+ All Categories
Home > Documents > Political connections, related party transactions, … connections, related party transactions, ......

Political connections, related party transactions, … connections, related party transactions, ......

Date post: 11-Mar-2018
Category:
Upload: buiquynh
View: 220 times
Download: 3 times
Share this document with a friend
49
Accepted Manuscript Political connections, related party transactions, and auditor choice: Evidence from Indonesia Ahsan Habib, Abdul Haris Muhammad, Haiyan Jiang PII: S1815-5669(17)30004-8 DOI: http://dx.doi.org/10.1016/j.jcae.2017.01.004 Reference: JCAE 102 To appear in: Journal of Contemporary Accounting & Economics Received Date: 5 March 2016 Accepted Date: 12 January 2017 Please cite this article as: Habib, A., Haris Muhammad, A., Jiang, H., Political connections, related party transactions, and auditor choice: Evidence from Indonesia, Journal of Contemporary Accounting & Economics (2017), doi: http:// dx.doi.org/10.1016/j.jcae.2017.01.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Transcript
Page 1: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

Accepted Manuscript

Political connections, related party transactions, and auditor choice: Evidence

from Indonesia

Ahsan Habib, Abdul Haris Muhammad, Haiyan Jiang

PII: S1815-5669(17)30004-8

DOI: http://dx.doi.org/10.1016/j.jcae.2017.01.004

Reference: JCAE 102

To appear in: Journal of Contemporary Accounting & Economics

Received Date: 5 March 2016

Accepted Date: 12 January 2017

Please cite this article as: Habib, A., Haris Muhammad, A., Jiang, H., Political connections, related party transactions,

and auditor choice: Evidence from Indonesia, Journal of Contemporary Accounting & Economics (2017), doi: http://

dx.doi.org/10.1016/j.jcae.2017.01.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Page 2: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

1

Political connections, related party transactions, and auditor choice:

Evidence from Indonesia

Ahsan Habib*

School of Accountancy

Massey University

Private Bag 102904

Auckland, New Zealand

Phone: +64 9 4140800

Email: [email protected]

Abdul Haris Muhammad

Staff of Directorate of Tax Regulations I, Directorate General of Taxes

Ministry of Finance, Indonesia

Jl. Gatot Subroto Kav.

40-42, Jakarta Selatan, Jakarta

Indonesia 12190 Email: [email protected]

&

Haiyan Jiang

School of Accountancy

Massey University

Private Bag 102904

Auckland, New Zealand

Email: [email protected]

*Contact author

We thank an anonymous reviewer for providing numerous constructive comments on the manuscript.

We also thank the Editor, Ferdinand Gul for comments.

Page 3: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

2

Political connections, related party transactions, and auditor choice:

Evidence from Indonesia

Abstract

This paper investigates how political connections in concert with related party transactions

(RPTs) determine auditor choice in Indonesia. Our study is motivated by conflicting findings

in the literature on whether politically connected firms appoint reputable auditors (Big 4

auditors). On one hand, politically connected firms are less likely to appoint Big 4 auditors if

they wish to cover up RPT-related tunneling activities by providing financial statements that

fail to reflect their true economic performance. On the other hand, politically connected

insiders who refrain from self-dealing would prefer higher-quality financial reporting and,

hence, appoint Big 4 auditors. Using data from Indonesia, we find support for the former. By

documenting the role of RPTs as a motivating factor for politically connected firms to choose

non-Big 4 auditors, we enrich the political connection and auditor choice literature.

Keywords: Political connections; Related party transactions (RPTs); Indonesia; Auditor

choice

JEL classification: G3, M0, M4

Page 4: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

3

1. Introduction

This paper investigates how political connections in concert with related party

transactions (RPTs) determine auditor choice in Indonesia. Competing arguments exist

regarding auditor choice by politically connected firms. On one hand, connected firms might

prefer to appoint non-Big 4 auditors to mask their tunneling and rent-seeking activities as

manifested in their financial statements. Such tunneling and rent seeking activities are

undertaken by politically connected firms in order to establish and maintain their political

connections (Morck et al., 2000; Ma et al., 2013). Given the value of political connections

(e.g. Fisman, 2001), politically connected firms may manipulate accounting numbers to

conceal true economic performance, in order to ensure that their diversionary practices

remain a secret (Guedhami et al., 2014). Appointing non-Big 4 auditors is a step towards that

direction.

On the other hand, connected insiders who refrain from self-dealing would prefer to

choose Big 4 auditors, so that outside investors will value their reporting transparency

positively. It is well accepted that investors value accounting transparency in order to identify

managerial misconduct detrimental to their value maximization interests (Dyck and Zingales,

2004). Following the arguments in the literature that Big 4 auditors make financial statements

more credible (DeFond and Zhang, 2014), it can be argued that connected firms would prefer

appointing Big 4 auditors. In line with this idea, Fan and Wong (2005) claim that high

quality auditors might be hired by the management to increase the credibility of their

financial reports in emerging markets.

Firms with incentives to tunnel resources from minority shareholders require channels

through which this can be achieved. We consider RPTs as one such tunneling channel, and

investigate whether connected firms with certain types of RPTs differ in terms of auditor

choice, compared to their non-connected counterparts. A related party transaction is a transfer

Page 5: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

4

of resources, services, or obligations between related parties, regardless of whether or not a

price is charged (International Accounting Standards 24.9) (IASB, 2009), where a related

party is a person or entity related to the entity preparing its financial statements. These

transactions are diverse, and often complex, business transactions between a firm and its own

managers, directors, principal owners or affiliates, and are reasons for concern because they

violate arm’s-length market transaction principles. There is abundant empirical evidence in

the extant literature that supports the opportunistic incentive for engaging in RPTs, in

particular, RPT loan and loan guarantees (Jian and Wong, 2010; Kohlbeck and Mayhew,

2010; Chen et al., 2011; Ryngaert and Thomas, 2012; Ying and Wang, 2013).1 Since it is

easier for politically connected firms to conduct RPTs, because of their large numbers of

affiliates with complex interrelationships, we propose that politically connected firms

undertaking opportunistic RPTs would hire non-Big 4 auditors.

However, an alternative argument could be advanced from the supply side, i.e.,

auditors’ assessment of the audit risk associated with politically connected firms. Auditors’

evaluation of engagement risk is aimed at minimizing the auditor’s exposure to litigation loss,

loss of reputation, as well as regulation risk (Knechel et al., 2007). Since selection of clients

by auditors is non-random, good quality auditors might avoid auditing risky clients, in our

case, politically connected firms, because of their strong incentives for engaging in tunneling

and rent-seeking activities. This perspective suggests that auditors’ choice of clients might

drive the results, instead of the discretionary appointment of auditors by politically connected

firms. Since Big 4 auditors are of high quality, they should be more concerned about their

clients’ opportunistic business transactions, including opportunistic RPTs, and may choose to

avoid clients with political connections. This argument might also suggest a negative

association between political connections and auditor choice. However, empirical evidence

1 For example, Kohlbeck and Mayhew (2010) document a detrimental effect of disclosed RPTs on firm

valuations and subsequent returns when compared with non-RPT firms.

Page 6: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

5

from Malaysia suggests that auditors charge higher audit fees for politically connected firms

instead of resigning from the engagement (Gul, 2006).

Guedhami et al. (2014), an important related paper published recently, find that

politically connected firms are more likely to appoint Big 4 auditors compared to their non-

connected peers. We, on the other hand, propose that politically connected firms have

incentives to appoint non-Big 4 auditors. This is due to our consideration of firm-level RPTs

that encourage politically connected firms to siphon resources from minority shareholders.

Since Guedhami et al. (2014) did not investigate the possibility that connected firms might

engage in rent-seeking using opportunistic RPTs, our prediction of a negative association is

justified. That is, Guedhami et al. (2014) considered the incentives for connected firms to be

transparent, whereas we propose that connected firms might extract rents and, hence, choose

non-Big 4 auditors.

Indonesia offers an interesting setting to explore this research due to its unique

institutional features. First, political connections affect firm value in Indonesia (Fisman,

2001). By using an event study of rumors of the health of former President Suharto during

1995 to 1997, Fisman (2001) finds that returns of shares having a close relationship with

Suharto lost value compared to those having no affiliation with the Suharto government. In

addition, Leuz and Oberholzer-Gee (2006) document that the performances of politically

connected firms in Indonesia fluctuate following the fortunes of their connections. Second,

Indonesia has a high ownership concentration (Claessens et al., 2000; Brown, 2006), with an

average 16.6% of market capitalization being confined in the hands of a single family:

Sudono Salim, a closed ally of Suharto (Brown, 2006). This number rises to a staggering

57.7% for the top ten families in Indonesia, centered on Suharto political and military

connections (Claessens et al., 2000; Brown, 2006). Such a high ownership concentration

gives rise to a type II agency problem, i.e., the risk of expropriation of minority shareholders’

Page 7: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

6

resources by their controlling owners. Third, RPTs are significant in Indonesia, as more than

90 per cent of listed firms in Indonesia conduct various forms of RPTs (Utama and Utama,

2013).2 In addition, Indonesian listed companies are part of complex economic groups with

interlocking directorships, reciprocal ownership arrangement and excessive cross-

ownerships, which expedites RPTs. Although the Capital Market and Financial Institutions

Supervisory Agency requires listed firms to disclose RPTs, companies’ complex ownership

structures and lack of transparent disclosures makes the monitoring of RPTs difficult

(Wulandari and Rahman, 2005). Habib et al. (2016, forthcoming) find that politically

connected firms conduct more opportunistic RPTs compared to their non-connected

counterparts and further engage in income-increasing earnings management to mask such

opportunistic RPTs. This evidence corroborates politically connected firms’ incentives to

appoint non-Big 4 auditors in Indonesia.

For our study, a firm-year observation is categorized as politically connected if at

least one of its large shareholders (having control of at least 10 per cent of voters directly or

indirectly), or its board of directors or board of commissioners, is a current or former (a)

member of the parliament, (b) minister or head of a local government, or (c) closely related to

a politician or party. We further decompose political connections into one of three mutually

exclusive categories: namely, government connections, military connections, and Suharto

connections.

Using a panel data of 1,428 RPT firm-year observations from 2007 to 2013, we find

that politically connected firms with RPTs in Indonesia tend to choose non-Big 4 auditors.

2 The Indonesian Capital Market Supervisory Agency (ICMSA), requires listed firms to announce RPTs to

the public as well as to the ICMSA no later than two working days after the transactions are undertaken if

the RPT has a value larger than 0.5% of the firm’s paid capital and greater than IDR 5,000,000,000

(US$509,840). Hence, most of individual RPTs are arranged below these thresholds in order to avoid public

announcement (Utama and Utama, 2014). However, RPTs having a value less than 0.5% of the firm’s paid capital still need to be reported to the ICMSA and be disclosed in the notes to financial statements (See

Appendix II for further details on RPT regulations in Indonesia).

Page 8: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

7

This negative association is more pronounced for firms with government connections as

opposed to firms with military connections. Firms with Suharto connections, on the other

hand, tend to choose Big 4 auditors. Our results remain robust to controls for a potential self-

selection problem.

