JOURNAL OF MANAGEMENT journal/UoSE V_10_BOOK-final.pdfISSN 1391-8230 Volume X No. 1 April 2014...

ISSN 1391-8230

April 2014Volume X No. 1

JOURNAL OF MANAGEMENT

Published by the Faculty of Management & Commerce

South Eastern University of Sri Lanka

Oluvil # 32360

Sri Lanka

The Journal of the Faculty of Management and Commerce


Journal of Management

EDITORIAL BOARD MEMBERS – JOURNAL OF MANAGEMENT

Editor in Chief: Dr. MIM. Hilal, Senior Lecturer, SEUSL

Associate Editors: Mrs. Sulaiha Beevi Athambawa, Senior Lecturer, SEUSL

Mr. KM. Mubarak, Senior Lecturer, SEUSL

Managing Editor: Ms. MMM. Mashroofa, Senior Assistant Librarian, SEUSL

Editorial Advisory Dr. SMM. Ismail, Snr. Lecturer, SEUSL

Board: Prof. Mahalliyawarachchi, Sabragamuwa University of Sri Lanka.

Prof. Sampath Amaratunge, University of Sri Jayawardenepura.

Dr. (Mrs.) FHA. Rauf, Snr. Lecturer, SEUSL

Dr. A. Jahufer, Snr. Lecturer, SEUSL

Dr. S. Gunapalan, Snr. Lecturer, SEUSL

Mrs. Sareena UMA Gaffoor, Snr. Lecturer, SEUSL

Ms. MAC. Sulaiha Beevi Athambawa, Snr. Lecturer, SEUSL

Mrs. A. Inun Jariya, Snr. Lecturer, SEUSL

Mr. ALMA. Shameem, Snr. Lecturer, SEUSL

Publication: The Journal of Management is published two times in a year in April

and October by the Faculty of Management & Commerce (FMC),

South Eastern University of Sri Lanka (SEUSL).

Subscription: Foreign: US $ 12 (SAARC Countries)

US $ 25 (Other Countries)

Local: Rs.500.00 inclusive of postage

Subscription will be accepted as standing orders. Single issues also will be provided on request

for Rs. 700.00 per copy. All orders must be accompanied with pre-payment in favour of the

“South Eastern University of Sri Lanka”. Sales request should be addressed to Assistant

Registrar, FMC, and SEUSL.

JOURNAL OF MANAGEMENT – Volume X No.1 - April 2014

Editorial Policies: Manuscript of research papers, reviews, and short communications in the field

of management sciences should be accepted for publication on the understanding that they have

not been published or submitted simultaneously for publication elsewhere. Manuscripts should

be clearly typed in single-spaced; length may be usually limited to 20 pages in A4 size paper.

Every manuscript must include a 100 – 150 words long abstract. By submitting an article, the

author transfers the copyright to Editorial Board of the Journal.

All papers submitted for publication are examined and evaluated and by referees, to assess the

suitability for publication. The Editorial Board reserves the right to edit the materials to confirm

to the style and format of the journal.

The Editorial Board is not responsible for any shortcomings in the form of inaccurate or

misleading data, option or statement appearing in the Journal leading to confusion and criticism.

So, the author is entirely responsible for the contents of the article subject to the above

consideration.

Correspondence: All correspondence regarding the Journal articles should be sent to:

Editor in Chief

Editorial Office

“Journal of Management”

Faculty of Management & Commerce


Oluvil #32360

Sri Lanka.

TP / Fax : 0094-67-2255069

E-mail : [email protected]

Copyright © Faculty of Management and Commerce, South Eastern University of Sri Lanka.

The Copyright covers the exclusive right to reproduce and distribute the article, including

reprints, photographic reproductions, microforms or any other reproductions of similar nature

and translations. No part of the article may be reproduced by the reader without permission in

writing from the copyright holder.

JOURNAL OF MANAGEMENT

Volume X No. 1 April 2014

Contents Page No

Contribution of Macroeconomic Factors on the Stock Market Performance in Sri LankaAboobacker Jahufer and Shehu Shariff Mohammed Irfan

Career Development - A knowledge based Economy for employability: from the Perspectives South Eastern University of Sri LankaMohamed Abdul Cader Salfiya Ummah and M.N Mohamed Bilal

Employment Generation for Women through Rural Tourism in KeralaMoli P. Koshy, Sabira N and Baby Salini P. V

Inter-Sectoral Dynamic Growth Linkages: Empirical Evidence form Sri LankaSelliah Sivarajasingham

Analyzing the Impact of Intrinsic Job Satisfaction of Government School TeachersSpecial Reference to Kalmunai Educational Zone, Sri LnakaAboobacker Jahufer and M.H.M. Sarjoon

Review of statistical modeling in technical analysis of Financial marketsW.G.S. Konarasinghe

The Effect of Unemployment on Socio-Economic Status of the People in Jaffna District, Sri LankaPaulina Mary Godwin Phillip and Thayaparan Aruppillai

Impact of Capital Structure on Profitability: A study of Listed ManufacturingCompanies in the Colombo Stock (SEC) Exchange in Sri Lanka.Sithy Safeena M.G. Hassan

1

10

15

25

34

42

51

60

SaNa

Typewriter

and Thamarasi Kularatne

SaNa

Typewriter

SaNa

Typewriter

SaNa

Typewriter

SaNa

Typewriter

SaNa

Typewriter

SaNa

Typewriter

ABSTRACT

The main aim of this research study is to analyzethe connections that exist amongst the all shareprice index of Colombo Stock Exchange (CSE)and four major macro economic factors:Inflation, Exchange Rate, Money Market Rateand Money Supply of Sri Lanka. To study thisresearch the data were collected monthly basisfrom January 2001 to December 2011 timeperiods. Co-integration analysis for macroeconomic factors and all share price index ofstock market were carried out to test for theexistence and Vector Error Correction Model(VECM), indeed extent of the co-movement thatis evident of co-integration can be viewed as thestatistical expression of the nature of equilibriumrelationships, with co-integrated variablessharing common stochastic trends. The Johansenco-integration test and VECM and its estimationprocedures are discussed in this paper. Long andshort run relationships exist among the Stockprice index and macroeconomic variables. Theresults of the study will aid us to gain insight intohow all share price index of CSE of Sri Lankaco-integration contributes to portfoliodiversification strategy. The results haveimplication for investors, both domestic andinternational.

