Leverage
Agency theory also suggests a strong linkbetween leverage and disclosure (Jensen andMeckling 1976) In highly levered firms there
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
this infrastructure agency conflicts are con-centrated around weak minority shareholdersrsquoand creditorsrsquo rights inconsistent and opaquedisclosure policies and lack of a culture ofvoluntary disclosure and convergence (in-separability) of ownership and management
8
These create an environment that fosters cor-ruption share dilution asset stripping tunnel-ling insider trading and market manipulation
Given these specific agency problems fos-tered by the ownership and regulatory struc-ture in large public Turkish firms we concen-trate on the disclosure and transparencycomponents of CG which are the mechanismsmost likely to improve the protection of minor-ity shareholdersrsquo and the creditorsrsquo rightsMcKinsey (2002) has carried out a global sur-vey in which institutional investors were askedto identify their CG priorities for companiesand policy makers The survey administeredin 2002 covers 201 responses from institutionalinvestors located in 31 countries Table 2reports the weights assigned by global institu-tional investors to the top seven most impor-tant CG factors for their investment decisionsin Turkey The survey results depict a patternsimilar to the average of all countries sur-veyed timely broad understandable account-ing disclosure and safeguarding of the share-holdersrsquo rights are the most important CGpriorities of institutional investors in theirdecisions to invest in Turkish companies
Unfortunately Turkey has not scored highin the scarce prior disclosure studies eitheras will be described below We believe thatthe aforementioned bleak picture is changingfast as a result of the reforms in the financialreporting environment the recent stability inTurkish politics economic reforms new regu-lation and more effective enforcement and inthis study we hope to quantify the expectedimprovement in disclosure practices of ISEfirms Some of these developments inlcude thefollowing
(a) accelerated privatisation of State Eco-nomic Enterprises
(b) the creation of private pension funds thatare expected to serve as institutionalinvestors and enhance monitoring inpublic firms
(c) monitoring of banks by the Banking Regu-lation and Supervison Agency (BRSA)
(d) the Capital Markets Boardrsquos (CMBrsquos)commitment to improvement of the regu-latory framework
(e) the CMBrsquos Corporate Governance Prin-ciples promulgated on a comply or ex-plain basis and covering areas of dis-closure minority rights board structureand management oversight The extentof compliance and the reasons for non-compliance to the principles would beindicated in the 2004 annual reports pub-lished in 2005
(f) the establishment of the Turkish Account-ing Standards Board and their mandate toconverge the local accounting standardsto the International Financial ReportingStandards (IFRS) by 2005 in line with thedirectives of the European Commission
(g) Of course Turkeyrsquos recent progress inachieving full membership of the EU willprovide the strongest inertia in establish-ing the ldquorule of lawrdquo and improved CGand TampD practices in Turkey
A model of incentives for good CGTampD and our main hypothesis
Our basic model of the relationships betweenthe conflicts of interest and information asym-metry between the stakeholders in the corpo-rate form CGTampD practices as mechanismsof checks and balances put in place to mitigatethem firm performance and the localglobaleconomy is depicted in Figure 1 In this paperwe explore the left-hand side of the model ie
Table 2 CG priorities of institutional investors for their investment decisions in 30 countries
CG factors Turkey () All 30 countries surveyed ()
Accounting disclosure 73 52Shareholder equality 64 47Market regulation and infrastructure 55 43Takeover markets 45 23Property rights 45 46Credit information 36 29Board independence 27 44
Sources McKinsey amp Company (2002) Global Investor Opinion Survey on Corporate Governance andArarat and Ugur (2003)
the quality of TampD practices of ISE firms andthe firm-specific agency-conflict related deter-minants of their TampD scores
Based on the agency theory of the firm andviewing the firm as a nexus of contracts proneto information asymmetry and conflicts ofinterest between its stakeholders (Jensen andMeckling 1976) we envision good CGTampDpractices as mechanisms of checks and bal-ances that have evolved to mitigate theseagency problems
9
Such best practices areexpected to lower the agency costs leading tohigher firm performance In turn the CGTampDpractices themselves are shaped by the legaltraditions and regulations in the firmrsquos countryof domicile as well as by international lawsand standards and the global economy Notonly do these impose new CGTampD rules andbest practices on the firm ndash such as the IFRS orCG principles of the OECD ndash but they alsocreate new opportunities in the form of positiveNPV projects and ease of access to foreigncapital on the one hand and new threats suchas regulatory intervention and litigation dueto perhaps non-compliance or misrepresenta-tion Since undertaking positive NPV projectsdepends on the firmrsquos ability to tap the equityand debt markets and since capital will flowto the highest return investments for a givenrisk level firms have the incentives to maxi-mise the return to their capital suppliers whileminimising their perceived risk by putting inplace optimal CG mechanisms Good TampDmechanisms are thus set in place to protect therights of shareholders creditors and other out-siders from extraction of private benefits byinsiders based on their superior informationThis is expected to minimise the informationalasymmetry increase investor awareness andtrust which in turn should reduce the uncer-tainty of the returns to capital suppliers andlead to lower cost of capital and hence higherfirm value (see for example Ashbaugh
et al
2004 Brown and Caylor 2004 Berglof and
Pajuste 2005)
10
This back and forth flowbetween the economic units and capital sup-pliers can be sustained only in the absence ofappropriation by insiders and other commonagency problems Thus in our model best CGTampD practices are used by firms not only tosignal their quality but also as monitoring toolsto maintain their quality
11
This simple model of incentives for goodCGTampD practices has been well-studied andreviewed extensively (see for example Wattsand Zimmerman 1986 Healy and Palepu2001 Core 2001 Rahman 2002) in literatureA recent example that draws on this literatureis the Albuquerque and Wang (2005) studywhich develops a dynamic general equilib-rium model to acknowledge the implicationsof agency conflicts through weak investorprotection for security prices They find thatcountries with weaker investor protectionhave higher overinvestment lower valuesand higher expected returns return volatilityand interest rates Ashbaugh
et al
(2004)discuss the agency problems created by theseparation of ownership and control andconclude that firms with better governancepresent less agency risk to shareholders lead-ing to lower cost of equity capital Accord-ingly we expect to observe higher quality CGTampD practices in firms that have solved or thatare less susceptible to or that have committedthemselves to reduce their agency conflictsAccordingly we pose the following hypoth-esis presented in the alternative form
H
1
Firms that have lower potential agency con-flicts are expected to have higher TampD scores
Prior research and the choice of the agency conflict proxies
Since this paper adapts an index of disclosureattributes to measure the TampD quality in a
Figure 1 A model of incentives for sound CGTampD practices
FIRM Nexus of contracts between stakeholders
CGTampD mechanisms
FIRM
Accounting Performance
Market Performance
COUNTRY
Local Economy
GLOBAL
Economy
specific country and also explores the deter-minants of the TampD scores obtained it isclosely related to both of these vast strands ofdisclosure literature Such indices have beendeveloped in several countries in the past 40years mostly carried out by rating firmsfinancial analysts academicians and databasecompanies They are indeed far too many tolist here Some examples are the disclosurequality evaluations published by the Asso-ciation for Investment Management andResearch (AIMR) over the decade 1980ndash90analystsrsquo ratings of firmsrsquo disclosure policiesdisclosure quality evaluations published bythe
Financial Post
and the country-based indexof accounting quality and transparency devel-oped by the Center for International FinancialAnalysis and Research (the CIFAR Index) in1993 and 1995 This latter index represents theaverage number of 90 items included in theannual reports of a sample of companiescovering 42 countries It has been used bymany academicians in their cross-countryresearch to uncover the relationship betweenaccounting transparency legal origin cultureand economic performance (see for exampleLa Porta
et al
1998 Hope 2003) To ourknowledge the CIFAR Index is the earliestdisclosure study that includes Turkey in itssample The average score of the 10 Turkishcompanies is measured as 586 the secondlowest after Portugal (537) among the 25code-law countries and after India (54) amongthe 17 common-law countries included in theCIFAR Index Of course this survey was car-ried out in 1995 and Turkey has come a longway since then
The more recent Patel
et al
(2002) studywhich uses the TampD scores of SampP measuredat the end of 2000 covers 354 firms in 19 emerg-ing markets The study includes only 12Turkish firms covering only 64 percent of themarket capitalisation of Turkish companiesincluded in the SampPInternational FinanceCorporation (IFC) Index As of the end of 2000the average TampD score of these 12 firms is 34the third lowest after the five Latin Americancountries and the Philippines ndash both groupsexhibiting an average score of 29
More closely related to this study are therecent surveys conducted by SampP to measurea broader-based firm-specific TampD quality inmany regions of the world often in collabora-tion with academicians in respective countriesIn 2002 SampP Governance Services publishedits first TampD study that includes companies inthe following SampPIFC indices Asia LatinAmerica Asia Pacific 100 TOPIX 150 (Japan)Then came the surveys in Russia in 20022003 and as a result of continued interestamong investors in 2004 Furthermore in
2003 SampP released its study of the SampP Europe350 companies
12
Our main objective in thisstudy is to quantify the improvement in dis-closure practices of Turkish firms in the recentyears by using a much larger sample (covering100 percent of SampPIFC Index market capital-isation of Turkish firms) and a more compre-hensive set of attributes encompassing notonly financial statement disclosure but alsodisclosure of other aspects of CG ie owner-ship board and management structures andprocesses and covering the disclosure inannual reports and company web sitesAccordingly we collaborated with SampP tocarry out the Turkish survey Our and theSampPrsquos roles in this study are explained atthe beginning of the next section SampP and theCorporate Governance Forum of Turkey(CGFT) of Sabanci University announced theresults to the public in a press release in May2002 and SampP officially published the resultsin their newsletter Ratings Direct dated 662005