We extend the auditor choice research by investigating how the choice of auditors in

Indonesia is systematically affected by firms’ political connections (See Appendix I for a

discussion on the external auditing environment in Indonesia). We further contribute to the

auditor choice literature by documenting the important role played by firm-level RPTs. This

latter finding is particularly insightful as this provides a contextual explanation for why an

opposite result to that of Guedhami et al. (2014) can be expected. Finally, we enrich the

political connection literature, as applied in auditing, by examining the effects of three

mutually exclusive categories of political connections separately. We find that firms having

connections with the government are more inclined to choose non Big 4 auditors.

The remainder of the paper proceeds as follows. Section 2 reviews the relevant

literature and develops hypotheses. Section 3 describes research design followed by sample

selection and descriptive statistics in Section 4. The following section explains main test

results and Section 6 concludes the paper.

2. Literature Review and hypotheses development

The classic agency problem between shareholders and managers gives rise to the

hiring of auditors, who provide independent assurance to corporate stakeholders that financial

statements prepared by corporate managers comply with generally accepted accounting

principles (GAAP) (Watts and Zimmerman, 1983). Auditing also plays a significant role in

enforcing and protecting investors’ rights, by detecting expropriation by insiders (Newman et

Page 9: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

8

al., 2005), and benefits management by signalling the reliability of management-provided

financial information. A firm’s decision to appoint a certain type of auditor is, therefore, a

crucial element of the auditing landscape.

Previous research on the possible determinants of auditor choice (both brand name

and industry specialist auditors) include culture (Hope et al., 2008), firm-level governance,

e.g., ownership structure, country-level investor protection (Fan and Wong, 2005; Wang et

al., 2008; Guedhami et al., 2009; He et al., 2014), managerial incentives (Chen et al., 2015),

and political connections (Guedhami et al., 2014). Our study belongs to the ‘political

connection’ domain.

Many benefits accrue to politically connected firms: preferential access to lenders

(Johnson and Mitton, 2003; Khwaja and Mian, 2005; Faccio, 2006; Leuz and Oberholzer-

Gee, 2006; Boubakri et al., 2012a); low cost of debt and equity (Houston et al., 2014); high

likelihood of being bailed out (Faccio et al., 2006); profitable government contracts

(Goldman et al., 2009); favorable regulations (Goldman et al., 2009); less monitoring and

oversight (Faccio, 2006); lower taxes (Faccio, 2006; 2010); and preferential import licenses

and tariffs (Goldman et al., 2009). On the other hand, political connections are also viewed as

harmful to the minority shareholders, as these connections can lead to rent-seeking activities

(Frye and Shleifer, 1997; Faccio, 2006; Boubakri et al., 2012b), tunneling (Qian et al., 2011),

and earnings management (Chaney et al., 2011).

In considering whether to choose a Big 4 auditor, controlling shareholders will assess

potential benefits derived, and costs incurred, as a result of their choice (He et al., 2014).

Controlling shareholders might voluntarily adopt bonding mechanisms in dealing with

adverse pricing and a high cost of capital caused by asymmetric information and illiquidity

(Jensen and Meckling, 1976). Fan and Wong (2005) claim that external independent auditors

might be hired as monitors, or as one of the bonding mechanisms, designed to alleviate

Page 10: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

9

information and agency problems. Big 4 auditors might offer high quality audit because they

have better monitoring capability (Watts and Zimmerman, 1983), are keen to maintain their

reputation, and are subject to heightened litigation (Hope et al., 2008; Guedhami et al., 2014).

In addition, Big 4 auditors might deliver a high quality of assurance consistently, owing to

their global operation (Humphrey et al., 2009; Guedhami et al., 2014). For politically

connected firms, being audited by Big 4 auditors can, therefore, reap benefits, such as better

transparency, higher valuation, lower earnings management, and cheaper cost of capital

(Guedhami et al., 2014; He et al., 2014).

On the other hand, the appointment of a Big 4 auditor may be costly, since controlling

shareholders have fewer opportunities for expropriation owing to the significant role of Big 4

auditors in monitoring financial reporting discretion of controlling shareholders (Guedhami

et al., 2009). Politically connected firms expecting to obtain benefits from their allies, tend to

appoint non-Big 4 auditors for the following reasons. First, connected firms are inclined to

render more opaque financial statements in order to conceal their tunneling and rent seeking

activities (He et al., 2014; Chen et al., 2015). Second, any increase in transparency will

decrease the ability of controlling shareholders and their political allies to enjoy the private

benefits of control (Leuz and Oberholzer-Gee, 2006; Piotroski et al., 2015). For politically

connected firms, transparency is costly, since it might result in unwanted scrutiny, restricting

the possibility of exploiting weak corporate governance (Bona-Sánchez et al., 2014;

Piotroski et al., 2015). Further, He et al. (2014) claim that when controlling shareholders

develop political connections, their political partners need more secrecy, so that their

reputations are maintained at the expense of financial reporting transparency. Third, the lack

of transparency of non-Big 4 auditing does not reduce their chance of getting easy credit from

state-owned banks anyway (Dinç, 2005; Guedhami et al., 2014). Supporting this idea,

Bushman et al. (2004) claim that in return for bribes, nepotism, and other political supports,

Page 11: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

10

the politicians might exercise their control over state owned banks to provide preferential

financing to their connected firms. Financial reporting transparency may have a minimal role

in this setting, diminishing the incentive for appointing a more expensive Big 4 auditor.

Moreover, Leuz and Oberholzer-Gee (2006) find that Indonesian firms with close

connections to the state avoid raising capital from abroad with more onerous disclosure

requirements that include, but are not limited to, Big 4 audits of the financial statements.3

H1: Politically connected firms are less likely to appoint Big 4 auditors.

However, a supply side argument concerning auditor choice of clients may also be

relevant in understanding the association between political connection and auditing in

Indonesia. Because of the non-random selection of clients by auditors, good quality auditors

might avoid auditing risky clients, in our case, politically connected firms, because of their

tunneling and rent-seeking activities. An extensive body of literature supports the view that

Big 4 auditors provide higher audit quality because of their exposure to greater litigation risk

3 Another perspective holds that politically connected top managers could give outside investors an

impression that the government favors the connected firms, which could conceivably be more effective

than hiring high-quality auditors. In such a case, outside investors would pay less attention to audit quality, so that the firms would be less motivated to hire high-quality auditors. Investors might also agree that

building political connections is more cost-effective than hiring reputable auditors.

Page 12: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

11

(“deep pocket” theory) and greater reputation risk (independence concern) (DeFond and

Zhang, 2014). This argument suggests that Big 4 auditors should be more concerned about

their clients’ opportunistic business transactions, and may choose to avoid risky clients to

minimize their exposure to litigation and reputational losses. However, high quality auditors

also tend to be larger, with the capacity to diversify client risk and, thus, are better able to

bear the risk of such clients. Hence it is not obvious whether high quality auditors avoid risky

clients (DeFond and Zhang, 2014). Empirical evidence, too, finds support in favor of

retaining politically connected firms, but charging a fee premium (Gul, 2006).

In Indonesia, connected firms can be classified further into government, military, and

Suharto connections. Our sample covers two consecutive periods of Susilo Bambang

Yudhoyono (SBY)’s presidency, from 2004 to 2014. With respect to government connection,

extant literature, argues that government plays a key role in controlling and allocating key

resources (Child, 1994; Li et al., 2008). Firms willing to maintain an ongoing relationship

with government need to share the rents extracted through expropriation of minority

resources and, as well, obfuscate their financial reports to mask tunneling activities.

Appointing a non-Big 4 auditor is a proactive decision to accomplish this.

H1A: Firms having political connections to government are more likely to hire non-Big 4

auditors.

On the other hand, Suharto-connected firms are more likely to appoint Big 4 auditors.

When Suharto was in power, firms having an affiliation with his regime through his families,

friends, and military connections enjoyed ample privileges (Brown, 2006), e.g., preferential

loans from state owned banks through memo-lending and exclusive import licenses (Leuz

and Oberholzer-Gee, 2006). However, after Suharto’s resignation, firms having connections

with Suharto had reduced access to government officials. They had difficulties in establishing

Page 13: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

12

a connection with the new government, and experienced loss of government contracts,

distributorships, and brokerage monopolies (Fukuoka, 2013).

With respect to military influence in Indonesia, it has been observed that, during the

Suharto regime both active and former military personnel held strategic posts at the national

and regional level, including managerial positions in state owned enterprises (Brown, 2006;

Bhakti et al., 2009; Sebastian and Iisgindarsah, 2013). Suharto handed over state owned

enterprises, previously seized from Dutch companies, to be managed by military personnel.

However, with the end of the Suharto era, foundations belonging to the military, Suharto’s

family and Golkar were under investigation (Brown, 2006). Therefore, Mietzner (2006)

concludes that the army have lost formal political influence considerably, and they do not

serve as a backbone for the incumbent regime anymore. With these benefits gone, firms

having Suharto as well as the military connections had less incentive to engage in tunneling

and financial report manipulation in order to obfuscate such tunneling. Based on these

arguments we hypothesize the following:

H1B: Firms having Suharto as well as military connections are more likely to hire Big 4

auditors compared to firms with government connections.

Note that H1 is opposite to what Guedhami et al. (2014) established. This is due to

our consideration of firm-level RPTs that encourage politically connected firms to siphon

resources from minority shareholders. Since Guedhami et al. (2014) did not investigate the

possibility that connected firms might engage in rent-seeking using opportunistic RPTs, our

prediction of a negative association is justified. That is, Guedhami et al. (2014) considered

the incentives for connected firms to be transparent, whereas we propose that connected firms

might extract rents and, hence, choose non-Big 4 auditors.

Our next hypothesis explicitly considers the mechanisms through which connected

firms might conduct tunneling activities, and whether that had a bearing on auditor choice by

Page 14: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

13

connected firms. We propose that RPTs is one such channel through which politically

connected firms might conduct resource diversion. RPTs are used to structure transactions,

e.g., tunneling, propping or earnings management, among their affiliates in order for insiders

to expropriate minority resources (Thomas et al., 2004; Cheung et al., 2009). In the context of

Indonesia, Habib et al. (2016, forthcoming) document that politically connected firms

conduct more RPTs compared to their non-connected counterparts, and that this effect is

more pronounced for firms with government connections. Further analysis reveals that the

connected firms use RPTs to tunnel resources, and engage in income-increasing earnings

management to mask such tunnelling activities. Since it is easier for politically connected

firms to conduct RPTs owing to the presence of a large number of affiliates with complex

inter-relationships among them, we propose that politically connected firms with

opportunistic RPTs would hire non-Big 4 auditors.

Extant literature has documented that opportunistic RPTs are primarily conducted

through RPTs involving loan guarantees and capital transfers (RPLOAN). Prior literature

reveals that RP loans and guarantees have been used by parent companies for tunnelling or

siphoning resources out of their listed subsidiaries (Berkman et al., 2009; Jiang et al., 2010).