Key Words: Co-integration Test, ADF-UnitRoot Test, Colombo Stock Exchange, MacroEconomic Variables, Vector Error CorrectionModel.

Introduction

CSE operates the only share market in Sri Lankaand is responsible for providing a transparent andregulated environment where companies andinvestors can come together. The CSE is acompany that is limited by guarantee establishedunder the Laws of Sri Lanka. At present theCSE functions as a market operator and throughits fully owned subsidiary, Central DepositorySystems (Pvt.) Limited (CDS), acts as a clearingand settlement system facilitator. The CSE hastwo main price indices, the All Share Price Index(ASPI) and the S&P Sri Lanka 20 Index (S&PSL 20). These index values are calculated on anon-going basis during the trading session, withthe closing values published at the end of eachsession.

An efficient capital market is one in whichsecurity prices adjust rapidly to the arrival ofnew information and therefore, the current pricesof securities reflect all information about thesecurity. What this means, in simple terms, isthat no investor should be able to employ readilyavailable information in order to predict stockprice movements quickly enough so as to makea profit through trading shares. Moreover,economic theory suggests that stock pricesshould reflect expectations about future corporateperformance and corporate profits generallyreflect the level of economic activities. If stockprices accurately reflect the underlyingfundamentals, then the stock prices should be

1

CONTRIBUTION OF MACROECONOMIC FACTORS

ON THE STOCK MARKET PERFORMANCE

IN SRI LANKA

Aboobacker Jahufer

Department of Mathematical Sciences, Faculty of Applied Sciences,South Eastern University of Sri Lanka

[email protected]

Shehu Shariff Mohammed IrfanDepartment of Mathematical Sciences, South Eastern University of Sri Lanka

[email protected]

employed as leading indicators of futureeconomic activities and not the other wayaround. Therefore, the causal relations anddynamic interactions among macroeconomicvariables and stock prices are important in theformulation of the nation’s macroeconomicpolicy.

To achieve the goal of the research study thisresearch paper is organized into five sections.Literature reviews are given in section two,section three described the methodology of theresearch study, in section four results anddiscussions are given and the last sectionconclusions and recommendations are given.

Literature Review

History has shown that the price of all shareprice index and other financial assets are animportant aspect of the dynamics of economicactivity, performing a vital role in nationaleconomies of Sri Lanka. Stock prices can be anindicator of social mood and are used as aleading indicator of the real economic activity.Rising share prices, for instance, tend to beassociated with increased business investmentand vice versa. Therefore, economic policymakers keep an eye on the control and behaviorof the stock market, as its smooth and risk freeoperation is essential for economic and financialstability. This study suggests that the movementof stock market indices is highly sensitive by thechanges in the macroeconomic variable. Themacroeconomic factors are the majordeterminants of the growth of an economy in SriLanka.

Addy et al., (2014), identify the relationshipbetween macroeconomic variable and Ghanastock exchange which was revealed that there isa long run relationship between some of themacro-economic variables and the stock market.Money supply is statistically significant at 1%level in explaining the variations in theperformance of Ghana stock exchange. Withthese results it is important to highlight that thereis the need to implement prudent macroeconomic

policies in order for the country to derivemaximum benefits from stock markets. Chandniet al., (2012) investigated that the relationshipbetween Indian stock exchange andmacroeconomic factors which show that there isa positive relation with call rate (interest rate)and negative relation with exchange rate.Anokye and George (2008) studied the role ofmacroeconomic variables in stock marketmovement in Ghana which was revealed thatthere is long run relationship between thevariables using Johansen's multivariate co-integration tests. And they found inflation topositively relate to Databank Stock Index.

Caroline et al., (2011) studied the relationshipbetween stock market, expected inflation rate,unexpected inflation rate, exchange rate, interestrate and GDP in the case of Malaysia, US andChina. They found that there is a long runequilibrium relationship between the variables.The results of VEC are no short run relationshipbetween the stock market, expected inflation,exchange rate, unexpected inflation, interest rateand GDP for Malaysia and US. However,China’s VEC result show there is a short runrelationship between expected inflation rateswith China’s stock market. Omran and Pointon(2001) analyzed short-run and long-runrelationships between the inflation rate and theperformance of the Egyptian stock market. Theresults revealed an expected behavior for thestock market response to the decrease in theinflation rate. From this analysis they concludedthat the inflation rate, clearly, has had an impactupon stock market performance in terms ofmarket activity and market liquidity. In fact, thisrelationship was negative and in the long andshort-run for all market activity and marketliquidity variables except for the number oftraded companies, in which case this relationshipwas in the long-run only.