13
This research is also closely related to alarge body of accounting and finance litera-ture that investigates the causes and conse-quences of voluntary disclosure In this paperwe limit our investigation to the agencytheory-based determinants of discretionarydisclosure reviewed in seminal studies suchas Verrecchia (1983 2001) Healy and Palepu(2001) and Core (2001) The theory predictsthat even though self-serving managers haveincentives to disclose their insider informa-tion they will withhold private informationwhen disclosure is too costly and that there isa strong country effect in what companiesdisclose (for empirical evidence on country-based differences see for example Berglofand Pajuste 2005 Hope
et al
2003) Priorresearch has shown that there is also consider-able within-country variation in disclosuredriven by firm characteristics
We draw on agency research studies onfirm-specific determinants of disclosure andwe consider the specific agency problemsfaced by Turkish firms in our choice of therelevant agency conflict proxies There existsan extensive body of prior research on deter-minants of disclosure that has accumulatedover the past 40 years a review of which isbeyond the scope of this paper Here we willjust mention a few important or recentpapers
14
The variables we explore undertwo categories of agency conflicts are sizemeasured as market value of equity (MVE)market-to-book value of equity ratio (MTBmeasured as MVEBVE) firm performancemeasured as earnings before interest andtaxestotal assets (ROA) and net incomeownerrsquos equity (ROE) leverage measured as
total liabilitiestotal assets (TLTA) and freecash flow (FCF)
Proxies for conflict with minority shareholders and regulatory institutions
Size and profitability
Watts and Zimmerman (1986) suggest thatlarger companies are more visible and thusthey are politically more sensitive We expectlarger and more profitable firms to havehigher TDS because they are closely followedby financial intermediaries have more com-prehensive disclosure standards in place tominimise the political costs of noncompliancewith generally accepted accounting prin-ciples (GAAP) and can better afford the costof voluntary disclosure Furthermore profit-able firms are likely to have less to hide fromtheir constituents and thus are expected toemploy less earnings management
15
Manyempirical studies have associated disclosurequantity and quality measured by a disclo-sure index with firm size and performanceand many have investigated the relationshipbetween firm characteristics and agencyproblems In their well-cited empirical workLang and Lundholm (1993) find that analystratings of disclosure are higher for firms thatperform well for larger firms firms witha weaker relation between annual stockreturns and earnings and firms that issuesecurities Doyle
et al
(2005) and Bryan andLilien (2005) document that material weak-nesses in internal control are more likely forsmaller younger financially weaker fastgrowth firms and worse performers Simi-larly using voluntary segment disclosureas a proxy for transparency Sidney andBertrand (2004) find that concentrated boardownership results in low disclosure and thiseffect is stronger when firm performance ispoor Black
et al
(2006) investigates the cross-sectional differences in Korean firmsrsquo CGpractices and again finds that firm size riskand long-term profitability and need forequity capital are positively related to betterCG Hope (2003) observes a positive correla-tion between firm size and the CIFAR indexfor annual report disclosure SimilarlyHossain
et al
(2005) has found that voluntarydisclosure of prospective information isrelated to firm size Veronina
et al
(2005) is arecent study which is similar to the presentone in that it measures the transparencypractices of 102 listed Russian firms in 2001and also investigates the cross-sectional dif-ferences in their transparency scores Usinga checklist of 441 items from InternationalAccounting Standards (IAS) they find that
use of a Big-5 auditor foreign listing sizegovernment shareholdings and indepen-dence of CEO and board chair are associatedwith transparency
Market-to-book ratio
MTB ratio has been used in the literature as aproxy for risk growth potential unreportedintangibles and firm prospects assessed bymarket participants A low ratio is associatedwith low growth potential and high free cashflows under the discretion of insiders Suchfirms have little need for external finance andthus voluntary disclosure (Core 2001) There-fore within the agency theory framework oneshould expect to see a positive relationshipbetween MTB and TampD scores For exampleBerglof and Pajuste (2005) report that moreinformation is publicly available in largerfirms firms with lower leverage higherfinancial performance higher market-to-bookratios and more concentrated ownership Onthe other hand a low MTB can also be con-sidered a sign of undervaluation by themarket The equityrsquos market value might below relative to its book value not because ofthe firmrsquos low growth potential but simplybecause future prospects of the firm are notproperly communicated to the public or thereis a general undervaluation in the market dueto local economic uncertainties For examplethe cash flows from valuable intangible assetssuch as human resources internally generatedpatents and RampD activities which are notrecognised in financial statements in line withGAAP may not have been impounded inprices in an inefficient information setting Insuch cases management may take actions toincrease transparency and put better CGpractices in action to remedy this unwantedperception and its negative effect on firmvalue Furthermore low MTB may be theresult of an increase in book value of equitydue to recently issued capital and severalresearchers have found that firms disclosemore when they have recently issued capital(see for example Frankel
et al
1995) Con-trary to the predictions of agency theory thisargument suggests a negative relationshipbetween MTB and TampD scores
16
Proxies for conflict between the shareholders and managers and creditors
Leverage
Agency theory also suggests a strong linkbetween leverage and disclosure (Jensen andMeckling 1976) In highly levered firms there
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
the quality of TampD practices of ISE firms andthe firm-specific agency-conflict related deter-minants of their TampD scores
Based on the agency theory of the firm andviewing the firm as a nexus of contracts proneto information asymmetry and conflicts ofinterest between its stakeholders (Jensen andMeckling 1976) we envision good CGTampDpractices as mechanisms of checks and bal-ances that have evolved to mitigate theseagency problems
9
Such best practices areexpected to lower the agency costs leading tohigher firm performance In turn the CGTampDpractices themselves are shaped by the legaltraditions and regulations in the firmrsquos countryof domicile as well as by international lawsand standards and the global economy Notonly do these impose new CGTampD rules andbest practices on the firm ndash such as the IFRS orCG principles of the OECD ndash but they alsocreate new opportunities in the form of positiveNPV projects and ease of access to foreigncapital on the one hand and new threats suchas regulatory intervention and litigation dueto perhaps non-compliance or misrepresenta-tion Since undertaking positive NPV projectsdepends on the firmrsquos ability to tap the equityand debt markets and since capital will flowto the highest return investments for a givenrisk level firms have the incentives to maxi-mise the return to their capital suppliers whileminimising their perceived risk by putting inplace optimal CG mechanisms Good TampDmechanisms are thus set in place to protect therights of shareholders creditors and other out-siders from extraction of private benefits byinsiders based on their superior informationThis is expected to minimise the informationalasymmetry increase investor awareness andtrust which in turn should reduce the uncer-tainty of the returns to capital suppliers andlead to lower cost of capital and hence higherfirm value (see for example Ashbaugh
et al
2004 Brown and Caylor 2004 Berglof and
Pajuste 2005)
10
This back and forth flowbetween the economic units and capital sup-pliers can be sustained only in the absence ofappropriation by insiders and other commonagency problems Thus in our model best CGTampD practices are used by firms not only tosignal their quality but also as monitoring toolsto maintain their quality
11
This simple model of incentives for goodCGTampD practices has been well-studied andreviewed extensively (see for example Wattsand Zimmerman 1986 Healy and Palepu2001 Core 2001 Rahman 2002) in literatureA recent example that draws on this literatureis the Albuquerque and Wang (2005) studywhich develops a dynamic general equilib-rium model to acknowledge the implicationsof agency conflicts through weak investorprotection for security prices They find thatcountries with weaker investor protectionhave higher overinvestment lower valuesand higher expected returns return volatilityand interest rates Ashbaugh
et al
(2004)discuss the agency problems created by theseparation of ownership and control andconclude that firms with better governancepresent less agency risk to shareholders lead-ing to lower cost of equity capital Accord-ingly we expect to observe higher quality CGTampD practices in firms that have solved or thatare less susceptible to or that have committedthemselves to reduce their agency conflictsAccordingly we pose the following hypoth-esis presented in the alternative form
H
1
Firms that have lower potential agency con-flicts are expected to have higher TampD scores
Prior research and the choice of the agency conflict proxies
Since this paper adapts an index of disclosureattributes to measure the TampD quality in a
Figure 1 A model of incentives for sound CGTampD practices
FIRM Nexus of contracts between stakeholders
CGTampD mechanisms
FIRM
Accounting Performance
Market Performance
COUNTRY
Local Economy
GLOBAL
Economy
specific country and also explores the deter-minants of the TampD scores obtained it isclosely related to both of these vast strands ofdisclosure literature Such indices have beendeveloped in several countries in the past 40years mostly carried out by rating firmsfinancial analysts academicians and databasecompanies They are indeed far too many tolist here Some examples are the disclosurequality evaluations published by the Asso-ciation for Investment Management andResearch (AIMR) over the decade 1980ndash90analystsrsquo ratings of firmsrsquo disclosure policiesdisclosure quality evaluations published bythe
Financial Post
and the country-based indexof accounting quality and transparency devel-oped by the Center for International FinancialAnalysis and Research (the CIFAR Index) in1993 and 1995 This latter index represents theaverage number of 90 items included in theannual reports of a sample of companiescovering 42 countries It has been used bymany academicians in their cross-countryresearch to uncover the relationship betweenaccounting transparency legal origin cultureand economic performance (see for exampleLa Porta