Empirical evidence shows that, compared to those with low levels of RP loans and

guarantees, Chinese firms with high levels of RP loans and guarantees demonstrate

significantly worse future performances including sharp declines in profitability, and a higher

likelihood of entering financial distress in the future (Jiang et al., 2010). Habib et al. (2016,

forthcoming) find that connected firms use primarily RP loans and guarantees to tunnel

resources in Indonesia. RP loans are generally not made as part of the normal course of

business, and most loans do not accrue interest. If RPTs involving loans allow politically

connected firms to siphon resources, then there is an incentive for those firms to manipulate

financial reports in order to obfuscate true economic performance. This argument suggests

Page 15: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

14

that politically connected firms with RPT loans are more likely to choose non-Big 4 auditors.

It is important to note that this argument does not imply that non-connected firms don’t use

RPT loans for opportunistic reasons. However, we do expect the effect to be more

pronounced for connected firms than for their non-connected counterparts. The following

hypothesis is developed:

H2: Firms with RP loan and guarantees are less likely to appoint Big 4 auditors and this is

more pronounced for politically connected firms.

Because we hypothesize that politically connected firms conduct opportunistic RPTs

for rent-seeking activities, the preceding hypotheses are valid in the context of such RPTs.

3. Research Design – Empirical Model

To test H1, we develop the following regression model:

)1.......(||1312

1110987

6543210

YearFEIndustryFEDACIC

ACROAFINANCESEGMENTINV

GROWTHLEVSIZEFOWNOWNCONPCONAUDITOR

where AUDITOR is dummy variable, coded 1 for Big 4 auditors and zero otherwise.

The local Indonesian audit firms that are affiliated with the Big 4 audit firms are:

Tanudiredja, Wibisana and Rekan (PWC); Purwantono, Suherman and Surja (EY); Osman

Bing Satrio and Rekan (Deloitte); and Siddharta Siddharta and Widjaja (KPMG). PCON is an

indicator variable coded 1 if the sample observations have political connections, 0 otherwise.

We expect a negative and significant coefficient on PCON to suggest that politically

connected firms will choose non-Big 4 auditors.

We include a set of control variables based on prior literature on the determinants of

auditor choice (DeFond et al., 2000; Wang et al., 2008; Guedhami et al., 2009, 2014). Larger

Page 16: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

15

firms (SIZE), firms with growth opportunities (GROWTH), firms with more inventories in

their balance sheet (INV), multi-segment firms (SEGMENT), new insurance of finance

(FINANCE), profitable firms (ROA), firms with larger foreign ownership (FOWN), and less-

leveraged (LEV) firms are more likely to appoint a Big 4 auditor. FOWN is the total

percentage of shares owned by foreign institutional investors. Foreign institutional investors

prefer Big 4 auditors to ensure the quality and the credibility of financial statements

(Guedhami et al., 2009; He et al., 2014). The association between ownership concentration

(OWNCON) and auditor choice is also expected to be positive (Fan and Wong, 2005).

SIZE is measured as the natural log of total assets. GROWTH is the market value of

equity divided by the book value of equity. ROA is net income divided by total assets. LEV is

the total long-term debt divided by total assets. AC is dummy variable coded 1 if the firm has

established an audit committee, and zero otherwise. We expect a positive coefficient on AC,

since independent audit committees demand higher quality audits (Carcello et al., 2002;

Abbott et al., 2003). IC is dummy variable coded 1 for a firm having independent

commissioners, and zero otherwise.4 We expect a negative and significant coefficient on

discretionary accruals (DAC) to suggest that the firms having higher DAC will prefer to

appoint non-Big 4 auditors in order to convey less information on their DAC. Appendix III

provides the variable definitions.5

4 Indonesia adopts a two-tier system for board structures: the board of directors (BODs) and the board of

commissioners (BOCs). The BODs serve as firm’ executives whereas the BOCs have responsibility to

supervise management policies and to advice the BODs. The role of the BOCs in the two-tier system is

comparable to that of the BODs in a one tier system. In order to improve corporate governance in

Indonesia, the Capital Market and Financial Institutions Supervisory agency requires listed firms to have

independent commissioners and audit committees. However, those boards did not carry out their legal

duties properly because both BOCs and BODs represent the interests of, and are often selected by the15

families who control the majority of the listed companies (Wulandari and Rahman, 2005). 5 We did not include some other control variables as in Guedhami et al. (2014) for reasons outlined below.

Data on FOREIGNSALE is not reported by companies in their published annual report. We did not include

CROSS-LIST because there are only two companies that are cross-listed [Aneka Tambang (Persero) Tbk

and Telekomunikasi Indonesia (Persero) Tbk]. CONTROLRIGHT (the percentage of voting rights belonging to the ultimate owner ) and CASHFLOWRIGHT (the percentage of voting rights belonging to

the ultimate owner) are unavailable because listed firms in Indonesia are only obliged to disclose their

legal owners instead of their beneficial or ultimate owners (Utama and Utama, 2013).

Page 17: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

16

To estimate DAC we use the cross-sectional modified Jones model, controlling for

firm performance (Kothari et al., 2005). We estimate the following model for all firms in the

same industry (using the SIC two-digit industry code) with at least eight observations in an

industry in a particular year:

)2......(..............................)()/(]/)[()/1(/

13

1211101

tt

tt tttttt ROA

TAPPE TA RECEIVABLE SALES TA TAACC

where ACC is total accruals, calculated as earnings before extraordinary items and

discontinued operations minus operating cash flows; TA is total assets in year t-1; ΔSALES is

change in sales from year t-1 to year t; ∆RECEIVABLE is change in accounts receivable from

year t-1 to year t; PPE is gross property plant and equipment; ROA is the prior year's return-

on-assets measured as earnings before extraordinary items and discontinued operations,

divided by total assets for the previous year. The coefficient estimates from Equation (2) are

used to estimate the non-discretionary component of total accruals (NDAC) for our sample

firms. The discretionary accruals are then the residual from equation (2), i.e. DAC=ACC-

NDAC.

An interesting aspect of the political connection landscape in Indonesia relates to the

different types of connection, which are not captured in a single PCON variable. We

categorize three mutually exclusive types of political connection: GCON (dummy variable, 1

for government connected firms, 0 otherwise); MCON (dummy variable, 1 for military

connected firms, 0 otherwise); and SCON (dummy variable, 1 for Suharto connected firms, 0

otherwise). The regression equation below is, therefore, estimated:

....(3).......... IndustryFEYearFEDACIC ACROAFINANCESEGMENTINVGROWTHLEVSIZE

FOWNOWNCONSCONMCONGCONAUDITOR

||1514131211

109876

543210

Page 18: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

17

We expect the coefficient on GCON to be negative, and that on MCON and SCON to

be positive.

Finally, we develop the following regression model to test for the effects of RPTs on

auditor choice, conditional on political connection:

)4...(||

*

151413

121110987

6543210

IndustryFEYearFEDACIC ACROAFINANCESEGMENTINVGROWTHLEV

SIZEFOWNOWNCONRPTPCONRPTPCONAUDITOR

We use three categories of RPTs. RPT_RATIO is ratio of gross RPT over total assets.

Deflation by assets is justified since RPTs vary with firm size. OPRPT_RATIO is operating

RPTs deflated by total assets. Finally, LOAN_RATIO is RPT loan deflated by total assets.

Operating RPTs, dealing with sales and purchases of goods and services, consist mainly of

trade relationships. From an efficiency-enhancing perspective, operating RPTs are argued to

be used within corporate groups as a way of optimizing internal resource allocation, reducing

transaction costs (Jian and Wong, 2010; Chen et al., 2012) and improving return-on-assets

(Khanna and Palepu, 2000). This is in contrast to RPT loan which is primarily a mechanism

for siphoning resources. Our variable of primary interest is the interactive variable

PCON*RPT. If politically connected firms make use of RPTs to siphon resources

(LOAN_RATIO), then we would expect a negative and significant coefficient on

PCON*LOAN_RATIO. The coefficient on PCON*OPRPT_RATIO is expected to be positive.

The coefficient on γ2 captures auditor choice by non-connected firms in the presence of

RPTs.

4. Sample selection and descriptive statistics

Data on the number and amount of RPT is hand-collected from audited financial

reports downloaded mainly from the website of the Indonesia Stock Exchange

Page 19: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

18

(http://www.idx.co.id/index-En.html). If not available, the data are derived from the websites

of Indonesian listed firms. In addition, the following corporate governance data are also

manually collected from audited financial statements or annual reports: board of directors,

board of commissioners, independent commissioners, audit committee, the names of audit

firms, the names and percentage of share ownerships, and information on reportable

segments. Financial statement data are collected from the Research Insight-Global Vantage

database. Since most of the data for RPT are in Indonesian Rupiah, the data for RPT are

translated into US$ by using the exchange rate available from the DataStream. Finally,

market data are retrieved from DataStream.

The criteria for defining politically connected firms follows Faccio (2006), Chaney et

al. (2011), and Guedhami et al. (2014) with necessary modification to the Indonesian context.

A firm-year observation will be categorized as politically connected if at least one of its large

shareholders (having at least 10 per cent direct or indirect voting rights) or its board of

directors and board of commissioners is a current or former (a) member of parliament, (b)

minister or head of a local government, or (c) is closely related to a politician or party.

Connection with government ministries is extended to close relatives (spouse, sons or

daughters, and other immediate family relationship). Close relationships with politicians or

parties encompass well-known friendships as identified by The Economist, Forbes or

Fortune; share ownership or a directorship held by former ministers, former heads of

government, former member of parliament and current politicians (Faccio, 2006; Chaney et

al., 2011); well documented relationships with political parties as utilized by Johnson and

Mitton (2003); and famous connections adopted by Fisman (2001) and Leuz and Oberholzer-

Gee (2006).

To establish those political connections, the names of boards of directors, boards of

commissioners and data of ownerships, including the name and the percentage of ownership,

Page 20: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

19

were gathered from the Indonesia Stock Exchange (http://www.idx.co.id/index-En.html),

company websites, audited financial reports and annual reports. The names of members of

parliament were collected from the website of the Indonesia House of Representatives

(http://www.dpr.go.id/id/anggota/), the names of members of cabinet were gathered from the

website of the cabinet secretariat of the Republic of Indonesia (http://setkab.go.id/en/profil-

kabinet.html). The names of heads of local government (governors) were collected from

(http://www.kemendagri.go.id/staff-directory/gubernur-dan-wakil-gubernur). The names of

members of parliament, members of cabinet, and heads of local government were matched

with the names of boards of directors, boards of commissioners and shareholders. In addition,

political connections could also be identified from the profiles of members of boards of

commissioners and directors described in the annual reports. Our primary variable of interest,

PCON, is an indicator variable coded 1 for the firm-year observation fulfilling at least one of

the abovementioned criteria, and 0 otherwise. We further categorize politically connected

firms into three mutually exclusive categories: namely, government connections, military

connections and Suharto connections.

The sample period is from 2007 until 2013, which covers two periods of the first

directly elected president, Susilo Bambang Yudhoyono (SBY)’s presidency, i.e., 2004-2014.