Wickremasinghe (2006) examined therelationship among stock price and sixmacroeconomic variables in Sri Lanka. Theresults of the Johansen co-integration testindicate that there is one co-integration

2


relationship among the stock price andmacroeconomic variables. Therefore, researcherproceeded to estimate the error correction modelto examine short-run and long-run relationship.This reveals that there are five long-runrelationships and two short-run relationshipsamong the variables. Money supply and USdollar exchange rate highly influence on stockmarket operation. Tahir and Wong (2004) studiedrelationship between four stock indices andexchange rate in Karachi Stock Exchange (KSE),the results obtained by using Johanson’s co-integration technique that there is no co-integrating relationship among the variables.Therefore, a long-run relationship between stockindices and exchange rate does not exist in theiranalysis, that is stock indices and exchange ratedo not move together in the long run associationof KSE. The exchange rate which indicates themovement of currency affects stock prices in away similar to the inflation variable.Depreciation of the local currency makes importexpensive compared to export, leading toincreased production cost of import companies.Mukherjee and Naka (1995) also found thatexchange rate positively relates to stock prices inJapan and Indonesia.

According to Nissim and Penman (2003),various studies have acknowledged that stockreturns and interest rates are negatively related.They examined the relationship between achange in the interest rate, earnings and stockreturn using data from the US and found thatunexpected changes in interest rates arepositively related to unexpected earnings in theyear of the interest rate change and in thefollowing year. This relationship is due to apositive association between interest rates andoperating income.

Keray (2009) investigated that theinterrelationship between stock prices andmonetary indicators for Jamaica. The Johansenco-integration test was used to determine theexistence of a long term relationships betweenstock prices and monetary variables such asmoney supply, interest rate, inflation rate and the

exchange rate. The variables were found to beco-integrated with significant relationshipsresearcher suggests that the JSE main index ispositively influenced by the inflation rate andmoney supply and negatively by the exchangerate, interest rate. The relationship betweenmoney supply and the stock market has beeninvestigated empirically.

Methodology

In the process of operationalisation the conceptwhich could be visualized by the researchers isas follows.

Objective of the Research Study

The objective of this study is to analyze theconnections that exist amongst the all share priceindex of CSE and four major macro economicfactors: Inflation, Exchange Rate, Money MarketRate and Money Supply of Sri Lanka.

Model Specification

The model for this study can be expressed inequation (1) and can represent this function in amathematical linear model as shown in equation(2).

ASPI=fEXH,INF,MMR,M (1)

ASPIt=β0+β1EXHt+β2INFt+β3MMRt+β4Mt+εt (2)

For the purpose of estimation, all the variables inthe above equation (2) are expressed in a log-linear form.

LASPIt=β0+β1LEXHt+β2LINFt+β3LMMREt+β4LMt+εt (3)

where; dependent variable is all share price index(ASPI) and the independent variables are:exchange rate (EXR), Inflation (INF), Moneymarket rate (MMR) and money supply (M).

3

Contribution of Macroeconomic Factors

on the Stock Market Performance in Sri Lanka

ASPI

EXH

INF

INTR

M

𝛽0 is a constant, 𝛽1,𝛽2,𝛽3 and 𝛽4 are thesensitivity of each of the macroeconomicvariables to stock prices and ε𝑡 is a stationaryerror correction term.

The test for stationary of the individual series inthe above econometric model (3) was undertakenby applying the Augmented Dickey Fuller (ADF)unit root test procedure using E-views version5.0.

Testing for Unit Roots

To have a meaningful understanding of therelationship between two or more economicvariables using regression technique, the timeseries data should satisfy some stationaryproperties. Hence any time series analysis shouldstart by checking the order of integration of eachvariable. The Augmented Dickey Fuller (ADF)(see Gujrati, 2007) test used to examines thepresence of unit roots in the data series. Thegeneral form of ADF test can be written asfollows

∆𝑌𝑡=𝑎+𝑏𝑡+𝑝𝑌𝑡−1+𝑖=1𝑘∇𝑌𝑡−1+𝑈𝑡 (4)

where, Yt is the individual time series, ΔYt, isthe first difference of the series Yt. Here, ΔYt =Yt - Yt-1, k is the lag order, t is the linear timetrend. Ut is serially uncorrelated random termwith zero means and constant variance, and 𝑎 -is constant.

Testing for Co-integration

Co-integration technique are used to examine thelong run relationship between economicvariables if they are integrated of order one (1).A long run relationship means that the variablesmove together over time, so that short rundisturbances from the long run trend will becorrected (Manning and Andrianacos, 1993).This study adopts Johanson and Juseliues (1990)method of co-integration. This method requiresthat variables entering the co-integrationrelationship to be integrated of the same orderand yields two likelihood statistics known as

trace and maximum eigen value statistics whichare given by:

𝜆𝑡𝑟𝑎𝑐𝑒𝑟=−𝑇𝑖=𝑟+1𝑛ln�(1−𝜆𝑖) (5)

𝜆𝑚𝑎𝑥𝑟,𝑟+1=−𝑇𝑙𝑛(1−𝜆𝑟+1) (6)

where, T is the number of observation; i is the itheigen value 𝜆𝑖 and r = 0, 1, 2… n-1. The tracestatistic tests the null hypothesis of at most r co-integration relations against the alternative ofmore than r co-integrating relations. Further theoptimum lag length used in the estimation isobtained on the basis of the Akaike InformationCriteria (AIC).

Data Collection

The required data for this study is collectedmainly from secondary sources with monthlybasis from January 2001 to December 2011. Thedata of all share price index was obtained fromColombo Stock Exchange web site. All the otherdata except money supply was collected fromCentral Bank web site and Central Bank annualreport. Money supply was obtained frominternational financial statistics database. The USdollar exchange rate expressed as the amount ofSri Lanka rupees per unit of US dollar (USD),the inflation rate was calculated by the changesin Colombo Consumer Price Index.

Results and Discussions

Descriptive Statistics

Table-1 presents a summary of descriptivestatistics of the variables. Sample mean,maximum, minimum, standard deviation,skewness, kurtosis, Jacque-Bera statistic and p-value have been reported. The high standarddeviation of LASPI with respect to the mean isan indication of high volatility in the stockmarket. From the p-values, the null hypothesisthat LEXH, LINF, LMMR and LM are normallydistributed at 5% level of significance cannot berejected.