et al
1998 Hope 2003) To ourknowledge the CIFAR Index is the earliestdisclosure study that includes Turkey in itssample The average score of the 10 Turkishcompanies is measured as 586 the secondlowest after Portugal (537) among the 25code-law countries and after India (54) amongthe 17 common-law countries included in theCIFAR Index Of course this survey was car-ried out in 1995 and Turkey has come a longway since then
The more recent Patel
et al
(2002) studywhich uses the TampD scores of SampP measuredat the end of 2000 covers 354 firms in 19 emerg-ing markets The study includes only 12Turkish firms covering only 64 percent of themarket capitalisation of Turkish companiesincluded in the SampPInternational FinanceCorporation (IFC) Index As of the end of 2000the average TampD score of these 12 firms is 34the third lowest after the five Latin Americancountries and the Philippines ndash both groupsexhibiting an average score of 29
More closely related to this study are therecent surveys conducted by SampP to measurea broader-based firm-specific TampD quality inmany regions of the world often in collabora-tion with academicians in respective countriesIn 2002 SampP Governance Services publishedits first TampD study that includes companies inthe following SampPIFC indices Asia LatinAmerica Asia Pacific 100 TOPIX 150 (Japan)Then came the surveys in Russia in 20022003 and as a result of continued interestamong investors in 2004 Furthermore in
2003 SampP released its study of the SampP Europe350 companies
12
Our main objective in thisstudy is to quantify the improvement in dis-closure practices of Turkish firms in the recentyears by using a much larger sample (covering100 percent of SampPIFC Index market capital-isation of Turkish firms) and a more compre-hensive set of attributes encompassing notonly financial statement disclosure but alsodisclosure of other aspects of CG ie owner-ship board and management structures andprocesses and covering the disclosure inannual reports and company web sitesAccordingly we collaborated with SampP tocarry out the Turkish survey Our and theSampPrsquos roles in this study are explained atthe beginning of the next section SampP and theCorporate Governance Forum of Turkey(CGFT) of Sabanci University announced theresults to the public in a press release in May2002 and SampP officially published the resultsin their newsletter Ratings Direct dated 662005
13
This research is also closely related to alarge body of accounting and finance litera-ture that investigates the causes and conse-quences of voluntary disclosure In this paperwe limit our investigation to the agencytheory-based determinants of discretionarydisclosure reviewed in seminal studies suchas Verrecchia (1983 2001) Healy and Palepu(2001) and Core (2001) The theory predictsthat even though self-serving managers haveincentives to disclose their insider informa-tion they will withhold private informationwhen disclosure is too costly and that there isa strong country effect in what companiesdisclose (for empirical evidence on country-based differences see for example Berglofand Pajuste 2005 Hope
et al
2003) Priorresearch has shown that there is also consider-able within-country variation in disclosuredriven by firm characteristics
We draw on agency research studies onfirm-specific determinants of disclosure andwe consider the specific agency problemsfaced by Turkish firms in our choice of therelevant agency conflict proxies There existsan extensive body of prior research on deter-minants of disclosure that has accumulatedover the past 40 years a review of which isbeyond the scope of this paper Here we willjust mention a few important or recentpapers
14
The variables we explore undertwo categories of agency conflicts are sizemeasured as market value of equity (MVE)market-to-book value of equity ratio (MTBmeasured as MVEBVE) firm performancemeasured as earnings before interest andtaxestotal assets (ROA) and net incomeownerrsquos equity (ROE) leverage measured as
total liabilitiestotal assets (TLTA) and freecash flow (FCF)
Proxies for conflict with minority shareholders and regulatory institutions
Size and profitability
Watts and Zimmerman (1986) suggest thatlarger companies are more visible and thusthey are politically more sensitive We expectlarger and more profitable firms to havehigher TDS because they are closely followedby financial intermediaries have more com-prehensive disclosure standards in place tominimise the political costs of noncompliancewith generally accepted accounting prin-ciples (GAAP) and can better afford the costof voluntary disclosure Furthermore profit-able firms are likely to have less to hide fromtheir constituents and thus are expected toemploy less earnings management
15
Manyempirical studies have associated disclosurequantity and quality measured by a disclo-sure index with firm size and performanceand many have investigated the relationshipbetween firm characteristics and agencyproblems In their well-cited empirical workLang and Lundholm (1993) find that analystratings of disclosure are higher for firms thatperform well for larger firms firms witha weaker relation between annual stockreturns and earnings and firms that issuesecurities Doyle
et al
(2005) and Bryan andLilien (2005) document that material weak-nesses in internal control are more likely forsmaller younger financially weaker fastgrowth firms and worse performers Simi-larly using voluntary segment disclosureas a proxy for transparency Sidney andBertrand (2004) find that concentrated boardownership results in low disclosure and thiseffect is stronger when firm performance ispoor Black
et al
(2006) investigates the cross-sectional differences in Korean firmsrsquo CGpractices and again finds that firm size riskand long-term profitability and need forequity capital are positively related to betterCG Hope (2003) observes a positive correla-tion between firm size and the CIFAR indexfor annual report disclosure SimilarlyHossain
et al
(2005) has found that voluntarydisclosure of prospective information isrelated to firm size Veronina
et al
(2005) is arecent study which is similar to the presentone in that it measures the transparencypractices of 102 listed Russian firms in 2001and also investigates the cross-sectional dif-ferences in their transparency scores Usinga checklist of 441 items from InternationalAccounting Standards (IAS) they find that
use of a Big-5 auditor foreign listing sizegovernment shareholdings and indepen-dence of CEO and board chair are associatedwith transparency
Market-to-book ratio
MTB ratio has been used in the literature as aproxy for risk growth potential unreportedintangibles and firm prospects assessed bymarket participants A low ratio is associatedwith low growth potential and high free cashflows under the discretion of insiders Suchfirms have little need for external finance andthus voluntary disclosure (Core 2001) There-fore within the agency theory framework oneshould expect to see a positive relationshipbetween MTB and TampD scores For exampleBerglof and Pajuste (2005) report that moreinformation is publicly available in largerfirms firms with lower leverage higherfinancial performance higher market-to-bookratios and more concentrated ownership Onthe other hand a low MTB can also be con-sidered a sign of undervaluation by themarket The equityrsquos market value might below relative to its book value not because ofthe firmrsquos low growth potential but simplybecause future prospects of the firm are notproperly communicated to the public or thereis a general undervaluation in the market dueto local economic uncertainties For examplethe cash flows from valuable intangible assetssuch as human resources internally generatedpatents and RampD activities which are notrecognised in financial statements in line withGAAP may not have been impounded inprices in an inefficient information setting Insuch cases management may take actions toincrease transparency and put better CGpractices in action to remedy this unwantedperception and its negative effect on firmvalue Furthermore low MTB may be theresult of an increase in book value of equitydue to recently issued capital and severalresearchers have found that firms disclosemore when they have recently issued capital(see for example Frankel
et al
1995) Con-trary to the predictions of agency theory thisargument suggests a negative relationshipbetween MTB and TampD scores
16
Proxies for conflict between the shareholders and managers and creditors
Leverage
Agency theory also suggests a strong linkbetween leverage and disclosure (Jensen andMeckling 1976) In highly levered firms there
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
specific country and also explores the deter-minants of the TampD scores obtained it isclosely related to both of these vast strands ofdisclosure literature Such indices have beendeveloped in several countries in the past 40years mostly carried out by rating firmsfinancial analysts academicians and databasecompanies They are indeed far too many tolist here Some examples are the disclosurequality evaluations published by the Asso-ciation for Investment Management andResearch (AIMR) over the decade 1980ndash90analystsrsquo ratings of firmsrsquo disclosure policiesdisclosure quality evaluations published bythe
Financial Post
and the country-based indexof accounting quality and transparency devel-oped by the Center for International FinancialAnalysis and Research (the CIFAR Index) in1993 and 1995 This latter index represents theaverage number of 90 items included in theannual reports of a sample of companiescovering 42 countries It has been used bymany academicians in their cross-countryresearch to uncover the relationship betweenaccounting transparency legal origin cultureand economic performance (see for exampleLa Porta
et al
1998 Hope 2003) To ourknowledge the CIFAR Index is the earliestdisclosure study that includes Turkey in itssample The average score of the 10 Turkishcompanies is measured as 586 the secondlowest after Portugal (537) among the 25code-law countries and after India (54) amongthe 17 common-law countries included in theCIFAR Index Of course this survey was car-ried out in 1995 and Turkey has come a longway since then
The more recent Patel
et al
(2002) studywhich uses the TampD scores of SampP measuredat the end of 2000 covers 354 firms in 19 emerg-ing markets The study includes only 12Turkish firms covering only 64 percent of themarket capitalisation of Turkish companiesincluded in the SampPInternational FinanceCorporation (IFC) Index As of the end of 2000the average TampD score of these 12 firms is 34the third lowest after the five Latin Americancountries and the Philippines ndash both groupsexhibiting an average score of 29