The first period of President SBY’s administration was from 20 October 2004 until 20

October 2009 and the second period was from 20 October 2009 until 20 0ctober 2014. The

financial year: 2014, is excluded from the observation owing to a political regime change

whereby a newly elected government took power.

Table 1, Panel A illustrates our sample selection process. The number of listed non-

financial firms on the Indonesian Stock Exchange in 2004 was 244 firms. This number

increased to 405 firms in 2013 giving us a total initial sample of 3,149 firm-year observations

for the period 2004 to 2013 inclusive. We deleted 772 firm-year observations pertaining to the

Page 21: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

20

2004 to 2006 sample period because we could retrieve valid data for only 25 firm-year

observations.6 From the remaining sample of 2,377 firm-year observations we deleted 481

firm-year observations with unavailable audit reports. This result in a total 1,896 firm-year

observations for matching with other variables required to run a regression on RPTs. We then

deleted 113 firm-year observations with zero RPT values and 85 firm-year observations with

negative equity values. Finally, we deleted a further 269 firm-year observations with missing

data on the relevant control variables, resulting in a final usable sample of 1,429 firm-year

observations for the period 2007 to 2013.

Panel B, Table 1 presents descriptive statistics for the variables used in the

regressions. We winsorize the continuous variables at 1% and 99% of their respective

distributions to control for the effects of outliers. About 43% of the firm-year observations

are audited by a Big 4 audit firm. Thirty nine percent of the observations have political

connections, split among GCON (24%), MCON (12%) and SCON (3%). The average

RPT_RATIO (Gross RPT/Total assets) is 0.50 of total assets. The corresponding values for

OPRPT_RATIO (Gross Operating RPTs/Total assets) and LOAN_RATIO (Gross

RPT_LOAN/Total assets) are 0.28 and 0.11 respectively.

These two categories do not add up to 0.50 because there are RPTs that do not belong

to operating and loan RPTs. However, we do not include other RPTs in our regression

specification because of a lack of theoretical prediction regarding its impact on auditor choice

by politically connected firms. Sample firms have growth opportunities (mean GROWTH is

6 We contacted the Financial Services Authority (FSA), formerly known as the Indonesian Capital Market

and Financial Institution Supervisory Agency, regarding the availability of hard copy annual reports. Since

05 July 2011, the FSA requires listed firms to submit both hard copy and soft copy audited financial

reports. However, the FSA does not allow public access to those audited financial reports. The Indonesia

Stock Exchange also used to receive hard-copies of audited financial reports, but those are kept in storage

outside of Jakarta. Currently, it maintains only soft-copies that can be downloaded from its website: a

procedure that we followed in collecting more recent annual reports.

Page 22: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

21

3.17), are low-levered (an average of 0.13), and profitable (average ROA of 7%). Eighty one

percent of the sample observations have established an audit committee.

The industry distribution of sample companies is presented in Panel C, Table 1,

revealing that materials account for 23.79% of the total sample observations, followed by

consumer discretionary and industrials with 20.08% and 17.35% of sample observations

respectively. Finally, Panel D presents a univariate test of the difference in means for the

variables between connected and non-connected firms. The proportion of firm years being

audited by Big 4 firms is much higher for connected as opposed to non-connected firms (an

average of 0.53 versus 0.37, t-stat of difference in means is 5.53). This is contrary to our

hypothesized negative association between PCON and AUDITOR. Politically connected firms

are larger, more levered, higher growth firms. There is no significant difference in gross RPT

ratios between connected and non-connected firms although OPRPT_RATIO is significantly

larger for the non-connected compared to their connected firm counterparts (t-stat of

difference in means is -1.72, p<0.10).

[TABLE 1 ABOUT HERE]

Correlations among the variables for auditor choice are presented in Table 2. The

correlation between PCON and AUDITOR is positive and significant at better than the 1%

level. This is contrary to our hypothesized negative association between PCON and

AUDITOR. Most of the independent variables are correlated positively with the dependent

variable, AUDITOR, at better than the 1% level. Untabulated correlation reveals that MCON

and SCON are positively correlated with AUDITOR, whilst GCON is negatively correlated.

[TABLE 2 ABOUT HERE]

5. Main test results

5.1 Political connections and auditor choice

Page 23: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

22

In a multivariate regression framework, we estimate the impact of political

connections on the likelihood that firms will hire a Big 4 auditor to examine the prediction in

H1. Next, we analyze the prediction in H2 that RPTs will moderate the association between

political connections and auditor choice. Table 3, Panel A presents the regression results for

the association between political connections and auditor choice (Big 4 versus non-Big 4)

(Columns 1 and 2) and the effect on that association of gross RPTs (Columns 3 to 5). We

control for industry and year fixed effects in all the regression models. We also report all

regression results with clustered standard errors at the firm level (Petersen, 2009; Gow et al.,

2010).

Our baseline regression in Column (1) reveals that the coefficient on PCON is

negative and significant (coefficient -0.363, z-statistic -2.22, p<0.05), suggesting that

politically connected firms are less likely to hire Big 4 auditors compared to their non-

connected counterparts. In terms of economic significance, the coefficient estimate of -0.363

means that political affiliations decrease the likelihood of appointing a Big 4 auditor by 5.8%

with all other variables are assigned their mean values.7 The negative coefficient could be

consistent with the argument that politically connected firms avoid Big 4 auditors in order to

conceal tunneling activities knowing that high quality auditors are competent to detect. This

supports H1. However, this negative coefficient could also be consistent with auditor’s risk

aversion argument, i.e., Big 4 audit firms might avoid auditing risky clients to mitigate their

7

Page 24: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

23

litigation and reputation losses. We perform additional tests in the sensitivity test section to

rule out this possibility.8

Column (2) reports this result with GCON showing a negative and significant

coefficient (coefficient -0.877, z-statistic -4.36, p<0.01) whilst that on SCON loads positively

(coefficient 2.358, z-statistic 4.05, p<0.01). In terms of economic significance, the coefficient

estimate of -0.877 on GCON means that government political affiliations decrease the

likelihood of appointing a Big 4 auditor by about 13% when all other variables are assigned

their mean values. These findings are consistent with H1A. However, we interpret the

coefficient on SCON cautiously, given the small number of observations. Among the other

firm-level determinants, we find that firm size, ownership concentration, foreign ownership,

growth, and the existence of an audit committee are related to auditor choice positively; while

leverage and profitability are related to auditor choice negatively.

γ

γ

8 We also performed a reverse regression technique whereby we regressed PCON on AUDITOR and the

remaining control variables. The untabulated result shows the coefficient on AUDITOR to be negative but

insignificant (coefficient -0.10, z-statistic -0.64).

Page 25: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

24

(4) and (5) consider the PCON categories, i.e., GCON and MCON,

and run Equation (4) without the interaction variable, while limiting the sample to PCON=1

only. The coefficient on RPT_RATIO is negative and significant for GCON only (coefficient -

2.417, z-statistic -4.96, p<0.01). The coefficient on MCON is insignificant and that on SCON

is undetermined because of the small sample size.

Page 26: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

25

[TABLE 3 ABOUT HERE]

5.3 Endogeneity tests

A firm’s decision to get politically connected is not random, and unobservable factors that

affect this decision may also be associated with the propensity to choose a certain type of

auditor. Selection problem, in this case, arise because one observes only the outcome of the

choice, in this case, political connection (1/0) made but not the outcomes of choices not

made. Selection bias, which is one form of endogeneity problem, can lead to inappropriate

inferences about treatment effects (Tucker, 2011). “Selection bias due to observables” arises

from a failure to control for differences researchers can observe, e.g., size, growth,

complexity, profitability. “Selection bias due to unobservable” arises because researchers use

a small set of observations. In the extant literature Heckman two-stage error correction

method has been the most popular and widely used approach for controlling the latter bias,

whereas the ‘Propensity score matching’ (PSM) has been used to control for bias due to

unobservable factors.

Page 27: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

26

To perform the Heckman test (1979) we proceed as follows. First we model firms’

decisions to form political connections using some observable firm characteristics based on

prior research (Faccio, 2006, 2010, Boubakri et al., 2008, Bunkanwanicha and

Wiwattanakantang, 2009). Lennox et al. (2012) argue that it is important to impose exclusion

restrictions in implementing the Heckman two-stage regression, even though the inverse

Mills ratio (IMR) can be identified by its nonlinear arguments. In other words, we need at

least one variable in the first-stage model that affects auditor choice through its effects on

political connections only.

Following Kim and Zhang (2016) and Guedhami et al. (2014), we include Industry %

of connected firms (%PCON_IND) and firms’ location (HQ) respectively, as the exclusion

variables. The variable %PCON_IND is related positively to the political connections of each

individual firm within the industry. However, we have no a priori reason to believe that

industry-level political affiliation has a direct impact on auditor choice through channels

other than political connections. Prior research suggests that firms located in capital cities are

more likely to form political connections (Agrawal and Knoeber, 2001), but are unlikely to be

associated with auditor choice. Our first-stage probit model takes the following form:

..(5)……………………………… +YearFE + FEIndustry ROASEGMENTGROWTH +LEV +SIZE +FOWN +OWNCON +HQ%PCON_IND =PCON

98

7654321

%PCON_IND is the percentage of politically connected firms in a firm’s industry

group. HQ is a dummy variable coded 1 if the firm is headquartered in the capital city and

zero otherwise. Other variables are defined as before. Table 4, Column (1) reports the first

stage estimation model. As predicted, the coefficients on %PCON_IND and on HQ, are

significantly positive [(coefficients 1.311 (p<0.1) and 0.219 (p<0.05)]. We calculate IMR

from the first stage probit model, and include it as an additional independent variable in the

Page 28: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

27

second stage regression model. We continue to find results that are consistent with results in

Panel B in Table 3. The coefficient on the interactive variable PCON*LOAN_RATIO is

negative and marginally significant (coefficient -1.51, z-statistic -1.89, p<0.10), while that on

PCON*OPRPT_RATIO is insignificant. We find the coefficient on IMR to be insignificant

suggesting that self-selection does not confound our results.

[TABLE 4 ABOUT HERE]

Propensity-matched (PSM) technique

PSM technique is utilized to mitigate selection problem arising from observables. Matching

on firm characteristics (covariates) is ideal when the number of characteristics over which the

treated and control groups differ is limited. Rosenbaum and Rubin (1983) propose matching

by a function of covariates: the probability of an individual being selected into the program

(treatment group). This matching method is referred to as “propensity score matching”

(PSM). We use nearest neighbor (NN), and kernel technique to perform the PSM model. The

NN procedure with replacement picks a single control firm according to the closest

propensity score. Kernel matching uses the entire sample of control firms as matches, where

each unit is weighted in proportion to its closeness to the treated observation (Rosenbaum and

Rubin 1983; 1985).

Proper implementation of PSM requires both the treatment and control groups to be

similar across a number of firm characteristics excluding the main variable on which they are

expected to differ, in our case, auditor. Therefore, we first document the covariates matching,

based on the calculated propensity score

Results are reported in Panel A in Table

5. Covariate balance is achieved if both the treatment and control groups appear similar along

Page 29: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

28

their observable dimensions, except for their choice of auditors.