4


Unit Root Test

Table 2: shows Augmented Dickey-Fuller (ADF)unit root test results to determine the order ofintegration and stationary of the variables. This

result indicates that all the data are non-stationary at 5% levels but first differences arestationary at 1% significant level. Consistentwith Figure 1: it can be concluded that all thevariables are I(1).

5



Mean 7.530462 4.644133 2.214171 2.381691 12.30404Maximum 8.961623 4.788099 3.343215 3.244154 12.99568Minimum 6.000424 4.455040 -0.287682 2.012233 11.59477Std. Dev. 0.780996 0.076484 0.629314 0.303068 0.409839Skewness -0.143921 -0.373144 -1.425305 0.880138 -0.128059Kurtosis 2.433424 2.117871 6.607508 3.070352 1.988640Jarque-Bera 2.221236 7.343046 116.2705 17.06935 5.986444Probability 0.329355 0.025438 0.000000 0.000197 0.050126Sum 994.0210 613.0255 292.2706 314.3832 1624.133Sum Sq. Dev. 79.90409 0.766317 51.88076 12.03236 22.00384Observations 132 132 132 132 132

LASPI LEXH LINF LMMR LM

Table 1: Summary Statistics of variable

LASPILEXH -2.3748 0.3909 -11.5870 0.0000LINF -3.2698 0.0760 -5.7269 0.0000

LMMR -2.3063 0.4272 -6.4652 0.0000LM -2.0141 0.5879 -12.7452 0.0000

Variablest-value p-value t-value p-value

Level First Differences

Table 2: ADF - Unit Root Test

Co-integration Test and Vector Error

Correction Model

The next step involves estimating the model anddetermining the rank, r to find the number of co-integrating relations in our model. The optimallag length was determined by both Schwarz(SIC) and Akaike Information Criterion (AIC)using 12 maximum lags in the general VARmodel. The aim is to choose the number ofparameters, which minimizes the value of theinformation criteria. The SIC has the tendency tounderestimate the lag order, while adding morelags increases the penalty for the loss of degreesof freedom. To make sure that there is noremaining autocorrelation in the VAR model,AIC is selected as the leading indicator. The

6


model lag length report indicates appropriate laglength of 3 for AIC.

We proceed to test for the presence of long-runrelationship among the variables by usingJohansen's Maximum Likelihood approach. Anintercept and no trend are specified for the co-integration test. Table-3 shows the trace statisticsuggests one co-integrating vectors and themaximum eigenvalue statistic indicates one co-integrating vector at the 5% significance level.This indicates co-movement between stock

market index and macroeconomic variables in along-run equilibrium path. The co-integrationgraph (see Figure 2) confirms that there are morethan “one” mean reversion effect in the co-integration vector over the period and signifies agood error correction behavior in the co-integration system. Consequently, the estimatedlong-run relationship via co-integration analysisand the error correction coefficients areappropriate. The long-run co-integrating relationbetween the macroeconomic factors and stockprices normalized on LASPI is given by:

1.000000 0.523455 -0.383817 0.690479 -2.041790 14.36956

(0.99978) (0.06721) (0.13817) (0.18714)

LEXHLASPI LINF LMMR LM C

The model can be re-parameterized as

𝐿𝐴𝑆𝑃𝐼𝑡=−0.5234𝐿𝐸𝑋𝐻𝑡+0.3838𝐿𝐼𝑁𝐹𝑡−0.6904𝐿𝑀𝑀𝑅𝑡+2.0417𝐿𝑀𝑡−14.3695

Table 3: Johansen Cointegration Test

Unrestricted Co-integration Rank Test (Trace)

None * 0.233766 78.74057 69.81889 0.0082At most 1 0.165112 44.65830 47.85613 0.0968At most 2 0.112829 21.55967 29.79707 0.3236At most 3 0.044230 6.235852 15.49471 0.6676At most 4 0.003473 0.445324 3.841466 0.5046

Hypothesized

No. of CE(s)

Eigenvalue Trace

Statistic

0.05

Critical Value

Prob.**

Trace test indicates 1 co-integrating eqn(s) at the 0.05 level* denotes rejection of the hypothesis at the 0.05 level**MacKinnon-Haug-Michelis (1999) p-values

Unrestricted Co-integration Rank Test (Maximum Eigenvalue)

None * 0.233766 34.08227 33.87687 0.0473At most 1 0.165112 23.09863 27.58434 0.1693At most 2 0.112829 15.32382 21.13162 0.2668At most 3 0.044230 5.790528 14.26460 0.6403At most 4 0.003473 0.445324 3.841466 0.5046

Hypothesized

No. of CE(s)

Eigenvalue Max-Eigen

Statistic

0.05

Critical Value

Prob.**

Max-eigenvalue test indicates 1 co-integrating eqn(s) at the 0.05 level* denotes rejection of the hypothesis at the 0.05 level**MacKinnon-Haug-Michelis (1999) p-values

7



Figure 2: A Plot of Co-integration Vectornormalized on LASPI

The coefficients of LINF and LM are positivesign with LASPI. And also LEXH and LMMRare negative sign with LASPI. Our expectedrelationship between LASPI and LINF should bepositive it is exist here. Contrary to ourexpectation LEXH has negative relationship withLASPI. The negative relationship betweenLASPI and LMMR is expected, from theinvestor background when increasing moneymarket rate is leads to invest any other financialinstitution or banks to gain high return theircapital with a risk free environment.