More closely related to this study are therecent surveys conducted by SampP to measurea broader-based firm-specific TampD quality inmany regions of the world often in collabora-tion with academicians in respective countriesIn 2002 SampP Governance Services publishedits first TampD study that includes companies inthe following SampPIFC indices Asia LatinAmerica Asia Pacific 100 TOPIX 150 (Japan)Then came the surveys in Russia in 20022003 and as a result of continued interestamong investors in 2004 Furthermore in
2003 SampP released its study of the SampP Europe350 companies
12
Our main objective in thisstudy is to quantify the improvement in dis-closure practices of Turkish firms in the recentyears by using a much larger sample (covering100 percent of SampPIFC Index market capital-isation of Turkish firms) and a more compre-hensive set of attributes encompassing notonly financial statement disclosure but alsodisclosure of other aspects of CG ie owner-ship board and management structures andprocesses and covering the disclosure inannual reports and company web sitesAccordingly we collaborated with SampP tocarry out the Turkish survey Our and theSampPrsquos roles in this study are explained atthe beginning of the next section SampP and theCorporate Governance Forum of Turkey(CGFT) of Sabanci University announced theresults to the public in a press release in May2002 and SampP officially published the resultsin their newsletter Ratings Direct dated 662005
13
This research is also closely related to alarge body of accounting and finance litera-ture that investigates the causes and conse-quences of voluntary disclosure In this paperwe limit our investigation to the agencytheory-based determinants of discretionarydisclosure reviewed in seminal studies suchas Verrecchia (1983 2001) Healy and Palepu(2001) and Core (2001) The theory predictsthat even though self-serving managers haveincentives to disclose their insider informa-tion they will withhold private informationwhen disclosure is too costly and that there isa strong country effect in what companiesdisclose (for empirical evidence on country-based differences see for example Berglofand Pajuste 2005 Hope
et al
2003) Priorresearch has shown that there is also consider-able within-country variation in disclosuredriven by firm characteristics
We draw on agency research studies onfirm-specific determinants of disclosure andwe consider the specific agency problemsfaced by Turkish firms in our choice of therelevant agency conflict proxies There existsan extensive body of prior research on deter-minants of disclosure that has accumulatedover the past 40 years a review of which isbeyond the scope of this paper Here we willjust mention a few important or recentpapers
14
The variables we explore undertwo categories of agency conflicts are sizemeasured as market value of equity (MVE)market-to-book value of equity ratio (MTBmeasured as MVEBVE) firm performancemeasured as earnings before interest andtaxestotal assets (ROA) and net incomeownerrsquos equity (ROE) leverage measured as
total liabilitiestotal assets (TLTA) and freecash flow (FCF)
Proxies for conflict with minority shareholders and regulatory institutions
Size and profitability
Watts and Zimmerman (1986) suggest thatlarger companies are more visible and thusthey are politically more sensitive We expectlarger and more profitable firms to havehigher TDS because they are closely followedby financial intermediaries have more com-prehensive disclosure standards in place tominimise the political costs of noncompliancewith generally accepted accounting prin-ciples (GAAP) and can better afford the costof voluntary disclosure Furthermore profit-able firms are likely to have less to hide fromtheir constituents and thus are expected toemploy less earnings management
15
Manyempirical studies have associated disclosurequantity and quality measured by a disclo-sure index with firm size and performanceand many have investigated the relationshipbetween firm characteristics and agencyproblems In their well-cited empirical workLang and Lundholm (1993) find that analystratings of disclosure are higher for firms thatperform well for larger firms firms witha weaker relation between annual stockreturns and earnings and firms that issuesecurities Doyle
et al
(2005) and Bryan andLilien (2005) document that material weak-nesses in internal control are more likely forsmaller younger financially weaker fastgrowth firms and worse performers Simi-larly using voluntary segment disclosureas a proxy for transparency Sidney andBertrand (2004) find that concentrated boardownership results in low disclosure and thiseffect is stronger when firm performance ispoor Black
et al
(2006) investigates the cross-sectional differences in Korean firmsrsquo CGpractices and again finds that firm size riskand long-term profitability and need forequity capital are positively related to betterCG Hope (2003) observes a positive correla-tion between firm size and the CIFAR indexfor annual report disclosure SimilarlyHossain
et al
(2005) has found that voluntarydisclosure of prospective information isrelated to firm size Veronina
et al
(2005) is arecent study which is similar to the presentone in that it measures the transparencypractices of 102 listed Russian firms in 2001and also investigates the cross-sectional dif-ferences in their transparency scores Usinga checklist of 441 items from InternationalAccounting Standards (IAS) they find that
use of a Big-5 auditor foreign listing sizegovernment shareholdings and indepen-dence of CEO and board chair are associatedwith transparency
Market-to-book ratio
MTB ratio has been used in the literature as aproxy for risk growth potential unreportedintangibles and firm prospects assessed bymarket participants A low ratio is associatedwith low growth potential and high free cashflows under the discretion of insiders Suchfirms have little need for external finance andthus voluntary disclosure (Core 2001) There-fore within the agency theory framework oneshould expect to see a positive relationshipbetween MTB and TampD scores For exampleBerglof and Pajuste (2005) report that moreinformation is publicly available in largerfirms firms with lower leverage higherfinancial performance higher market-to-bookratios and more concentrated ownership Onthe other hand a low MTB can also be con-sidered a sign of undervaluation by themarket The equityrsquos market value might below relative to its book value not because ofthe firmrsquos low growth potential but simplybecause future prospects of the firm are notproperly communicated to the public or thereis a general undervaluation in the market dueto local economic uncertainties For examplethe cash flows from valuable intangible assetssuch as human resources internally generatedpatents and RampD activities which are notrecognised in financial statements in line withGAAP may not have been impounded inprices in an inefficient information setting Insuch cases management may take actions toincrease transparency and put better CGpractices in action to remedy this unwantedperception and its negative effect on firmvalue Furthermore low MTB may be theresult of an increase in book value of equitydue to recently issued capital and severalresearchers have found that firms disclosemore when they have recently issued capital(see for example Frankel
et al
1995) Con-trary to the predictions of agency theory thisargument suggests a negative relationshipbetween MTB and TampD scores
16
Proxies for conflict between the shareholders and managers and creditors
Leverage
Agency theory also suggests a strong linkbetween leverage and disclosure (Jensen andMeckling 1976) In highly levered firms there
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
total liabilitiestotal assets (TLTA) and freecash flow (FCF)
Proxies for conflict with minority shareholders and regulatory institutions
Size and profitability
Watts and Zimmerman (1986) suggest thatlarger companies are more visible and thusthey are politically more sensitive We expectlarger and more profitable firms to havehigher TDS because they are closely followedby financial intermediaries have more com-prehensive disclosure standards in place tominimise the political costs of noncompliancewith generally accepted accounting prin-ciples (GAAP) and can better afford the costof voluntary disclosure Furthermore profit-able firms are likely to have less to hide fromtheir constituents and thus are expected toemploy less earnings management
15
Manyempirical studies have associated disclosurequantity and quality measured by a disclo-sure index with firm size and performanceand many have investigated the relationshipbetween firm characteristics and agencyproblems In their well-cited empirical workLang and Lundholm (1993) find that analystratings of disclosure are higher for firms thatperform well for larger firms firms witha weaker relation between annual stockreturns and earnings and firms that issuesecurities Doyle
et al
(2005) and Bryan andLilien (2005) document that material weak-nesses in internal control are more likely forsmaller younger financially weaker fastgrowth firms and worse performers Simi-larly using voluntary segment disclosureas a proxy for transparency Sidney andBertrand (2004) find that concentrated boardownership results in low disclosure and thiseffect is stronger when firm performance ispoor Black
et al
(2006) investigates the cross-sectional differences in Korean firmsrsquo CGpractices and again finds that firm size riskand long-term profitability and need forequity capital are positively related to betterCG Hope (2003) observes a positive correla-tion between firm size and the CIFAR indexfor annual report disclosure SimilarlyHossain
et al
(2005) has found that voluntarydisclosure of prospective information isrelated to firm size Veronina
et al
(2005) is arecent study which is similar to the presentone in that it measures the transparencypractices of 102 listed Russian firms in 2001and also investigates the cross-sectional dif-ferences in their transparency scores Usinga checklist of 441 items from InternationalAccounting Standards (IAS) they find that
use of a Big-5 auditor foreign listing sizegovernment shareholdings and indepen-dence of CEO and board chair are associatedwith transparency
Market-to-book ratio
MTB ratio has been used in the literature as aproxy for risk growth potential unreportedintangibles and firm prospects assessed bymarket participants A low ratio is associatedwith low growth potential and high free cashflows under the discretion of insiders Suchfirms have little need for external finance andthus voluntary disclosure (Core 2001) There-fore within the agency theory framework oneshould expect to see a positive relationshipbetween MTB and TampD scores For exampleBerglof and Pajuste (2005) report that moreinformation is publicly available in largerfirms firms with lower leverage higherfinancial performance higher market-to-bookratios and more concentrated ownership Onthe other hand a low MTB can also be con-sidered a sign of undervaluation by themarket The equityrsquos market value might below relative to its book value not because ofthe firmrsquos low growth potential but simplybecause future prospects of the firm are notproperly communicated to the public or thereis a general undervaluation in the market dueto local economic uncertainties For examplethe cash flows from valuable intangible assetssuch as human resources internally generatedpatents and RampD activities which are notrecognised in financial statements in line withGAAP may not have been impounded inprices in an inefficient information setting Insuch cases management may take actions toincrease transparency and put better CGpractices in action to remedy this unwantedperception and its negative effect on firmvalue Furthermore low MTB may be theresult of an increase in book value of equitydue to recently issued capital and severalresearchers have found that firms disclosemore when they have recently issued capital(see for example Frankel