Panel B, Table 5 shows the PSM regression results using the NN approach in

Columns (1) to (4) and Kernel technique in Columns (5) to (8). We find results that are

consistent with the main results: (i) politically connected firms choose non-Big 4 auditors, (ii)

firms with government and military connection choose non Big 4 auditor whereas firms with

Suharto connection choose Big 4 auditors, (iii) connected firms with RPTs choose non Big 4

auditors which is more pronounced for RP loans. Overall, our PSM analysis provides robust

evidence about the association between political connection and auditor choice and the

mediating role of firm-level RPTs.

[TABLE 5 ABOUT HERE]

6. Conclusions

This study investigates the association between political connections and auditor

choice in Indonesia. We also investigate whether firm-level RPTs moderate the association

between the two. Our study is motivated by conflicting views on whether politically

connected firms appoint reputable auditors. We find that politically connected firms in

Indonesia tend to appoint less reputable auditors. Connected firms have been found to engage

in tunneling and rent-seeking activities in order to establish and maintain their political

connections. Given the value of political connections (e.g. Fisman, 2001), politically

connected firms may manipulate accounting numbers to conceal their true economic

performance and, hence, prefer less reputable auditors. We also find that connected firms’

Page 30: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

29

preference for appointing less reputable auditors is more pronounced when these firms

engage in significant RPTs.

The findings from this study might be generalized to other emerging countries, which

have characteristics similar to those of Indonesia, where political connections hold a

significant role, and politically connected firms conduct considerable RPTs. The findings of

the research might benefit prospective investors and minority shareholders, who have limited

information and knowledge of the true economic incentives for RPTs conducted by listed

firms that have political connections. The minority shareholders should be aware that RPTs

might be used opportunistically by controlling shareholders of politically connected firms to

carry out tunnelling that is followed by subsequent earnings management. In addition, with

regard to auditor choice, the findings of this study might benefit prospective investors and

minority shareholders regarding the rationale as to why politically connected firms appoint

non-Big 4 auditors. The research findings show that the detrimental effect to minority

interests of RPT loans conducted by politically connected firms justifies the regulatory

restrictions on RPTs and the disclosure requirements imposed on listed firms.

Page 31: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

30

References

Abbott, L. J., Parker, S., Peters, G. F., Raghunandan, K. 2003. The association between audit

committee characteristics and audit fees. Audit. J. Pract.Th. 22, 17-32.

Agrawal, A., Knoeber, C.R. 2001. Do some outside directors play a political role? J. Law. Econ. 44, 179–98.

Armstrong, C.S., Jagolinzer, A.D., Larcker, D.F. 2010. Chief Executive Officer equity incentives and

accounting irregularities. J. Account. Res. 48, 225-271.

Basioudis, I. G., Fifi, F. 2004. The market for professional services in Indonesia. Int. J. Aud. 8, 153–164.

Berkman, H., Cole, R. A., Fu, L. J. 2009. Expropriation through loan guarantees to related parties:

Evidence from China. J. Bank. Finan. 33, 141-156. Bhakti, I. N., Yanuarti, S., Nurhasim, M. 2009. Military politics, etnnicity and conflict in Indonesia.

Centre for Research on Inequality, Human Security and Ethnicity - Working Paper, 62, 1-36.

Bona-Sánchez, C., Pérez-Alemán, J., Santana-Martín, D. J. 2014. Politically connected firms and earnings informativeness in the controlling versus minority shareholders context: European

evidence. Corp. Gov. 22, 330–346.

Boubakri, N., Cosset, J.C., Saffar, W. 2008. Political connections of newly privatized firms. J.

Corp. Fin. 14, 654–73. Boubakri, N., Cosset, J.C., Saffar, W. 2012a. The impact of political connections on firms’ operating

performance and financing decisions. J. Fin. Res. 35, 397-423.

Boubakri, N., Guedhami, O., Mishra, D., Saffar, W. 2012b. Political connections and the cost of equity capital. J. Corp. Fin. 18, 541-559.

Brown, R. A. 2006. The rise of the corporate economy in Southeast Asia. Oxon, United Kingdom,

Routledge.

Bushman, R. M., Piotroski, J. D., Smith, A. J. 2004. What determines corporate transparency? J. Account. Res. 42, 207-252.

Carcello, J. V., Hermanson, D. R., Neal, T. L., Riley, R. A. 2002. Board characteristics and audit fees.

Contem. Account. Res. 19, 365-384.

Page 32: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

31

Chaney, P. K., Faccio, M., Parsley, D. 2011. The quality of accounting information in politically

connected firms. J. Account. Econ. 51, 58-76. Chen, C.L., Hsu, C.S., Kung, F.H. 2015. Political connections, managerial incentives and auditor

choice: Evidence from China. Pacific Account. Rev. 27, 441-465.

Chen, J. J., Chen, P., Xiao, X. 2011. Related Party transactions as a source of earning management.

App. Fin. Econ. 21, 165-181. Chen, S., Wang, K., Li, X. 2012. Product market competition, ultimate controlling structure and

related party transactions. China J. Account. Res. 5, 293-306.

Cheung, Y.L., Jing, L., Lu, T., Rau, P. R., Stouraitis, A. 2009. Tunneling and propping up: An analysis of related party transactions by Chinese listed companies. Pacific-Basin Fin. J. 17,

372-393.

Child, J. 1994. Management in China during the age of reform. Cambridge, Cambridge University Press.

Claessens, S., Djankov, S., Lang, L. H. P. 2000. The separation of ownership and control in East

Asian corporations. J. Fin. Econ. 58, 81-112.

Davis, L. R., Ricchiute, D. N., Trompeter, G. 1993. Audit effort, audit fees, and the provision of nonaudit services to audit clients. Account. Rev. 68, 135-150.

DeFond, M. L., Wong, T. J., Li, S. 2000. The impact of improved auditor independence on audit

market concentration in China. J. Account. Econ. 28, 269-305. DeFond, M. L., Zhang, J. 2014. A review of archival auditing research. J. Account. Econ. 58, 275-

326.

Dinç, I. S. 2005. Politicians and banks: Political influences on government-owned banks in emerging markets. J. Fin. Econ. 77, 453–479.

Dyck, A., Zingales, L. 2004. Private benefits of control: An international comparison. J. Fin. 59.

Faccio, M. 2006. Politically connected firms. Am. Econ. Rev. 96, 369-386.

Faccio, M. 2010. Differences between politically connected and nonconnected firms: A cross-country analysis. Fin. Mgt. Autumn, 905-927.

Faccio, M., Masulis, R. W., Mcconnell, J. J. 2006. Political connections and corporate bailouts. J. Fin.

61. 2597-2635. Fan, J. P. H., Wong, T. J. 2005. Do external auditors perform a corporate governance role in emerging

market? Evidence from East Asia. J. Account. Res. 43, 35-72.

Financial Conduct Authority. 2015. Asia Resource Minerals plc (formerly Bumi plc) fined £4.65m for

breaching the Listing Rules. Financial Conduct Authority. Available: https://www.fca.org.uk/news/asia-resource-minerals-plc-formerly-bumi-plc-fined-4-65m-for-

breaching-the-listing-rules# [Accessed 19 October 2015].

Fisman, R. 2001. Estimating the value of political connections. Am. Econ. Rev. 91, 1095-1102. Frye, T., Shleifer, A. 1997. The invisible hand and the grabbing hand. Am. Econ. Rev. 87, 354-358.

Fukuoka, Y. 2013. Indonesia's 'democratic transition' revisited: A clientelist model of political

transition. Democratization 20, 991-1013. Goldman, E., Rocholl, J., So, J. 2009. Do politically connected boards affect firm value? Rev. Fin. St.

22, 2331-2360.

Gow, I. D., Ormazabal, G., Taylor, D. J. 2010. Correcting for cross‐sectional and time‐series

dependence in accounting research. Account. Rev. 85, 483-512. Guedhami, O., Pittman, J. A., Saffar, W. 2009. Auditor choice in privatized firms: Empirical evidence

on the role of state and foreign owners. J. Account. Econ. 48, 151-171.

Guedhami, O., Pittman, J. A., Saffar, W. 2014. Auditor choice in politically connected firms. J. Account. Res. 52, 107-162.

Gul, F.A. 2006. Auditors’ response to political connections and cronyism in Malaysia. J. Account.

Res. 44(5), 931-963.

Habib, A., Haris, A., Jiang, H. 2016. Political connections and related party transactions: Evidence from Indonesia. Int. J. Account. Forthcoming.

He, X., Rui, O., Zheng, L., Zhu, H. 2014. Foreign ownership and auditor choice. J. Account. Public

Pol. 33, 401-418. Heckman, J. J. 1979. Sample-selection bias as a specification error. Econometrica 47, 153-62.

Page 33: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

32

Hope, O.K., Kang, T., Thomas, W., Yoo, Y. K. 2008. Culture and auditor choice: A test of the

secrecy hypothesis. J. Account. Public Pol. 27, 357–373. Houston, J. F., Jiang, L., Lin, C., Ma, Y. 2014. Political connections and the cost of bank loans. J.

Account. Res. 52, 193-243.

Humphrey, C., Loft, A., Woods, M. 2009. The global audit profession and the international financial

architecture: Understanding regulatory relationships at a time of financial crisis. Account. Org. Society. 34, 810–825.

IASB 2009. Related Party Disclosures. International Accounting Standard 24. Available at:

http://www.iasplus.com/en/standards/ias/ias24. Jensen, M. C., Meckling, W. H. 1976. Theory of the firm: Managerial behavior, agency costs and

ownership structure. J. Fin. Econ. 3, 305–360.

Jian, M., Wong, T. J. 2010. Propping through related party transactions. Rev. Account. St. 15, 70-105. Jiang, G., Lee, C. M. C., Yue, H. 2010. Tunnelling through intercorporate loans: The China

experience. J. Fin. Econ. 98, 1-20.

Johnson, S., Mitton, T. 2003. Cronyism and capital controls: Evidence from Malaysia. J. Fin. Econ.

67, 351–382. Khanna, T., Palepu, K. 2000. Is group affiliation profitable in emerging markets? An analysis of

diversified Indian business groups. J. Fin. 55, 867-891.

Khwaja, A. I., Mian, A. 2005. Do lenders favor politically connected firms? Rent provision in an emerging financial market. Quart. J. Econ. 120, 1371-1411.

Kim, C.F., Zhang, L. 2016. Corporate political connections and tax aggressiveness. Contem. Account.

Res. 33, 78-114. Knechel, R.W., Salterio, S., Ballou, B., 2007. Auditing: Assurance and Risk. South-Western College

Pub.

Kothari, S. P., Leone, A. J., Wasley, C. E. 2005. Performance matched discretionary accrual

measures. J. Account. Econ. 39, 163-197. Kohlbeck, M., Mayhew, B. W. 2010. Valuation of firms that disclose related-party transactions. J.

Account. Public Pol. 29, 115–137.

Lennox, C.S., Francis, J.R., Wang, Z. 2012. Selection models in accounting research. Account. Rev. 87(2), 589–616

Leuz, C., Oberholzer-Gee, F. 2006. Political relationships, global financing, and corporate

transparency: Evidence from Indonesia. J. Fin. Econ. 81, 411-439.