Table 4: shows that the error correction termestimated t value of 0.0590 is less than thecritical value of t and p-value is 0.953 which isgreater than 0.05, R-squared and Adjusted R-squared are 0.222 and 0.1102 respectively. Asmuch it can be concluded that the null hypothesisof 2 short run relationship for the model All

share price index = f (Inflation, Exchange rate,Money market rate, Money supply), which arebetween the money market rate with the stockmarket index and money supply with stockmarket index. All variables in lag 1, lag 2 and lag3 for p-values, ASPI (0.4485, 0.7981, 0.5604),exchange rate (0.2058, 0.7903, 0.4248), inflation

Table 4: VECM estimation for LASPI

ECT 0.002465 0.041774 0.059012 0.953D(LASPI(-1)) 0.078839 0.103641 0.76069 0.4485D(LASPI(-2)) -0.02587 0.100909 -0.25637 0.7981D(LASPI(-3)) -0.05592 0.095743 -0.58403 0.5604D(LEXH(-1)) 0.789899 0.620647 1.272702 0.2058D(LEXH(-2)) -0.17261 0.647685 -0.2665 0.7903D(LEXH(-3)) 0.516378 0.644541 0.801157 0.4248D(LINF(-1)) 0.031031 0.033299 0.931894 0.3534D(LINF(-2)) -0.03064 0.031231 -0.98119 0.3286D(LINF(-3)) -0.03571 0.031995 -1.11621 0.2667D(LMMR(-1)) -0.17705 0.065982 -2.68321 0.0084D(LMMR(-2)) -0.12195 0.069805 -1.74704 0.0834D(LMMR(-3)) -0.11536 0.065611 -1.75827 0.0815D(LM(-1)) 1.038802 0.315169 3.296011 0.0013D(LM(-2)) 0.118518 0.321957 0.368116 0.7135D(LM(-3)) -0.18578 0.323674 -0.57397 0.5671C 0.005264 0.010055 0.523544 0.6016

R-squared 0.222373 Mean dependent var 0.020939Adjusted R-squared 0.110283 S.D. dependent var 0.076837S.E. of regression 0.072476 Akaike info criterion -2.28799Sum squared resid 0.583063 Schwarz criterion -1.9092Log likelihood 163.4313 Durbin-Watson stat 1.98353

Variables Coefficient Std. Error t-Statistic Prob.

(0.3534, 0.3286, 0.2667), money market rate(0.0084, 0.0834, 0.0815) and money supply(0.0013, 0.7135, 0.5671) are found to beinsignificant especially in lag 2 and lag 3 inexplaining changes in price index of stockmarket because the p value is greater than thesignificant level of 5%.

Conclusion

This study mainly focus to investigate therelationship between stock market, inflation rate,exchange rate, money market rate and moneysupply in the case of Colombo Stock Exchangein Sri Lanka. To test stationary and the order ofintegration of all the series, the ADF wasconducted and show that all the variables areintegrated in the same order I(1). The Johansentest for co-integration result indicates that thereis a long run equilibrium relationship betweenthe variables. With the big sample, the optimallag structure for each of the VAR models wasselected by minimizing Akaike’s InformationCriteria. In the final analysis lag of 3 are used.Johansen test procedure confirmed that there is atleast 1 co-integration equation at 5% significantvalue for the ASPI and Macroeconomicvariables. The result of VEC shows that short runrelationship between the stock market index,money market rate and money supply.

Overall from these findings, somerecommendations are suggested. Since the resultsshow there is long run co-integration relationshipbetween stock markets and those variables in SriLanka. A word of caution, the investors whoinvest in the stock market might take risk ingaining benefit from the portfolio diversificationbecause the macroeconomics linked to all shareprice index which lead to stock market.Therefore, stock market returns may beadversely affected by the money market rate andexchange rate, because of increasing moneymarket rate may lead to falls the share marketcapital or stocks.

References

Addy, F. K., Sampson, V. and Yakubu, A. S.(2014), Relationship between stock marketperformance and macroeconomic variables inGhana, Issues in Business Management andEconomics. 2(3), pp. 46-53.

Anokye M. A. and George T. (2008),Macroeconomic Factors and Stock MarketMovement: Evidence from Ghana. School ofManagement, University of Leicester, UK.

Caroline, G., Rosle, M., Vivin, V.C. and Victoria,C. (2011), The relationship between inflationand stock market: evidence from Malaysia,United States and China. International Journalof Economics and Management Sciences. Vol. l, No. 2, pp. 1-16.

Chandni, M., Avneet, K.A. and Saakshi, C. (2012),A Study of the effect of MacroeconomicVariables on Stock Market: Indian Perspective.http://mpra.ub.uni-muenchen.de/43313/MPRA.

Gujrati, D.N. (2007), Basic Econometrics, ForthEdition, McGraw Hill, India.

Johansen, S. and Juselius, K. (1990), MaximumLikelihood estimation and Inference on Co-integration with Applications to the Demandfor Money. Oxford Bulletin of Economics andStatistics. Vol. 52, No. 2.

Keray, R. (2009), Is There a Long RunRelationship Between Stock Prices andMonetary Variables? Evidence from Jamaica,Financial Stability Department, Bank ofJamaica.

Omran, M. and Pointon, J. (2001), Does theinflation rate affect the performance of thestock market? The case of Egypt, EmergingMarkets Review – Vol. 2, pp. 263-279.

Manning, L. M. and Adriacanos, D. (1993), Dollarmovements and inflation: a Co-integrationanalysis. Applied Economics, Vol. 25, pp. 1483-1488.