et al
1995) Con-trary to the predictions of agency theory thisargument suggests a negative relationshipbetween MTB and TampD scores
16
Proxies for conflict between the shareholders and managers and creditors
Leverage
Agency theory also suggests a strong linkbetween leverage and disclosure (Jensen andMeckling 1976) In highly levered firms there
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
is a higher demand for and supply of informa-tion and creditors themselves produce infor-mation about the borrower
17
Furthermore asa result of monitoring by informed creditorsand strict debt covenants the debtor firm hasto commit itself to the discipline of debt pay-ments and cannot as freely expropriate the freecash flows (Jensen 1986) Bebchuck
et al
(2000)suggest that controlling-minority structures(CMSs) like stock pyramids cross-ownershipsand dual class shares such as those found inTurkey could benefit from the constrainingfunction of leverage and thus reduce the highagency costs associated with such ownershipstructures Thus we expect a higher commit-ment to reduce the agency conflicts and thushigher disclosure in firms with higher leverageeven in CMSs However empirical studieshave provided conflicting results For examplewhile Ho and Wong (2001) found no relation-ship between disclosure and leverage severalstudies have found a significant relationship(for a positive relationship see for exampleHossain
et al
1994 and Dellas and Hess 2005for a weakly negative one see Hope 2003)We have no priors for the large family-ownedgroup firms some of which can be consideredCMSs and many of whom own their ownbanks When the bank is directly or indirectlyowned by the borrower the role of leverage asa monitoring andor disciplining device ndash bothfor the majority owner group and the manage-ment team ndash is suspect
Free cash flow
FCF defined as the cash flow at the discretionof management after all positive NPV projectsare undertaken has been an extensively usedmeasure of the agency conflict between self-serving managers and diffuse shareholdersFirms with higher FCFs are subject to higheragency costs and this results in undervalua-tion of the firmrsquos equity Indeed firms withhigher FCFs and lower MTB ratios have beentargets for leveraged buyout transactions (seeLehn and Poulsen (1989) for example) Ourhypothesis predicts a positive relationshipbetween TD score and FCF measure ascompanies with accumulated cash have anincentive to be as transparentinformative aspossible through their disclosure to mitigatethe negative impact of the excessive cashholding on the equity value However whenconcentrated ownership structures like thosein the ISE do not leave the control of cash inthe hands of managers the role of FCF as asource of agency conflict between the man-agers and shareholders and its role in dis-closure practices are suspect
Sample data requirements and methods of analysis
Our sample consists of the 52 largest marketcap and most liquid (based on trade volume)corporations traded in the ISE 44 of which arecurrently followed by SampP and included intheir SampPIFC Emerging Markets Index Thestudy analyses disclosure practices of thesefirms from an investor perspective It thereforefocuses on sources of information that aremost readily accessible by local and inter-national investors ndash typically the latest avail-able English language and local languagecompany annual reports and the English andTurkish websites The study leverages SampPrsquosexpertise in corporate governance index con-struction and financial analysis We carriedout the study ourselves but make use of theirlist of 98 best TampD attributes use their classi-fication of overall TampD into three subcatego-ries and utilise their scoring methodology
Transparency and disclosure are evaluatedby searching for the inclusion of best practiceinformation items (ldquoattributesrdquo) in the 2003annual reports and websites of our samplefirms Our primary data source is the annualreports prepared in English If we do not finda specific information item in this primarysource then we sequentially search for thatitem in the English website then in the Turk-ish annual report and finally in the Turkishwebsite As SampP uses US disclosure best prac-tices as an implicit benchmark we have modi-fied their list of 98 attributes used in priorsurveys A similar modification was also car-ried out in Russia The list of attributes hasbeen extended to 106 information items in thecase of Turkey in order to incorporate somelocal market culture and regulation-specificissues the CMBrsquos new CG principles concen-trated ownership due to family ownershipand foundersrsquo shares cross-holding relation-ships between the holding companies andtheir subsidiaries high inflation rate and themandate to follow IFRS starting with the firstinterim reports in 2005
The 106 attributes grouped into the follow-ing three sub-categories are presented inAppendix 1
1 Ownership structure and investor relations(OwnStr)
2 Financial transparency and information disclo-sure (FinDisc)
3 Board and management structure and processes(BrdMgmt)
The inclusion of each attribute is scored ona trinary basis as ldquoyesrdquo (included) or ldquonordquo (notincluded) and ldquoNArdquo (not applicable) to en-sure objectivity Each ldquoyesrdquo answer is equal to
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
one point and the overall TampD score (TDS) foreach firm is calculated as
(1)
where
j
=
the attribute category subscript
j
=
1 2 3
k
= the attribute subscript k = 1 106Sjk = the number of info items disclosed (answered as ldquoyesrdquo) by the firm in each category and TOTS = the total maximum possible ldquoyesrdquo answers for each firm18
Companies are finally ranked based on theirTDS in deciles An overall ranking will reflectthe total number of the 106 maximum possibleattributes disclosed in a companyrsquos annualreport and website Individual rankings foreach of the three sub-categories are calculatedlikewise by reference to the maximum pos-sible number of yes answers for each sub-category We use the TDS rankings to repre-sent the market participantsrsquo assessments ofthe completeness clarity and transparency ofa firmrsquos disclosure policies
In exploring the determinants of TDS weconduct a cross-sectional analysis of the re-lationship between the TampD scores and thefirm-specific proxies for agency variables dis-cussed above Both financial statement pricereturn data are obtained from the electronicdatabase in the ISE website
Results
In Table 3 we present a comparison of theaverage scores of Turkish firms with those ofother geographical regions of the world andwith European countries in the Europe 300index of the SampPIFC database The scores areobtained from SampPrsquos Transparency and Dis-closure Studies in 2001ndash2002 The comparisonindicates that UK and US composite have thehighest rankings globally Continental Europeand developed Asia are somewhat lower andemerging Asia and Latin America have thelowest disclosure scores particularly in termsof Board and Management disclosures Thescores vary a lot between regions and betweendifferent categories of TampD In all regionsand countries reported including Turkey thehighest TampD scores are observed in theFinDisc category while the lowest scores areobtained in the OwnStr and BrdMgmt cate-gories in all regions except for Japan LatinAmerica Emerging Asia and Turkey In theseexception regionscountries BrdMgmt scoresare the lowest Apart from Japan these are allemerging markets
TDS TOTS= sumsum Sjkkj
The average TampD scores of sample ISE firmsare reported in the last row of Panel A Whilethe overall score is 41 out of 100 points theOwnStr FinDisc and BrdMgmt scores are 3964 and 20 respectively Overall Turkish com-panies have a disclosure pattern similar toEmerging Asia but have higher FinDisc andlower BrdMgmt scores SampP also assigns adecile score to each regioncountry by record-ing the scores in deciles while rounding themup to the next highest digit The decile scoresof all reported regions and countries includ-ing the 350 firms in 15 European countries andRussia are given in Panels B and C In compar-ison to other European countriesrsquo overall TampDscore in deciles (6) Turkey has a somewhatlower average score (5) which is comparableto many continental European countries
Descriptive statisticsTable 4 presents the descriptive statistics forour sample The average 2003 overall TDscores and average scores in the three TampDcategories depicted in Panel A indicate that thetransparency and disclosure scores of ISE com-panies are not impressive especially in termsof board and management structures and pro-cesses The highest scores are obtained in thecategory of financial transparency and infor-mation disclosure which is not surprising asmost of the attributes in that category are man-datory disclosures required by the CMBrsquosaccounting principles that all publicly tradedfirms are required to follow and which moreor less follow IAS We also include the meanmedian min and max values of our firm-specific agency conflict proxies such as sizemarket-to-book ratio leverage accountingprofitability of the sample firms as definedearlier Most of the sample firms have recentlybecome public firms and on average they havetraded on the ISE for 115 years They haveslightly negative returns in 2003 due to thecontinuation of the economic slump of 2002their return on assets and return on equityratios have been moderate on average 6 per-cent and 17 percent respectively and theirmean leverage about 50 percent
Evidence on the relation between agency problems and the TampD scoresIn this section we test if potential conflicts ofinterest caused by agency problems explainthe differences in the TDS scores by estimat-ing the following regression equation where iis the firm subscript j stands for the overallscore (and scores in each sub-category) andthere is no time subscript as the regressionsare run for only 2003 values of the inde-pendent variables Our explanatory variables
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
are free cash-flow (FCF) accounting per-formance (ROE) leverage (TLTA) marketcapitalisation (MVE) and market-to-bookratio (MTB) as defined earlier
(2)TDS TL TA MVE MTB
ROE FCFi j i i i
i i
= + + ++ + +b b b b
b b e0 1 2 3
4 5
TDS ratings are based on 2003 annual reportsConsistent with this all the explanatoryvariables are measured by using 2003 end ofyear values This is also consistent with ourhypothesised relationship between the TDSscores and the variables included in the modelFor example firms with higher realised profitsover the year tend to share more information