Li, H., Meng, L., Wang, Q., Zhou, L.A. 2008. Political connections, financing and firm performance: Evidence from Chinese private firms. J. Dev. Econ. 87, 283–299.

Ma, L., Ma, S., Tian, G. 2013. Political connections, founder-managers, and their impact on tunneling

in China's listed firms. Pacific-Basin Fin. J. 24, 312-339. Mietzner, M. 2006. The politics of military reform in post-Suharto Indonesia: Elite conflict,

nationalism, and institutional resistance. Washington, D.C., East-West Center Washington.

Morck, R. K., Stangeland, D. A., Yeung, B. 2000. Inherited wealt, corporate control, and economic growth: The Canadian disease. In: Morck, R. K. (ed.) Concentrated Corporate Ownership.

Chicago: University of Chicago Press.

Newman, D. P., Patterson, E. R., Smith, J. R. 2005. The role of auditing in investor protection.

Account. Rev. 80, 289-313. Petersen, M. A. 2009. Estimating standard errors in finance panel data sets: Comparing approaches.

Rev. Fin. St. 22, 435-480.

Piotroski, J. D., Wong, T. J., Zhang, T. 2015. Political incentives to suppress negative information: Evidence from Chinese listed firms. J. Account. Res. 53, 405-459.

Qian, M., Pan, H., Yeung, B. 2011. Expropriation of minority interest shareholders in politically

connected firms. Working Paper - NUS Business School. Rosenbaum, P. R., Rubin, D. B. 1983. The central role of the propensity score in observational studies

for causal effects. Biometrika 70, 41-55.

Rosenbaum, P. R., Rubin, D. B. 1985. Constructing a control group using multivariate matched

sampling methods that incorporate the propensity score. Am. Stat. 39, 33-38. Ryngaert, M. D., Thomas, S. 2012. Not all related-party transactions (RPTs) are the same: ex ante

versus ex post RPTs. J. Account. Res. 50, 845–882.

Page 34: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

33

Sebastian, L. C., Iisgindarsah 2013. Taking stock of military reform in Indonesia. In: Rüland, J.,

Manea, M.G., Born, H. (eds.) The Politics of Military Reform Expoeriences from Indonesia and Nigeria. Verlag Berlin Heidelberg Springer.

Thomas, W. B., Herrmann, D. R., Inoue, T. 2004. Earnings management through affliated

transactions. J. Int. Account. Res. 3, 1-25.

Tucker, J. W. 2010. Selection bias and econometric remedies in accounting and finance research. J. Account. Lit. 29, 31-57.

Utama, C. A., Utama, S. 2013. Corporate governance, size and disclosure of related party

transactions, and firm value: Indonesia evidence. Int. J. Disc. Gov. 1-25. Utama, C. A., Utama, S. 2014. Determinants of disclosure level of related party transactions in

Indonesia. Int. J. Disc. Gov. 11, 74-98.

Wang, Q., Wong, T. J., Xia, L. 2008. State ownership, the institutional environment, and auditor choice: Evidence from China J. Account. Econ. 46, 112– 134.

Watts, R. L., Zimmerman, J. L. 1983. Agency problems, auditing, and the theory of the firm: Some

evidence. J. Law. Econ. 26, 613-633.

Whisenant, S., Sankaraguruswamy, S., Raghunandan, K. 2003. Evidence on the joint determination of audit and non-audit fees. J. Account. Res. 41, 721-744.

Wulandari, E. R., Rahman, A. R. 2005. Political patronage, cross-holdings and corporate governance

in Indonesia. In: Gul, F. A., Tsui, J. S. L. (eds.) The governance of East Asian corporations. N.Y.: Palgrave Macmillan.

Ying, Q., Wang, L. 2013. Propping by controlling shareholders, wealth transfer and firm

performance: Evidence from Chinese listed companies. China J. Account. Res. 6, 133-147.

Appendix I: Audit environment in Indonesia

The Indonesian Institute of Accountants9 has responsibility for establishing accounting

standards, auditing standards, and ethics codes for accountants. The International aid agencies

such as the World Bank and IMF and foreign investors recommend that the government both

enforce tougher disclosure rules and harmonize accounting practices in Indonesia. In

response to those requests, following the enactment of the Sarbanes-Oxley Act in July 2002,

the Indonesian government issued the Minister of Finance Regulation Number

423/KMK.06/2002 dated 30 September 2002 regarding Public Accountant Services10

.

Basioudis and Fifi (2004) claim that accounting standards and corporate governance

regulations in Indonesia have been influenced significantly by the developed western

countries and, sometimes, go even further than those of western regulations. They point to

9 Ikatan Akuntan Indonesia (IAI - the Indonesian Institute of Accountants) is the organization officially

recognized by the government. IAI was established on 23 December 1957. 10

The amendment of the previous Minister of Finance Regulation number 43/KMK.017/1997. This regulation was further revised by the issuance of the Minister of Finance Regulation number

17/PMK.01/2008, dated 05 February 2008, and was upgraded to the law by the enactment of Law number

5, year 2011, regarding Public accountants on 03 May 2011.

Page 35: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

34

auditor rotation as stipulated in Regulation number 423/KMK.06/2002, in which audit firms

are not allowed to audit an entity for five consecutive years, and audit partners can audit an

entity for only three consecutive years11

. Further, they argue that the requirements for audit

rotation in Indonesia are more burdensome compared to those of other jurisdictions in the

world, as even the Sarbanes-Oxley Act did not require rotation of audit firms at all, or

rotation of audit partners until after five years.

All financial statements were to be prepared in accordance with the Indonesian

accounting standards which are mostly derived from the International and US accounting

standards. Further, financial statements were to be audited according to the Indonesian

auditing standards, which are significantly similar to the generally accepted auditing

standards in the US. In providing an opinion on the fairness of financial statements, auditors

refer primarily to the Indonesian Accounting Standards, followed by relevant domestic rules

and regulations, and International Accounting Standards for matters that are not covered by

the abovementioned regulations ((Basioudis and Fifi, 2004).

Wulandari and Rahman (2005) point out that even though the Indonesian accounting

standards are substantially similar to the International Accounting Standards, some

Indonesian companies still have low quality financial reports, owing to the lack of effective

and efficient enforcement mechanisms. Further, they argue that auditors might have been

responsible for playing a role in the Asian financial crisis, as many of listed companies were

granted unqualified opinions prior to the crisis, but went bankrupt during the financial crisis.

They claim that auditors had little understanding and awareness of accounting or auditing

standards. In addition, auditors’ opinions were sometimes based on the work of other

professionals, such as appraisers, when they took for granted the value of certain assets

belonging to the auditee (Wulandari and Rahman, 2005), and auditors rely heavily on 11

The rotation of auditors was revised by the issuance of the Minister of Finance Regulation number

17/PMK.01/2008 by which rotation of audit firms is now required after six consecutive years (one year

longer than the previous regulation).

Page 36: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

35

management representations without conducting further investigation to gather corroborative

evidence in supporting their audit opinions (Rahman, 2000 as cited in Wulandari and

Rahman, 2005).

Empirical evidence on whether audit quality is valuable in Indonesia is not available.

We used our data to provide some evidence on this important issue. Since extant research has

established that financial reporting quality, as proxied by less earnings management, is better

for firms audited by Big 4 and industry specialist audit firms, we regress abnormal accruals

on auditor identity and other control variables. Untabulated results show the coefficient on

AUDITOR, is significantly negative (coefficient -0.023, t-stat -2.41), suggesting that Big 4

auditors in Indonesia constrain earnings management. This result may provide a perspective

on connected firms’ reluctance to appoint Big 4 audit firms, i.e., to conceal their tunneling

and rent-seeking activities. We also regress audit opinion (OPINION) (a dummy variable

coded 1 if clients receive a qualified audit opinion, and 0 otherwise) on the AUDITOR

indicator variable and other relevant determinants of OPINION. Untabulated results show the

coefficient on AUDITOR is negative and significant (coefficient -0.42, z-statistic -1.85,

p<0.10). However, this effect is pronounced for the PCON=1 group only (coefficient -0.76,

z-statistic -1.75, p<0.10). This suggests that connected firms tend to get more qualified

opinions compared to their non-connected counterparts and this, again, might explain

connected firms’ preference for non-Big 4 auditors.

Page 37: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

36

Appendix II: Development of RPT regulations in Indonesia

Bapepam-LK enacted Regulation Number VIII.G.7, which requires extensive disclosure on

RPTs in the notes to financial statements. Regulation Number VIII.G.7 under the Decree of

the Head of the Indonesian Capital Market Supervisory Agency number KEP-06/PM/2000

dated 13 March 2000 requires issuers and listed firms to disclose the breakdown and the total

amount of RPTs under each category of assets, liabilities, sales, and purchases (expenses),

along with their percentage to total assets, total liabilities, total sales, and total purchases

(expenses); nature, types and elements of RPTs; pricing policies and terms of RPTs and a

statement that RPTs are similar to those undertaken with third parties; reasons and basis for

the recognition of doubtful accounts for RPT receivables. If the amount of a RPT for each

category exceeds IDR 1,000,000,000 (US$ 101,96812

) or if non-core business RPTs are

conducted, separate disclosures and explanations are needed. From 25 June 2012, a new RPT

regulation was enacted, i.e., Regulation Number VIII.G.7, under a Decree of the Head of the

Indonesian Capital Market Supervisory Agency number KEP-347/BL/2012, in which the 12

(1US$=IDR 9,807 in 2013 (DataStream)

Page 38: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

37

disclosure requirements are quite similar, but RPTs are now classified based on three

categories of parties conducting the RPT, namely, family members, related parties, and

government affiliated.

In addition, Regulation Number IX.E.1 under the Decree of the Head of the

Indonesian Capital Market Supervisory Agency, number KEP-412/BL/2009, dated 25

November 2009, requires issuers and listed firms to announce to the public any RPT, and

report the evidence of that announcement to the Indonesian Capital Market Supervisory

Agency no later than two working days after the transactions are undertaken, except for

RPTs having a value less than 0.5% of the firm’s paid capital and less than IDR

5,000,000,000 (US$509,840), in which case no public announcement is required, but such an

RPT shall be reported to the Indonesian Capital Market Supervisory Agency. In addition, if

RPTs are identified as conflict of interest transactions13

, they need to get approval from the

independent shareholders, except for RPTs having a low value as specified above.

Recent RPT cases in the United Kingdom involved an Indonesian listed firm. Asia

Resource Mineral Plc (formerly Bumi Plc) suffered fines in the amount of £4,651,200 from

the Financial Conduct Authority because its subsidiary, namely, PT Berau Coal Energy Tbk

(listed in Indonesian Stock Exchange) conducted an RPT which breached UKLA Listing

Rules (Authority, 2015). Those two companies are controlled by Bakrie Group, and are

classified as politically connected firms in our study.

13

According to Bapepam-LK rule IX.E.I, conflict of interest takes place in the transaction when the

economic interest of the firm differs from the personal interest of directors, commissioners, or controlling

shareholders, such that the transaction might ruin the firm.