8


Mukherjee, T.K. and Naka, (1995), DynamicRelations between the MacroeconomicVariables and the Japanese Stock Market AnApplication of a Vector Error CorrectionModel, Journal of Empirical research, Vol. 18,pp. 223-237

Nissim, D. and Penman, S.H. (2003), TheAssociation between Changes in Interest Rates,Earnings, and Equity Values, ContemporaryAccount. Res., Vol. 20, No. 4, pp. 775–804.

Tahir, M.F. and Wong, W.K. (2004), Linkagebetween Stock Market Prices and ExchangeRate: A Causality Analysis for Pakistan,Colombo Stock Exchange Guide book 2013March.

Wickremasinghe, G. B. (2006), Macroeconomicforces and stock prices: Some empiricalevidence from an emerging stock market,Victoria University.

9



10

ABSTRACT

University graduates are facing strongcompetition for top jobs within the country aswell as in overseas. Therefore they have toprepare and develop their careers well beforethey start seeking for jobs as early as theirfreshmen year at university. For a country likeSri Lanka which has the vision “To be themiracle of Asia” it is very important to identifythe gaps of ‘KSA’ of the younger generation ofthe country because ultimately they are the futureof the nation. The concept of career developmentis being considered important and beingdiscussed very much than ever before. It isindeed much more important to focus on careerdevelopment today because unlike earlier, theenvironmental changes are rapid and thus, it isimportant to develop more flexible andcompetent graduates in a country to make itsfuture a bright one. Though many of us don’t paymuch attention on the importance of the careerdevelopment it is indeed important and plays amajor role in deciding an individual’s future. Ofan individual’s perspective it is very important inorder to identify his or her potential, interests,and capabilities in order to find out what shouldbe improved. Thus, the aim of this article is tohighlight the importance of career developmentand identifying the employability skills thatinfluence a person’s career development ofundergraduates in Sri Lanka. This exploratorystudy uses a semi-structured, face-to-faceinterview with 210 undergraduates and 72graduates passed out from South Eastern

University of Sri Lanka based on the simplerandom sampling method. Based on the findings,the skills for employability and careerdevelopment revealed by both undergraduatesand graduates the Language competency is themost important skill with a contribution of 62%for career development, while team work,communication, positive attitude and practicalskills contribute 39%, 35%, 34%, and 33%respectively. The research skills, problem solvingskills, initiative and adoptability skills alsomakes some useful contributions. This study willbe a useful output for the entire society, speciallythe youths to succeed and meet the futurerequirements in any market, especially in adeveloping country like Sri Lanka. Universitiesdo play a key role in developing individualcharacteristics, which is a dominant factor thataffects an individual’s career development.Focusing more in language, communication andteam working skills are important because it willbe a big challenge in the future in cratingglobally employable graduates. Finally, whateverthe barriers we may have, ourselves are thebiggest barrier for the development of our careerbecause of the attitudes and thoughts that weform regarding our own knowledge skills andabilities. So it is imperative to develop thecompetencies of undergraduates with theintention of making them employable in anymarket around the world.

Key Words: Career development, employabi -lity skills, graduates, undergraduates.

CAREER DEVELOPMENT- A KNOWLEDGE BASED

ECONOMY FOR EMPLOYABILITY:

FROM THE PERSPECTIVES SOUTH EASTERN UNIVERSITY

OF SRI LANKA

Mohamed Abdul Cader Salfiya Ummah

South Eastern University of Sri [email protected]

M.N Mohamed Bilal

South Eastern University of Sri [email protected]

Introduction

We all have a general idea of what careerdevelopment is, and the fact is that we don’twidely pay attention to these types of topicsthinking that it will not have much impact on ourlives. Before we move further, let us know whatis really meant by Career Development. “Careerdevelopment is an ongoing process that occursover the lifespan and includes home, school andcommunity experiences related to an individual’sself-concept and its implementation in lifestyleas one lives life and makes a living.” Pietrofesa& Splete (1975), As in “Career DevelopmentChallenges for the 21st century work place” byConlon J Thomas, (2004). Career development isindeed a lifelong learning process or a learningcurve which is not just a matter of one or twodays. So, it can be called as a continuousprocess.

Having introduced career development, it is veryimportant to know for what reasons is it soimportant and what are the key factors that drive,in a person choosing his or her career. Accordingto an article written by Budhathoki K Dhruba in2004 on “Human Resource Management: CareerDevelopment” in which he states that “PEOPLEARE THE SOURCES of all productive effort inorganizations. Organizational effectivenessdepends on the performance of people workingin organizations”.

People are the main resource who drives theorganization towards success. The HR’s successin any organization would mean thesuccessfulness of the entire organization. For acountry like Sri Lanka which has the vision “TOBE THE MIRACLE OF ASIA” it is veryimportant to identify the gaps of ‘KSA’ of theyounger generation of the country becauseultimately they are the future of the nation. Of anindividual’s perspective it is very important inorder to identify his or her potential, interests,and capabilities in order to find out what shouldbe improved. Thus, the aim of this article is tohighlight the importance of employability skills

for career development of university graduateswhich is the need of an hour.

Literature Review

There are many researchers that have beenconducted on various related topics. The previousstudies are focused in every aspect of careerdevelopment. Especially, the employee andemployer expectations, the factors that influencea person’s choice of career path and so on.