Table 3 Comparison of TampD scores in different geographical regions with Turkey
Panel A The scores in three categories of TampD and the composite scores (in )
Source SampPrsquos Transparency and Disclosure Study 2002 (with the exception of results for Turkey)aRepresents the sum of all the above listed European companiesbExcludes the 133 UK companies
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
with the public naturally in their current state-ments Similarly firms with low MTB ratiosbased on end of year figures would tend to bemore transparent both about their financialsand managerial practices in their immediateannual reports to mitigate any miscommuni-cation about their future prospects that couldwell be the reason for a poor MTB ratio
We report the results with the last four vari-ables only The FCF variable was excludedfrom the regression equations for two reasonsFirst we were able to find information onlyon 31 firms to estimate an appropriate FCFmeasure However as far as the signs andsignificance of the coefficients are concernedour initial model that included FCF providedsimilar results to those reported here but thecoefficient of FCF was not significant Our sec-ond reason for excluding FCF stems from theinstitutional structure of Turkish corporationsSince as discussed earlier family membersusually sit in the highest executive positionsand in the boards the rest of the managers arenot likely to have much power and authorityover the use of free cash flows As a resultagency problemcost associated with FCF ina typical corporate form ndash with diffusedownership and fully delegated decision mak-ing authority to the management team ndash issomewhat limited This argument defies therole of FCF as a proxy for agency conflict
Table 5 shows that our model is a valid onewith an F-value of 406 Given the usual limi-tations of multiple regression analysis usedin our study we checked our independentvariables for a possible correlation With theexception of a ROE with leverage ndash whichseem to have a moderate positive correlationndash no significant correlation exists among thevariables Furthermore usual diagnosticsapplied to residuals show no violation ofhomoscedasticity or normality assumptions ofthe models
We find that size and market-to-book aresignificant in explaining the variation in TDSwhile leverage is not We conjecture that largerfirms with higher followings by investors andwith higher political costs of non-complianceor litigation threat have higher quality disclo-sures as expected On the other hand thenegative relation between MTB and TDS indi-cates that riskier low growth or undervaluedfirms also disclose more to signal their qualityto perhaps counter the negative assessmentsof the market participants about their pros-pects Leverage is not significant in explainingthe changes in TDS probably due to loss ofpower in the tests due to the familiesrsquo owner-ship of banks from whom they obtain credit19
When we add either ROA or ROE to themodel they both have significant coefficientsas we expected20
Table 4 Descriptive statistics TampD scores and financial profiles of 52 ISE firms in 2003
TDS = Transparency and disclosure scores of the sample firms measured in three disclosure categories basedon the 106 question survey presented in Appendix 1 and the SampP scoring methodology MVE = (year endclose price) (Number of outstanding shares) and measured at the end of 2003 in million TL MTBratio = market value of equitybook-value of equity DTA(leverage) = total liabilitiestotal assetsROA = firm performance measure defined as earnings before interest and taxestotal assets ROE = firmperformance indicator measured as net incomeownerrsquos equity All variables are measured as of 31December 2003
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
Apart from their explanatory power for theoverall scores size and market-to-book arealso significant in explaining the ownershipstructure and board and management struc-ture sub-category scores None of the variablestried are significant in explaining differencesin financial disclosure scores probably becausethe information items included in this sub-category are mostly mandatory disclosureitems (with the exception of voluntary earlyadoption of IFRS consolidated reporting andinflation accounting) and hence there is lessof a variation in these scores
Summary and concluding remarks
Collaborating with SampP we score the TampDpractices of 52 large and liquid ISE firms inthree categories of disclosure by examiningtheir 2003 annual reports and websites Weconclude that the TampD quality of Turkish firmsare at best moderate with the highest and low-est scores observed in the Financial Disclosureand Board and Management processes cate-gories respectively The results indicate thatthe annual reports and websites are weak interms of voluntary disclosure In the explor-atory part of our study that investigates thefirm-specific determinants of the TDS we findthat while size accounting profitability andmarket-to-book ratio explain the differences inthe scores in the ownership and board andmanagement sub-categories as well as in over-all scores leverage remains insignificant in allfour models
There are several limitations of the studyOur sample is small in size and by construc-tion composed of the largest and mostfollowed firms in the ISE and thus may notbe representative of the population of Turkishfirms Second we concentrate on a subset ofdisclosure and transparency attributes inannual reports and websites to the exclusion
of timeliness truthfulness and accessibility ofinformation and other sources of CG informa-tion Finally the study gives at best a snap-shot picture of disclosure quality in ISE firmsand its relation to firm-specific variablesbecause the tests are based on only one yearrsquosdata Thus it lacks longitudinal generalis-ability and ignores a possible lead lag relation-ship in the link between TampD scores and ouragency conflict proxies To mitigate some ofthese limitations and enrich our insight wewill expand our data set to ISE-100 firms andreplicate the study in 2005 Our expectationis that the TampD scores and their relationshipwith our agency conflict proxies will improvedue to the required compliance with IFRS andthe CMBrsquos CG principles for public firms in2005 We are also planning to investigate theright-hand side of our model presented inFigure 1 that is the relationship between theTampD scores and firm-level and country-levelfinancial performance We are particularlyinterested in unveiling whether good TampDincreases the firmrsquos and the marketrsquos access todomestic and foreign capital and hence leadsto higher levels of growth
Acknowledgements
We would like to thank Utpal Bhattacharya forhis invaluable comments on an earlier versionof this paper This paper was presented atthe 28th Annual Congress of the EuropeanAccounting Association (EAArsquo05) held inGothenburg Sweden May 18ndash21 2005 and atthe 1st International Symposium of the Istan-bul Chamber of Certified Public Accountants(ISMMO ndash TURMOBrsquo05) held in AntalyaApril 20ndash24 2005 and we thank the partici-pants of both conferences for helpful com-ments and suggestions The list of attributesthe scoring methodology and the sales datafor other surveyed countries are obtained
Table 5 Cross-sectional regressions of TD scores on profitability size market-to-book ratio and leverage The regression coefficientstheir p-values (in parentheses) and the overall F-statistics are presented for each model specification
aThe TD scores and the explanatory variables lnMVE (natural log of market value of equity) MTB (market value of equitybook-valueof equity) DTA (total liabilitiestotal assets) and ROE (net incomeownerrsquos equity) ratios are measured as of 31 December 2003
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
from Standard amp Poorrsquos and are gratefullyacknowledged We are also thankful to AydosOumlzel Erkin Solak Esra Aksu and MericcedilBıccedilakccedilı our dear assistants for their carefuland competent reading of the annual reportsto Amra Balic of SampP London for her exper-tise in rating and for training the assistantsand finally to the Corporate GovernanceForum of Turkey (CGFT) of Sabanci Univer-sity for financial support
Notes
1 Some local examples are the recent resolve torewrite the Turkish Commercial Code that hasbeen in effect since 1957 and the new CG Prin-ciples of the Turkish Capital Markets Board whichwill be effective in 2005 Another one isthe recent Reporting Standard (RS) 1 ldquoTheOperating and Financial Reviewrdquo in the UKAnnouncing its publication Ian Mackintoshthe Chairman of the ASB said ldquoThis reportingstandard represents an important step forwardin encouraging greater transparency and moreopen communication between companies andtheir shareholders hellip This should be of greatbenefit to many partiesrdquo (Association of Inter-national Accountants 2005)
2 See for example the G-Index of Gompers et al(2003) and Gov-7 of Brown and Caylor (2005)
3 In their official newsletter dated 6605entitled ldquoCorporate Governance the TurkishTransparency and Disclosure Surveyrdquo Stan-dard amp Poorrsquos states that they view corporatetransparency as an important factor affecting acompanyrsquos attractiveness to investors and as avital element of corporate governance
4 Yurtoglu (2003) documents that families ulti-mately own about 80 percent of the 305 compa-nies traded in the ISE as of 2001
5 In line with these three categories TampD isdefined as adherence to the principles ofldquofull-disclosurerdquo and ldquoopenness and informa-tivenessrdquo in public presentation of financialstatements as well as the ownership and boardmanagement structures and processes followedby the company As such an attempt to quan-tify it like ours would not measure the truth-fulness of the information disclosed in thefinancial reports or the actual quality of boardand management structures and investorrelations that is it does not measure the ldquoCGqualityrdquo of the firm Implicitly assuming truth-ful disclosure it basically measures the ldquoqualityof disclosurerdquo of its governance and financialreporting mechanisms
6 An important reason for exploring the TDquality using data from a single country is thatmulti-country studies may yield mixed resultsas a result of their differences in terms of thesizes and liberalisation of their markets andinstitutions political and economic risk ex-posure the average market capitalisation andtheir reporting standards We use only ISE data
in this study to control for these differences thatare expected to affect both the TD scores offirms and the variables we use as agency con-flict proxies Yurtoglu (2003) specifically statesthat Turkey is especially suitable for exploringthe link with agency costs because it sharesmany features of weak CG such as concen-trated family ownership weak regulation andenforcement pyramidal groups and dual classshares
7 In large ISE firms the conflict between creditorsand shareholders has not been an issue becausemany banks are owned by the business-groupsthemselves that often have control over thebanksrsquo lending decisions Furthermore familymembers usually sit in the highest executivepositions and in the boards and the rest ofthe managers do not have much power andauthority over the use of free cash flows As aresult the managerndashshareholder conflict is alsononexistent