Page 39: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

38

Appendix – III: Variable definitions

Variables Definitions

AUDITOR Dummy variable, 1 for the firm audited by Big 4 auditors, 0 for otherwise.

PCON Dummy variable, 1 for politically connected firms, 0 for otherwise.

GCON Dummy variable, 1 for government connected firms, 0 for otherwise. MCON Dummy variable, 1 for military connected firms, 0 for otherwise.

SCON Dummy variable, 1 for Suharto connected firms, 0 for otherwise.

RPT_RATIO Gross RPT divided by total assets of the firm.

OPRPT_RATIO Gross operating RPTs divided by total assets of the firm. LOAN_RATIO Gross RPTLOAN divided by total assets of the firm.

OWNCON Total percentage of shares owned by the five largest shareholders

FOWN Total percentage of shares owned by foreign institutional investors. SIZE Natural logarithm of total assets.

LEV Total long term debt divided by total assets.

GROWTH Market value of equity divided by book value of equity. INV The ratio of inventory to total assets

SEGMENT Natural logarithm of number of business segments

FINANCE A dummy variable that takes the value of one if the sum of new long-term debt

plus new equity exceeds 20% of total assets ROA Return on assets (earnings before extraordinary items plus discontinued operations

for the preceding year divided by total assets for the same year).

AC Dummy variable, 1 for the firm having audit committee, 0 for otherwise. IC Dummy variable, 1 for the firm having independent commissioners, 0 for

otherwise.

ACC Total accruals derived from earnings before extraordinary items and discontinued operations minus operating cash flows.

|DAC| Absolute discretionary accruals calculated with the Modified Jones model (1995).

To estimate DAC we use the cross-sectional modified Jones model, controlling for

firm performance (Kotari et al., 2005). We estimate the following model for all firms in the same industry (using economic sector code):

Page 40: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

39

)2.....()()/(]/)[()/1(/

3

1211101

t

tt tttttt

ROA TAPPE TADEBTORS SALES TA TAACC

The coefficient estimates from Equation (2) are used to predict non-discretionary

component of total accruals (NDAC) for our sample firms. Thus, discretionary accruals is the residual from equation (2), i.e. DAC=ACC-NDAC.

ΔSALES Change in sales from year t-1 to year t.

∆DEBTORS Change in accounts receivable from year t-1 to year t. PPE Gross property, plant, and equipment.

ROA Return on assets (earnings before extraordinary items plus discontinued operations

for the preceding year divided by total assets for the same year).

TABLE 1: Sample selection procedure and descriptive statistics

PANEL A: Sample selection procedure

Sample Selection Process Observations Number of non-financial firm-year observations from 2004 to 2013 3,149

Less: Firm year observations pertaining to 2004-2006 sample period dropped because of

too few available annual reports

( 772)

Less: Firm year observations with unavailable audit reports during 2007-2013 period ( 481)

Less: Firm year observations with zero RPT values ( 113)

Less: Number of firm-year observations with negative book value (distress firms) ( 85)

Number of firm-year observations with complete non zero RPT data 1,698

Less: missing data of other control variables ( 269)

Number of firm-year observations for the baseline regression 1,429

PANEL B: Descriptive statistics

Variables Mean SD 25% Median 75%

AUDITOR 0.43 0.50 0.00 0.00 1.00

PCON 0.39 0.49 0.00 0.00 1.00

GCON 0.24 0.43 0.00 0.00 0.00

MCON 0.12 0.32 0.00 0.00 0.00

SCON 0.03 0.19 0.00 0.00 0.00

RPT_RATIO 0.50 0.88 0.04 0.21 0.66

OPRPT_RATIO 0.28 0.57 0.00 0.03 0.27

LOAN_RATIO 0.11 0.29 0.00 0.01 0.08

OWNCON 0.71 0.19 0.59 0.74 0.85

FOWN 0.28 0.30 0.00 0.15 0.52

SIZE 19.00 1.67 17.89 18.99 20.15

LEV 0.13 0.15 0.00 0.08 0.21

GROWTH 3.17 5.52 0.94 1.66 3.28 INV 0.16 0.16 0.02 0.12 0.24

SEGMENT 0.90 0.53 0.69 1.1 1.39

Page 41: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

40

FINANCE 0.14 0.34 0.00 0.00 0.00

ROA 0.07 0.12 0.01 0.05 0.11

AC 0.81 0.39 1.00 1.00 1.00

IC 0.92 0.27 1.00 1.00 1.00

|DAC| 0.10 0.10 0.03 0.07 0.13

PANEL C: Industry distributions

Sector Code Economic Sector Description Observations % of

observations 1000 Materials 340 23.79

2000 Consumer Discretionary 287 20.08

3000 Consumer Staples 222 15.54

3500 Health Care 42 2.94

4000 Energy 41 2.87

5000 Real Estate Management & Development 168 11.76

6000 Industrials 248 17.35

8000 Information Technology 40 2.80

8600 Telecommunication Service 41 2.87

1,429 100.00

Panel D: Univariate test Variables PCON=1 [n=562] PCON=0 [n=867] t-test of difference in mean

AUDITOR 0.53 0.37 5.53***

RPT_RATIO 0.47 0.51 -1.02

OPRPT_RATIO 0.24 0.30 -1.72*

LOAN_RATIO 0.11 0.12 -0.42

OWNCON 0.51 0.50 0.68

FOWN 0.23 0.31 -4.55***

SIZE 20.03 18.32 21.95***

LEV 0.16 0.11 5.23***

GROWTH 3.73 2.81 3.08***

INV 0.14 0.18 -4.32***

SEGMENT 3.06 2.55 7.29***

FINANCE 0.14 0.13 0.75

ROA 0.08 0.06 1.88*

AC 0.89 0.76 6.00***

IC 0.91 0.92 -0.57

|DAC| 0.09 0.10 -0.99

Note: Variable definitions are in Appendix III. ***, ** are significant at p<0.01 and <0.05 respectively (two-tailed test)

Page 42: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

41

TABLE 2: Correlation analysis

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

AUDITOR (1) 1.00

PCON (2) 0.15 1.00

RPT_RATIO (3) -0.11 -0.03 1.00

LOAN_RATIO (4) 0.14 -0.05 0.33 1.00

OPRPT_RATIO (5) -0.17 -0.01 0.59 -0.11 1.00

OWNCON(6) -0.04 0.11 0.49 0.08 0.12 1.00

FOWN (7) 0.18 -0.15 -0.04 0.17 -0.02 -0.06 1.00

SIZE (8) 0.16 -0.12 0.04 0.14 0.00 -0.03 0.30 1.00

LEV (9) 0.40 0.50 -0.02 -0.01 -0.07 0.11 -0.21 -0.09 1.00

GROWTH (10) 0.00 0.13 -0.05 -0.16 0.15 -0.08 -0.14 -0.01 0.27 1.00

INV(11) 0.10 0.08 -0.02 -0.01 0.05 0.07 0.02 0.02 0.01 0.00 1.00

SEGMENT(12) 0.00 -0.12 -0.09 0.11 -0.05 -0.10 0.06 -0.02 -0.09 -0.30 -0.06 1.00

FINANCE(13) 0.05 0.14 -0.04 0.00 -0.05 -0.02 -0.17 -0.12 0.20 0.02 -0.06 -0.03 1.00

ROA (14) -0.02 0.02 0.06 -0.02 0.08 0.04 0.00 -0.01 0.04 0.17 0.08 -0.13 -0.01 1.00

AC (15) 0.22 0.04 -0.05 0.08 -0.16 0.13 0.09 0.01 0.17 -0.28 0.20 0.05 -0.02 0.24 1.00

IC (16) 0.13 0.16 -0.06 -0.06 -0.12 0.08 -0.09 -0.06 0.23 0.06 -0.01 -0.07 0.03 0.04 0.08 1.00

|DAC| (17) 0.05 -0.01 0.06 0.03 -0.01 0.06 -0.05 -0.05 0.11 0.07 -0.05 -0.04 0.02 0.05 -0.01 0.20 1.00

Note: Italicized and bold-faced correlations are significant at p<0.01. Variable definitions are in Appendix III. The correlation is based on a full sample of 1,429 firm-year observations.

Page 43: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

42

TABLE 3: Political connections, RPTs, and auditor choice

)1.......(..............................||13121110987

6543210

YearIndustryDACICACROAFINANCESEGMENTINV

GROWTHLEVSIZEFOWNOWNCONPCONAUDITOR

PANEL A: Political connections, RPTs, and auditor choice Only PCON=1 Only PCON=1

RPT_RATIO RPT_RATIO RPT_RATIO

Expected

Sign

(1) (2) (3) (4) (5)

Baseline Baseline PCON GCON=1 MCON=1

Variables Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

PCON - -0.363** - 0.007 - -

[-2.22] [0.04]

GCON - - -0.877*** - - -

[-4.36]

MCON - - -0.193 - - -

[-0.88]

SCON + - 2.358*** - - -

[4.05]

RPT_RATIO - - - -0.271** -2.417*** -1.16

[-2.21] [-4.96] [-0.78]

PCON*RPT_RATIO - - - -1.163*** - -

[-4.12]

OWNCON + 2.150*** 2.229*** 2.620*** 4.340*** 5.609***

[6.85] [7.03] [8.12] [5.30] [3.63]

FOWN + 1.866*** 1.857*** 1.936*** 4.229*** 5.163***

[7.59] [7.36] [7.32] [6.01] [5.11]

SIZE + 0.805*** 0.840*** 0.906*** 0.731*** 1.427***

[13.02] [12.81] [12.94] [3.37] [3.76] LEV - -1.283** -1.429** -1.606** -3.125 -6.753***

[-2.03] [-2.22] [-2.36] [-1.63] [-2.63]

GROWTH + 0.047** 0.052*** 0.047** 0.049* 0.116

[2.47] [2.74] [2.46] [1.81] [0.93]

INV + 1.225*** 1.441*** 1.062** 3.881** -0.632

[2.60] [3.01] [2.09] [2.24] [-0.33]

SEGMENT + 0.049 0.072 0.015 -0.022 0.005

[0.36] [0.51] [0.11] [-0.05] [0.03]

FINANCE ? -0.341 -0.312 -0.387 0.066 0.289

[-1.49] [-1.33] [-1.63] [0.13] [0.37]

ROA + 1.984** 1.803** 1.512* 11.362*** -1.338 [2.44] [2.15] [1.76] [3.36] [-0.40]

AC + 0.726*** 0.847*** 0.633*** -0.577 4.304***

[3.69] [4.18] [3.28] [-0.78] [4.38]

IC + 0.066 0.262 0.184 -0.101 -0.524

[0.26] [1.11] [0.72] [-0.18] [-0.45]

|DAC| ? 0.123 0.090 -0.035 -1.352 7.542*

[0.15] [0.11] [-0.04] [-0.48] [1.79]

Constant -16.710*** -17.614*** -18.435*** -14.519*** -31.578***

[-12.41] [-12.62] [-12.77] [-3.36] [-4.10]

Industry YES YES YES YES YES

Year YES YES YES YES YES

Pseudo R2 0.29 0.31 0.32 0.46 0.48

Observations 1,429 1,429 1,429 342 155

***, **, and * represent statistical significance at the 1%, 5%, and 10% levels respectively (two-tailed test). Variable

definitions are in Appendix III.