The term “Knowledge based economy” resultsfrom a fuller recognition of the role ofknowledge and technology in economic growth.Knowledge as embodied in human beings (ashuman capital) and in technology has alwaysbeen central to economic development. A paperby Jeffery H Gary et al. in 1992 on the topic“Empowering Rural Parents to Support YouthCareer Development” reveals that the ruralsituation presents unique challenges to the careerdecision making process of young people. Unliketheir urban counterparts, youth in rural andremote areas generally have to leave home toattend postsecondary school and more often thannot, have to leave home to partake in any type oftraining which would put them on a career path.Rural parents are faced with difficulties that aresomewhat different than urban parents. One ofthe major factors for rural parents is the financialcost of sending their children to urban centers topursue their career goals. If we consider anotherresearch which was conducted by Shafie.L.A andNayan S. in 2010 on “Employability Awarenessamong Malaysian Undergraduates” This researchwas conducted among 61 students of Diploma inScience at a local university in Malaysia. 40 ofthem were female students and 21 were malestudents. The sample was selected from the listof registered final semester students of Diplomain Science. Based on the finding, the participantschose Personal attributes as the most importantemployability skill. Personal attributes includeloyalty, commitment, honesty, integrity,enthusiasm, reliability, personal presentation,common sense, positive self-esteem, and a senseof humor, motivation, adaptability, a balanced

11

CAREER DEVELOPMENT- A KNOWLEDGE BASED ECONOMY FOR EMPLOYABILITY:

FROM THE PERSPECTIVES SOUTH EASTERN UNIVERSITY OF SRI LANKA

attitude to work and home life and ability to dealwith pressure. This was indicated by theparticipants as personal attributes encompassed25 % of employability skills. The secondimportant employability skill was Team work at17 %. Team work included working as anindividual and a member of a team, coaching andmentoring. This might be caused by theparticipants’ observations that workplace tasksusually require team work and collaborativetasks through classroom learning activities. Thethird important employability skill is Self-Management at 15 %. Self -managementincluded having personal vision and goals,evaluating own performance, takingresponsibility and articulate own ideas andvision. The forth important employability skillwas Communication at 10 %. Learning wascategorized as the fifth important employabilityskill at 8% which included managing their ownlearning, being open to new ideas, contributingto their learning community. Initiative andenterprise was the sixth important employabilityskill at 7 %. It was interesting to note thatTechnology, Planning and Organizing andProblem Solving were given equal importance at6 % as the three least important employabilityskills.

On his paper “Career Choice Factors of HighSchool Students” Borchert.T.M (2002) foundwhat the most preferable occupations were, hereveals that based on his research GermantownHigh School students made multiple referencesto ‘teachers, nurses, and marketing careerchoices. This reflects the interest that is presentin high school students in the coming years.There many professionals in those areas now, aswell as a need for many teachers, nurse, andbusiness majors in the future. I am amazed at thevariety of professions listed. Of the seventy-eightrespondents there were over fifty differentprofessions. The fact that students listed a largevariety of professions, leads the researcher tobelieve that students are exploring careerchoices. Where else would the diversity comefrom? Students have also stated very specificallythe area within some career choices. This alsoleads the researcher to believe that unless

students were asking questions and taking aninterest in the career choice process, the answerswould be more superficial.

In addition there are so many other factors thatresearchers have considered; the futurechallenges are one of them which are alsoimportant. Overall, the literature provides us withgreat evidence of that career development isindeed important.

Research Methodology

This research was conducted among 210undergraduates from all faculties and 72graduates of the south Eastern university of SriLanka, based on simple random samplingmethod. Nature of study was using exploratory,using a semi-structured face-to-face interview.Also this study is qualitative using case study

Results and findings

According to our findings we would like todiscuss some of the key things that we found outthrough our semi structured interviews withsome undergraduates as well as some passed outgraduates of our university. Many graduates andundergraduates stated different views about howthe university has contributed to their careersuccess through the subjects in the curriculum,workshops, short term courses such astranslation, sports education, career guidance andcounseling and extracurricular activities such ascultural shows, sports festivals.

According to one respondent a 3rd year studentof the faculty of Management and Commerce itis really good to have such programmes in whichshe acquired a lot of skills which she did nothave prior to entering the university and shethinks that student should be given theopportunity to organize such programmes and allthe students should take part in thoseprogrammes. Another respondent- a 3rd yearstudent of the faculty of Management andCommerce states that the university has giventhe motivation and it is really a booster becausethe chances have been created and the university

12


is always concerned about the students and hethinks that the participation of all the students tothese programmes should be made compulsory.

A 3rd year student of the faculty of IslamicStudies and Arabic Language says that hisambition is to become a lecturer and he lackedvarious skills prior to entering the university andsuch programmes helped him to develop hispersonality, attitude. Further he says that he wantto learn, teach, do research’s and to spread theresearch’s. He thinks the good things should becontinued and maintained, while anotherrespondent- a 2nd year student of the faculty ofmanagement and commerce says she was a veryshy and a less interactive character and now thefear and the tension is derived out from herthrough participating in such events that isconducted by the university. Her view is thenumber of programmes that are organized shouldincrease in order to make students moreinteractive. Many other students also gavesimilar views.

Discussing on some specific barriers thatstudents think which are blocking thedevelopment of their careers almost 90% of therespondents say or think that it is the languageand the communication skills is the main barrierswhich should be improved.

Discussion with some graduates also was veryuseful where they also shared different views.One respondent who specialized in Accountingand who is now a lecturer and worked in Rainco(Pvt) Ltd states that, entering the universityenabled to improve his English and the 1 yearindustrial exposure at Rainco was the key wherehe moved to metropolitan areas and followedprofessional courses, then got a permanent job atRainco. He says that communication skills aremost important for career development.