8 Such stock pyramids cross-ownership anddual-class voting and non-voting equity struc-tures are predominant in Turkey Bebchuk et al(2000) call these controlling-minority structuresand posit that they lead to much larger agencycosts than both highly leveraged and concen-trated ownership structures
9 This is similar to the view in Florackis andOzkan (2004) that investigates managerialownership and compensation and ownershipconcentration as two corporate governancedevices that mitigate agency costs
10 Another advantage of compliance with goodTampD practices is that it mitigates the politicalcosts of non-compliance and hence reduces therisk of higher taxes litigation and too muchregulation Field et al (2005) indeed find thatdisclosure deters certain types of litigation
11 Taken literally the model predicts limitlessdisclosure which of course is prohibited by itscost Only to the extent the aforementionedbenefits of disclosure exceed its costs we expectthe firm to increase its voluntary disclosure andtransparency Furthermore excessive trans-parency may limit the competitive advantage ofthe firm and the limitations in place againstextraction of private benefits by insiders mayreduce the incentives of controlling share-holders to undertake positive NPV projects andto monitor the managers (see for exampleCore 2001 for the trade-off between lower costof capital and litigation costs against higherproprietary and incentive costs)
12 Standard amp Poorrsquos (2002) states that the SampPrsquosTransparency and Disclosure Study will even-tually cover about 1600 companies which rep-resent approximately 70 percent of the worldrsquostradable market capitalisation
13 The newsletter entitled ldquoCorporate Gover-nance the Turkish Transparency and Disclo-sure Surveyrdquo discusses their scoring method-ology TampD studies carried out in other coun-tries announces the top 5 companies in the ISElists the most and the least practised onesamong the 106 TampD attributes includes a tableof cross-country comparisons which we extend
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
as our Table 3 and a reproduction of our Table 4Panel A
14 The reader is referred to a very useful andcomprehensive review of empirical literatureentitled ldquoDevelopments of firm-to-outsidercommunication researchrdquo by Schadewitz andBlevins (2005) on the web
15 Faulkender et al (2005) finds that the extent ofthe potential disagreement between investorsand managers depends on prior firm perfor-mance
16 Size and book-to-market ratio are also impor-tant variables that explain excess returnsboth in developed and many emerging markets(Fama and French 1993 1998) includingTurkey (Aksu and Onder 2003) As such thesetwo variables may also act as controls for thisomitted variable
17 This is especially true for firms with significantamount of bank debt in their capital structureRecent theories of financial intermediation havefocused on the role of banks to produce andtransmit information in capital markets whenthere are informational asymmetries Eitherbecause of their information-gathering costadvantages or intimate customer relationshipdeveloped over time banks are assumed tohave access to valuable information that maynot be available to other market participants
18 In determining the denominator TOTS weexclude the NA answers for info items notapplicable to the firm For example in question35 under BrdMgmt category ldquoWhether boardmembers are employees of parent cordquo is notincluded in the total possible score (TOTS) incase the company is not a holding companySimilarly question 36 in the Financial Disclo-sure category ldquoList of group transactionsrdquo isexcluded from TOTS if the firm is not a groupfirm
19 When we exclude the highly leveraged banksand financial institutions from our sample leav-ing only 43 firms the coefficient of leverage hasthe correct positive sign and becomes weaklysignificant This is reasonable as high leveragedoes not necessarily mean financial distress ormonitoring intensity in banks
20 As the ROE measure already incorporates theeffect of leverage hence might be a possiblereason behind the insignificance of leverage wetried the runs with ROA instead of ROE Thisdid not have any effect on the signs and signi-ficances of the coefficients in the model Wealso tried an indicator variable for whetherthe company is a group firm with a highlyconcentrated ownership This was found to beinsignificant
References
Aksu M and Onder T (2003) Size and Book-to-market Effects as Proxies for Fundamentals andas Determinants of Returns in the ISE Workingpaper Sabanci University Available at httpssrncomabstract=250919 (accessed July 2005)
Albuquerque R and Wang N (2005) Agency Con-flicts Investment and Asset Pricing Workingpaper University of Rochester and ColumbiaUniversity Available at httppapersssrncompapertaf abstract_id=637303 (accessed January2006)
Ararat M and Ugur M (2003) Corporate Gover-nance in Turkey An Overview and Some PolicyRecommendations Corporate Governance 3 58ndash75
Ashbaugh H S Collins D W and Lafond R(2004) Corporate Governance and the Cost ofEquity Capital Working paper University ofWisconsin Available at httppapersssrncompapertafabstract_id=63968 (accessed April2005)
Association of International Accountants (2005) RS1 The Operating and Financial Review AIAAccountancy E-Newsletter 13 May
Bebchuk L Kraakman R and Triantis G (2000)Stock Pyramids Cross-Ownership and Dual ClassEquity The Mechanisms and Agency Costs ofSeparating Control from Cash-flow Rights In RMorck (ed) Concentrated Corporate OwnershipChicago IL University of Chicago Press pp 445ndash460
Beeks W and Brown P (2005) Do Better GovernedAustralian Firms Make More Informative Dis-closures Working paper Lancaster UniversityAvailable at httpssrncomabstract=650062(accessed May 2005)
Berglof E and Pajuste A (2005) What Do FirmsDisclose and Why Enforcing Corporate Gover-nance and Transparency in Central and EasternEurope Working paper Stockholm School ofEconomics
Black B S Jang H and Kim W (2006) PredictingFirmsrsquo Corporate Governance Choices Evidencefrom Korea Journal of Corporate Finance 12 660ndash691
Brown L D and Caylor M L (2004) CorporateGovernance and Firm Performance Workingpaper Georgia State University
Brown L D and Caylor M L (2005) CorporateGovernance and Firm Valuation Working paperGeorgia State University
Bryan S H and Lilien S B (2005) Characteristicsof Firms with Material Weaknesses in InternalControl An Assessment of Section 404 ofSarbanes Oxley Working paper Wake ForestUniversity
Center for International Financial Analysis andResearch (1993 and 1995) International Accountingand Auditing Trends Volumes I and II PrincetonNJ CIFAR Publications Inc
Core J E (2001) A Review of the Empirical Disclo-sure Literature Discussion Journal of Accountingamp Economics 31 441ndash456
Dellas H and Hess M (2005) Financial Develop-ment and Stock Returns A Cross-Country Anal-ysis Journal of International Money and Finance 24891ndash912
Doyle J W Ge W and McWay S (2005)Determinants of Weaknesses in Internal Controlover Financial Reporting Working paperUniversity of Michigan Utah and New YorkUniversity
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
Fama E F and French K R (1993) Common RiskFactors in the Returns on Stocks and BondsJournal of Financial Economics 33 3ndash56
Fama E F and French K R (1998) Value versusGrowth The International Evidence The Journalof Finance 53 1975ndash1999
Faulkender M Milbourn T T and Thakor A(2005) Does Corporate Performance DetermineCapital Structure and Dividend Policy Workingpaper Washington University
Field L Lowry M and Shu S (2005) Does Disclo-sure Deter or Trigger Litigation Journal ofAccounting and Economics 39 487ndash507
Florackis C and Ozkan A (2004) Agency Costsand Corporate Governance Mechanisms Evi-dence for UK Firms Working paper Universityof York
Frankel R McNichols M and Wilson G P (1995)Discretionary Disclosure and External FinancingThe Accounting Review 70 135ndash149
Gompers P Ishii J and Metrick A (2003) Corpo-rate Governance and Equity Prices QuarterlyJournal of Economics 118 102ndash155
Gugler K Mueller D C and Yurtoglu B B (2004)Corporate Governance and the Returns on Invest-ment Journal of Law and Economics 47 589ndash633
Healy P and Palepu K (2001) A Review of theEmpirical Disclosure Literature Journal ofAccounting amp Economics 31 405ndash440
Ho S and Wong K S (2001) A Study of the Re-lationship Between Corporate GovernanceStructures and the Extent of Voluntary Disclo-sures Journal of International Accounting Auditingand Taxation 10 139ndash156
Hope Ole-Kristian (2003) Firm-level Disclosuresand the Relative Roles of Culture and LegalOrigin Journal of International Financial Manage-ment and Accounting 14 218ndash248
Hossain M Tan L M and Adams M (1994)Voluntary Disclosure in an Emerging CapitalMarket Some Empirical Evidence from Com-panies Listed on the Kuala Lumpur Stock Ex-change The International Journal of Accounting 29334ndash351
Hossain M Ahmad K and Godfrey J (2005)Investment Opportunity Set and Voluntary Dis-closure of Prospective Information Journal ofBusiness Finance and Accounting 12 871ndash907
Institute of International Finance Inc (2005) Cor-porate Governance Seen as a Key Issue in theDevelopment of Turkeyrsquos Stock Markets pressrelease Washington DC April 12 Available athttpwwwiifcomressreleasequaggaid=108(accessed May 2005)
Jensen M C and Meckling W H (1976) Theory ofthe Firm Managerial Behavior Agency Costsand Ownership Structure Journal of FinancialEconomics 3 305ndash360
Jensen M C (1986) Agency Cost of Free CashFlow Corporate Finance and Takeovers Ameri-can Economic Review 76 323ndash339
La Porta R Lopez-de Silanes F and Shleifer A(1997) Legal Determinants of External FinanceJournal of Finance 52 1131ndash1150
La Porta R Lopez-de Silanes F Shleifer A andVishny B (1998) Law and Finance Journal ofPolitical Economy 106 1113ndash1155
Lang M and Lundholm R (1993) Cross-sectionalDeterminants of Analyst Ratings of CorporateDisclosures Journal of Accounting Research 31246ndash271
Lehn K and Poulsen A (1989) Free Cash Flow andStockholder Gains in Going Private TransactionsJournal of Finance 44 771ndash787
McKinsey amp Company (2002) Global InvestorOpinion Survey on Corporate GovernanceAvailable at httpwwwmckinseycom (ac-cessed April 2005)
OECD (1999) Principles of Corporate GovernanceOECD Paris
Patel S Balic A and Bwakira L (2002) MeasuringTransparency and Disclosure at Firm-level inEmerging Markets Emerging Markets Review 3325ndash337
Rahman A R (2002) Incomplete Financial Con-tracting Disclosure Corporate Governance andFirm Value Available at httpssrncomabstract=348304 (accessed June 2005)
Sadka G (2004) Financial Reporting Growth andProductivity Theory and International EvidenceWorking paper University of Chicago Availableat httppapersssrncompapertafabstract_id=652301 (accessed April 2005)
Schadewitz H J and Blevins D R (2005) FromDisclosure to Business Communication AReview of the Transformation Available athttpwwwwestgaedu~bquest1997disclosuhtml (accessed April 2005)