Page 44: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

43

PANEL B: Categories of RPTs, political connections, and auditor choice

Operating RPTs RPT loan and guarantees Operating RPTs and RPT loans (1) (2) (3) (4) (5) (6) (7) (8) (9)

Predicted

sign

Full sample Only

PCON=1 sample

Only

PCON=1 sample

Full sample Only

PCON=1 sample

Only

PCON=1 sample

Full sample Only PCON=1

sample Only PCON=1

sample

GCON=1 MCON=1 GCON=1 MCON=1 GCON=1 MCON=1

Variables Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

PCON - -0.22 - - -0.338* - - -0.125 - -

[-1.19] [-1.92] [-0.66]

OPRPT_RATIO + 0.47*** -0.185 0.348 - - - 0.430*** -0.333 0.380

[3.40] [-0.75] [0.55] [3.08] [-1.26] [0.59]

LOAN_RATIO - - - -1.180** -3.160*** 0.882 -1.058** -3.719*** 0.880

[-2.42] [-3.49] [0.84] [-2.19] [-4.07] [0.83]

PCON*OPRPT_RATIO + -0.517 - - - - - -0.56 - -

[-1.16] [-1.29]

PCON*LOAN_RATIO - - - - -1.373* - - -1.550** - -

[-1.73]

[-1.96]

Other control variables YES YES YES YES YES YES YES YES YES

Industry YES YES YES YES YES YES YES YES YES

Year YES YES YES YES YES YES YES YES YES

Pseudo R2 0.29 0.37 0.47 0.31 0.42 0.47 0.31 0.42 0.47

Observations 1,429 342 155 1,429 342 155 1,429 342 155

Page 45: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

44

TABLE 4: Endogeneity tests

Heckman (1979) self-selection tests

..(5)……………………………… +YearFE + FEIndustry ROASEGMENTGROWTH +LEV +SIZE +FOWN +OWNCON +HQ%PCON_IND =PCON

98

7654321

)1.......(..............................|| 141312111098

76543210

YearIndustryIMRDACICACROAFINANCESEGMENT

INVGROWTHLEVSIZEFOWNOWNCONPCONAUDITOR

(1) (2) (3) (4) (5)

1st stage

probit

model

DV=PCON

Expected

sign for the

2nd stage

variables

PCON=1 &

GCON=1

PCON=1 &

MCON=1

Variables Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

PCON - - -0.029 -0.199 - -

[-0.15] [-1.07]

RPT_RATIO - - -0.310** - - -

[-2.38]

OPRPT_RATIO - + - 0.390*** -0.338 0.493

[2.71] [-1.29] [0.81]

LOAN_RATIO - - - -1.120*** -3.266*** 1.337

[-2.75] [-3.80] [1.29]

PCON*RPT_RATIO - - -1.146*** - - -

[-3.97]

PCON*OPRPT_RATIO - + - -0.535 - -

[-1.20]

PCON*LOAN_RATIO - - - -1.510* - -

[-1.89]

HQ 0.219** NA - - - -

[2.16]

%PCON_IND 1.311* NA - - - -

[1.76]

OWNCON -0.092 + 2.715*** 2.436*** 3.910*** 4.808***

[-0.43] [8.24] [7.21] [4.35] [2.98]

FOWN -0.328** + 2.137*** 2.145*** 3.944*** 4.905***

[-2.37] [5.04] [5.37] [5.69] [3.95]

SIZE 0.478*** + 0.866** 0.576 0.664*** 0.792

[14.49] [2.04] [1.41] [2.59] [1.34]

LEV -0.310 -1.654** -0.954 -1.647 -7.587**

[-1.00] [-2.25] [-1.51] [-0.86] [-2.48]

GROWTH 0.032*** + 0.042 0.031 0.037 0.066

[3.00] [1.37] [1.04] [1.11] [0.46]

INV - + 1.076** 1.075** 3.384* -2.623

[2.09] [2.03] [1.87] [-1.41]

SEGMENT 0.172*** + -0.069 -0.181 -0.269 -0.115

[2.22] [-0.34] [-0.89] [-0.72] [-0.21]

FINANCE - ? -0.382 -0.304 0.277 0.137

[-1.60] [-1.38] [0.57] [0.17]

ROA -0.515 + 1.691 1.683* 11.204*** -0.344

[-1.35] [1.62] [1.92] [3.44] [-0.09]

AC - + 0.665*** 0.694*** -0.139 4.633***

[3.39] [3.33] [-0.20] [4.54]

IC - + 0.188 0.023 -0.728 -1.407

[0.73] [0.08] [-1.24] [-1.22]

Page 46: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

45

|DAC| - ? -0.043 0.037 -0.723 8.041*

[-0.05] [0.05] [-0.26] [1.68]

IMR - ? -0.344 -1.514 0.886 -4.662

[-0.16] [-0.72] [1.02] [-1.48]

Constant -9.742*** -17.425* -17.63*** -13.587** -15.837

[-11.90] [-1.75] [-8.75] [-2.34] [-1.15]

Industry YES YES YES YES YES

Year YES YES YES YES YES

Pseudo R2 0.26 0.31 0.30 0.44 0.44

Observations 1,429 1,429 1,429 342 155

Page 47: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

46

Table 5: Propensity-matched technique

PANEL A: Propensity-matched variables

NN method Kernel method

Variable Treated Control t-stat p-value Treated Control t-stat p-value

OWNCON 0.51 0.52 -0.75 0.451 0.51 0.52 -0.61 0.54

FOWN 0.23 0.29 -3.15 0.002 0.23 0.28 -2.49 0.013

SIZE 20.04 19.97 0.83 0.407 19.91 19.90 0.15 0.883

LEV 0.16 0.17 -1.43 0.152 0.15 0.16 -1.17 0.243

INV 0.14 0.14 -0.16 0.874 0.15 0.14 0.07 0.941

SEGMENT 0.99 0.93 1.91 0.057 0.98 0.93 1.66 0.097

FINANCE 0.14 0.13 0.83 0.408 0.14 0.13 0.46 0.646

ROA 0.08 0.09 -1.62 0.115 0.08 0.09 -1.62 0.115

AC 0.89 0.89 0.08 0.939 0.89 0.89 -0.05 0.963

|DAC| 0.09 0.10 -0.74 0.458 0.09 0.10 -0.34 0.737

Page 48: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

47

PANEL B: Regression results

(1) (2) (3) (4) (5) (6) (7) (8)

NN NN NN NN Kernel Kernel Kernel Kernel

Variables Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

Coefficient

[z-stat]

PCON - -0.613** - -0.401* -0.569*** -0.765*** -0.360* -0.679***

[-3.54] [-1.91] [-2.79] [-4.74] [-1.77] [-3.56]

GCON - - -0.724*** - - - -1.203*** - -

[-3.62] [-6.22]

MCON - - -0.967*** - - - -0.647*** - -

[-4.20] [-2.83]

SCON + - 2.696*** - - - 2.247*** - -

[3.85] [3.62]

RPT_RATIO - - - -0.664*** - - - -0.554** -

[-2.81] [-2.48]

OPRPT_RATIO + - - - -0.237 - - - -0.225

[-1.00] [-1.06]

LOAN_RATIO - - - - -1.667*** - - - -1.276**

[-2.92] [-2.38]

PCON*RPT_RATIO - - - -0.723** - - - -1.153*** -

[-2.11] [-3.36]

PCON*OPRPT_RATIO + - - - 0.016 - - - -0.012

[0.06] [-0.04]

PCON*LOAN_RATIO - - - - -2.047* - - - -1.590*

[-1.80] [-1.76]

OWNCON + 2.980*** 3.332*** 3.753*** 3.429*** 2.192*** 2.331*** 2.872*** 2.508***

[6.96] [7.51] [8.20] [7.60] [5.56] [5.73] [6.83] [6.06]

FOWN + 4.040*** 4.356*** 4.230*** 4.181*** 2.141*** 2.124*** 2.221*** 2.171***

[12.35] [12.69] [12.50] [12.38] [6.69] [6.44] [6.56] [6.58]

SIZE + 0.855*** 0.840*** 0.973*** 0.882*** 0.748*** 0.783*** 0.906*** 0.793***

[9.97] [9.50] [10.10] [9.58] [8.90] [8.81] [9.47] [8.84]

LEV - -1.317** -1.106 -1.355* -0.900 -0.576 -0.638 -0.824 -0.497

[-1.97] [-1.61] [-1.95] [-1.30] [-0.89] [-0.96] [-1.22] [-0.74]

GROWTH + -0.029* -0.032* -0.043** -0.040** -0.001 0.003 -0.012 -0.006

[-1.72] [-1.91] [-2.43] [-2.30] [-0.04] [0.19] [-0.66] [-0.35]

Page 49: Political connections, related party transactions, … connections, related party transactions, ... Political connections, related party transactions, and auditor ... than 0.5% of

48

INV + 0.326 0.889 0.771 0.910 1.044* 1.452** 0.994 1.170*

[0.56] [1.44] [1.25] [1.44] [1.69] [2.29] [1.51] [1.78]

SEGMENT + -0.386** -0.172 -0.365** -0.413** 0.145 0.206 0.158 0.072

[-2.33] [-0.99] [-2.13] [-2.42] [0.89] [1.22] [0.92] [0.43]

FINANCE ? -0.261 -0.193 -0.372 -0.335 -0.326 -0.295 -0.444* -0.321

[-1.07] [-0.77] [-1.43] [-1.29] [-1.34] [-1.17] [-1.72] [-1.26]

ROA + 5.829*** 5.699*** 5.266*** 5.446*** 4.808*** 4.667*** 4.602*** 4.312***

[6.34] [6.24] [5.46] [5.67] [5.22] [4.93] [4.81] [4.55]

AC + 1.062*** 1.161*** 0.877*** 0.970*** 0.927*** 1.129*** 0.765*** 0.875***

[3.87] [4.06] [3.06] [3.44] [3.27] [3.80] [2.61] [3.01]

IC + 0.357 0.745** 0.514 0.379 -0.142 0.067 0.016 -0.146

[1.09] [2.12] [1.45] [1.09] [-0.44] [0.20] [0.05] [-0.43]

|DAC| ? 0.717 0.049 0.711 0.689 -1.201 -1.391 -1.407 -1.367

[0.78] [0.05] [0.75] [0.71] [-1.26] [-1.39] [-1.42] [-1.39]

Constant -17.826*** -18.295*** -20.159*** -18.379*** -15.018*** -16.078*** -17.872*** -15.473***

[-9.26] [-9.15] [-9.54] [-8.99] [-8.36] [-8.44] [-8.87] [-8.11]

Industry YES YES YES YES YES YES YES YES

Year YES YES YES YES YES YES YES YES

Pseudo R2 0.43 0.45 0.45 0.44 0.29 0.33 0.34 0.31

Observations 1,150 1,150 1,144 1,144 1,398 1,398 1,391 1,391


Recommended