Shleifer A and Wolfenzon D (2000) InvestorProtection and Equity Markets Working paperHarvard Institute of Economic Research PaperNo 1906 Available at httpssrncomabstract=245448 (accessed May 2005)
Sidney L and Bertrand B (2004) Director Owner-ship and Voluntary Segment Disclosure HongKong Evidence Journal of International FinancialManagement and Accounting 15 235ndash261
Standard amp Poorrsquos (2002) (a Division of TheMcGraw-Hill Companies Inc) Transparencyand Disclosure Study Europe Available athttpwwwgovernancestandardandpoorscomavailable under Governance Services Trans-parency and Disclosure Studies as of April2003
Standard amp Poorrsquos (2005) (a Division of TheMcGraw-Hill Companies Inc) Corporate Gover-nance Turkish Transparency and DisclosureSurvey Available at httpwwwgovernancestandardandpoorscom available under Gover-nance Services Transparency and DisclosureStudies as of May 2006
Veronina T Morris R D and Gray S (2005)Corporate Financial Transparency in Russia AnEmpirical Study of Russian Company PracticesWorking paper University of New South Walesand University of Sydney
Verrecchia R E (1983) Discretionary DisclosureJournal of Accounting and Economics 5 179ndash194
Verrechia R E (2001) Essays on Disclosure Journalof Accounting and Economics 32 97ndash180
Watts R and Zimmerman J (1986) PositiveAccounting Theory Englewood Cliffs NJPrentice-Hall
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
Yurtoglu B B (2000) Ownership Control andPerformance of Turkish Listed Firms Empirica27 193ndash222
Yurtoglu B B (2003) Corporate Governanceand Implications for Minority Shareholders inTurkey Corporate Ownership and Control 1 72ndash86
Mine Aksu has been teaching as an AssistantProfessor of Accounting at Sabanci Universityin Istanbul Turkey for four and a half yearsShe also serves in the research group of theCorporate Governance Forum of Turkey in thesame institution and as the Country Directorfor Turkey to the International AccountingSection of the American Accounting Associa-tion She previously taught at Temple Univer-sity in Philadelphia and at Koc University inIstanbul Her research areas include financialdistress and debt restructurings employeestock options market anomalies financialdisclosure and other corporate governanceissues She is a graduate of Bogazici University
in Istanbul and received a PhD in Accountingfrom Syracuse University in upstate NewYork
Arman Kosedag is an Associate Professor ofFinance at the Campbell School of Business atBerry College Rome Georgia He has taughtat the Graduate School of ManagementndashSabanci University Istanbul Turkey and atOklahoma State University Stillwater Okla-homa He has published in the QuarterlyJournal of Business and Economics Journal ofBusiness Finance amp Accounting InternationalReview of Financial Analysis and Review ofFinancial Economics His current researchinterest focuses on corporate governance div-idend policy and leveraged buyouts (inc-luding reverse LBOs and reLBOs) ArmanKosedag received his BA degree in BusinessAdministration from Istanbul University andholds Master of Science and PhD degrees inFinance both awarded by Louisiana StateUniversity
Appendix 1 List of attributes used in the survey
Ownership
1) No of issued and os ordinary shares2) No of issued and os other shares (pref non-vot recap)3) Par value of each ordinary share4) Par value of each other share (pref non-vot recap)5) No of auth but unissued amp os ordinary shares6) No of auth but unissued amp os other shares7) Top 1 shareholder8) Top 3 shareholders9) Top 5 shareholders
10) Top 10 shareholders11) No and identity of shareholders holding more than 312) No and identity of shareholders holding more than 513) No and identity of shareholders holding more than 1014) Identity of shareholders holding at least 5015) Float 16) Descriptions of share classes17) Review of shareholders by type 18) Percentage of cross-ownership19) Existence of Corporate Governance Charter or Code of Best Practice20) Reproduction of its Corporate Governance CharterCode of Best Practice21) Mention of Articles of Association22) Details about Articles of Association (ie Charter Articles of Incorp)23) Voting rights for each voting share24) How or who nominates directors to board25) How shareholders convene an EGM26) Procedure for putting Inquiry Rights to the board27) Procedure for proposals at shareholders meetings28) Review of last shareholders meeting (eg minutes)29) Calendar of important shareholder dates
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
30) Any (in)formal voting agreements or blocks (relevant to family ownership)31) Shareholding by senior managers32) Ultimate beneficiaries in case of institutional co or cross shareholdings
Financial disclosure
1) Its accounting policies2) Accounting standards it uses for its accounts3) Accounts according to the local accounting standards4) Accounts according to internationally recognised accounting standard (IASGAAP)5) BS according to international accounting standard (IASGAAP)6) IS according to international accounting standard (IASGAAP)7) CF according to international accounting standard (IASGAAP)8) Accounts adjusted for inflation9) Basic earnings forecast of any kind
10) Detailed earnings forecast11) Financial information on a quarterly basis12) Segment analysis (broken down by business line)13) Name of its auditing firm14) Reproduction of the auditorsrsquo report15) How much it pays in audit fees to the auditor16) Any non-audit fees paid to auditor17) Consolidated financial statements (or only the parentholding company)18) Methods of asset valuation19) Information on method of fixed assets depreciation20) List of affiliates in which it holds a minority stake21) Reconciliation of its domestic accounting standards to IASUS GAAP22) Ownership structure of affiliates23) Details of the kind of business it is in24) Details of the products or services producedprovided25) Output in physical terms disclosed (No of users etc)26) Characteristics of assets employed 27) Efficiency indicators (ROA ROE etc)28) Any industry-specific ratios29) Discussion of corporate strategy30) Any plans for investment in the coming year(s)31) Detailed info about investment plans in the coming year(s)32) Output forecast of any kind33) Overview of trends in its industry34) Its market share for any or all of its businesses 35) Listregister of related party transactions36) Listregister of group transactions37) English Annual report on the web site
Board and management
1) List of board members (names)2) Details about directors (other than nametitle)3) Details about current employmentposition of directors provided4) Details about previous employmentpositions provided5) When each of the directors joined the board6) Classification of directors as an executive or an outside director7) Named chairman listed8) Details about the chairman (other than nametitle)
Appendix 1 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
Appendix 2 The list of ISE firms included in our survey
9) Details about role of the board of directors at the company10) List of matters reserved for the board11) List of board committees12) Existence of an audit committee13) Names on audit committee14) Existence of a remunerationcompensation committee15) Names on remunerationcompensation committee16) Existence of a nomination committee17) Names on nomination committee18) Existence of other internal audit functions besides Audit committee19) Existence of a strategyinvestmentfinance committee20) No of shares in the company held by directors21) Review of last board meeting (eg minutes)22) Whether they provide director training23) Decision-making process of directorsrsquo pay24) Specifics of directorsrsquo salaries (eg numbers)25) Form of directorsrsquo salaries (eg cash shares etc)26) Specifics on performance-related pay for directors27) Decision-making of managersrsquo (not Board) pay28) Specifics of managersrsquo (not on Board) salaries (eg numbers)29) Form of managersrsquo (not on Board) salaries30) Specifics on performance-related pay for managers31) List of senior managers (not on the Board of Directors)32) Backgrounds of senior managers33) Details of the CEOs contracted34) No of shares held by managers in other affiliated companies35) Whether board members are employees of parent company (in case company is a
consolidated affiliatesubsidiary)36) Whether any group policies exist re nature of relationship between parent and affiliates (with
respect to CG of affiliatessubsidiaries)37) Whether any members of senior management are related (family joint business etc) to any
major shareholder
Company Tickersymbol
Sectorcode
Sector
Ak Enerji Elektik Uretimi Otoproduktor Gurbu AS
AKENR 55 Utilities
Akbank TAS AKBNK 40 FinancialsAkcansa AS AKCNS 15 MaterialsAksa Akrilik Kimya Sanayil AS AKSA 25 Consumer discretionaryAksigorta AKGRT 40 FinancialsAlarko Holding AS ALARK 20 IndustrialsAnadolu Efes Biracilik ve Malt Sanayii AEFES 30 Consumer staplesArcelik AS ARCLK 25 Consumer discretionaryAselsan Askeri Elektronik Sanayii ve
Ticaret ASASELS 20 Industrials
Aygaz AS AYGAZ 55 UtilitiesBeko Elektronik BEKO 25 Consumer discretionaryBosch Fren BFREN 20 Industrials
Appendix 1 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
Company Tickersymbol
Sectorcode
Sector
Appendix 2 Continued
Brisa Bridgestone Sabanci Lastik Sanve Tic
BRISA 25 Consumer discretionary
BSH Profilo Elektrikli Gerecleri BSPRO 25 Consumer discretionaryCelebi Hava Servisi AS CLEBI 20 IndustrialsCimsa Cimento San ve Tic AS CIMSA 15 MaterialsDogus Holding DOAS Construction wholesale retailDogan Holding AS DOHOL 20 IndustrialsDogan Yayin Holding AS DYHOL 25 Consumer discretionaryEczacibasi Ilac San ve Tic AS ECILC 35 Health CareEnka Insaat ve Sanayi AS ENKAI 20 IndustrialsEregli Demir ve Celik Fabrikalari AS AEFES 15 MaterialsFinansbank FINBN 40 FinancialsFord Otosan Otomobil Sanayi AS FROTO 25 Consumer discretionaryHaci Omer Sabanci Holding AS SAHOL 40 FinancialsHurriyet Gazetecilik ve Matbaacilik
ASHURGZ 25 Consumer discretionary
Ihlas Holding IHLAS 40 FiancialsIs Gayrimenkul Yatirum Ortakligi AS ISGYO 40 FinancialsKardemir 15 MaterialsKoc Holding AS KCHOL 20 IndustrialsKordsa Sabanci Dupont Endustri KORDS 25 Consumer discretionaryMigros AS MIGRS 30 Consumer staplesNetas Northern Electric
Telekomunikasyon ASNETAS 45 Information technology
Petrokimya Holding AS PETKM 15 MaterialsPetrol Ofisi PTOFS 10 EnergySasa Dupont Sabanci Polyester SASA 15 MaterialsT Garanti Bankasi GARAN 40 FinancialsT Sise Cam Fabrikalari AS SISE 25 Consumer discretionaryTansas Perakende Magazcilik Ticaret
ASTNSAS 30 Consumer staples
Tofas Turk Otomobil Fabrikasi AS TOASO 25 Consumer discretionaryTrakya Cam Sanayil AS TRKCM 20 IndustrialsTurk Traktor TTRAK 20 IndustrialsTupras Turkiye Petrol Rafinerileri AS TUPRS 10 EnergyTurk Dis Ticaret Bankasi DISBA 40 FinancialsTurk Hava Yollari AS THYAO 20 IndustrialsTurkcell Iletisim Hizmetleri AS TCELL 50 Telecommunication servicesTurkiye is Bankasi AS ISBNK 40 FinancialsUlker ULKER 30 Consumer staplesVestel Elektronik Sanayi ve Ticaret AS VESTL 25 Consumer discretionaryYapi ve Kredi Bankasi AS YKBNK 40 FinancialsYazici Holding YAZIC Manuf wholesale retail Zorlu Enerji ZOREN 55 Utilities