According to another passed out student whospecialized in Marketing and is a senior lecturerin Marketing, the base was the curriculum andthe specialization program in Colombo gave hima learning opportunity. Again it was the one yeartraining at Nestle Lanka ltd which gave theopportunity to learn more practical knowledge

and skills and after being an academic it gavemore opportunity to interact with local andforeign scholars. He says that it helped todevelop his career though it was challenging.Another one who completed his degree incommerce and now the owner of PrivateCompany states that the curriculum included anentrepreneurship course unit which induced himto start his own business. Creativity, innovationand initiative were the skills that turned his lifeas an entrepreneur. One recent passed outgraduate who specialized in Accounting, who isworking at Uniliver, Sri Lanka, says that researchskills were the base while another respondentwho is a lecturer who specialized in IslamicFinance says since Islamic Finance was newlyintroduced, visiting lectures from variousindustries were invited where he got industryexposure and several links, and worked as atrainee for 03 months at ADIL Capital (Pvt) Ltd.Another respondent who is also a lecturer whospecialized in Management Information Systems(MIS) says that it was again the curriculum andthe industrial exposure that helped him todevelop his career. One graduate who is now anaccountant in Dubai says that his familymembers supported him to get this job while thedegree and other skills also helped to develop hiscareer, while another one who is now working inQatar as an accountant says it was hisenvironment that created him chances to learnsome professional courses since he was fromColombo, while the accounting special degreealso assisted in developing his career.

When we compare with the currentundergraduates with the graduates of the SouthEastern University, there is a clear differencewhere the graduates feel it was the curriculumand the industrial exposure that helped themdevelop their career.

Based on the findings, the skills foremployability and career development revealedby both undergraduates and graduates theLanguage competency is the most important skillwith a contribution of 62% for careerdevelopment, while team work, communication,positive attitude and practical skills contribute

13

CAREER DEVELOPMENT- A KNOWLEDGE BASED ECONOMY FOR EMPLOYABILITY:

FROM THE PERSPECTIVES SOUTH EASTERN UNIVERSITY OF SRI LANKA

39%, 35%, 34%, and 33% respectively. Theresearch skills, problem solving skills, initiativeand adoptability skills also makes some usefulcontributions.

These skills are gained mainly by the university,through sound curriculum, internship trainingand special programmes on soft skillsdevelopment, which are conducted by careerguidance and welfare units.

Conclusions and Recommendations

In any country, the university students orundergraduates as commonly known, are beingtermed or called as ‘The Cream of The Society’because the society values them so much. If wejust think for a movement WHY ALL THESE?If we are brave enough it should be easilyunderstood that the society or the communityexpects something from us that of which can besome voluble contribution to the society. Havingdiscussed some real factors of what ouruniversity does to develop the career and howsome of our passed out graduates andundergraduates feel about that, we should like toconclude stating that universities do play a keyrole in developing individual characteristics,which is a dominant factor that affect anindividual’s career development. Focusing morein language, communication and team workingskills are important because it will be a bigchallenge in the future in crating globallyemployable graduates. Finally, whatever the

barriers we may have, ourselves are the biggestbarrier for the development of our career becauseof the attitudes and thoughts that we formregarding our own knowledge skills and abilities.

References

Borchert.T.M, (2002), Career Choice Factors ofHigh School Students, The Graduate CollegeUniversity of Wisconsin-Stout

Bilal, M.N.M. & Salfiya Ummah, M.A.C. (2013),The Role of Universities for CareerDevelopment of Undergraduates forEmployability, Abstract, ASAIHL InternationalConference, Colombo

Budhathoki. D.K. (2004), “Human ResourceManagement: Career Development” TheJournal of Nepalese Business Studies Vol. I No. 1

Conlon J. Thomas (2004), “Career DevelopmentChallenges for the 21st Century Workplace”,University of Minnesota, 2004

Jeffery H Gary, et.al. (1992) “Empowering RuralParents to Support Youth Career Development”Canadian Journal of Counselling / RevueCanadienne de Counseling , Vol. 26, No.4

Joeli W. (2013), Career Development under theLiberal Arts Education, Abstract, ASAIHLInternational Conference, Colombo.

Shafie.L.A and Nayan.S. (2010), “EmployabilityAwareness among Malaysian undergraduates”International Journal of Business andManagement Vol. 5, No. 8.

Shyamalee M.M.M.V. et al. (2013), “EmployerPerception on Employability Skills Requiredfor Entry Level Engineering Graduates”,Abstract, ASAIHL International Conference,Colombo.

14


Contribution for Career Development

Contribution for Career Development

ABSTRACT

Rural tourism is highly associated with the livesof local community where women play a majorrole.

Kerala being endowed with huge potential forrural tourism offers tremendous opportunities forthe countryside, particularly womenfolk. Peoplein rural areas are generally considered to bepoor and marginalized and unable to tap theopportunities due to their ignorance, lesseducation, absence of exposure, lack ofinitiatives etc. Thus some external or internalintervention is preferable to drive them forpoverty reduction and community capacitybuilding. Thus, the study tries to find out theagents which initiate and persuade the women toinvolve in rural tourism for their livelihood andidentify activities that can generate employmentfor women

Key Words: rural tourism, employmentopportunities, women participation.

Introduction and Background

Ministry of tourism, Government of Indiadefined rural tourism as any form of tourismthat showcases the rural life, art, culture andheritage at rural locations, thereby benefiting thelocal community economically and socially aswell as enabling interaction between the touristsand the locals for a more enriching tourismexperience. It is essentially an activity whichtakes place in the countryside. The southernIndian state of Kerala being endowed with hugepotential for rural tourism offers tremendousopportunities for countryside, particularly thewomenfolk. Rural tourism is highly associatedwith the lives of local community where womenplay a major role. People in rural areas aregenerally considered to be poor andmarginalized and unable to tap the opportunitiesdue to their ignorance, lower levels ofeducation, absence of exposure, lack ofinitiatives etc. Thus some external or internalintervention is a catalyst to drive them to takeup activities for poverty reduction andcommunity capacity building.

Pro Poor tourism model is one of suchinnovative approach on tourism development andmanagement which result in increasing net

EMPLOYMENT GENERATION FOR WOMEN

THROUGH RURAL TOURISM

IN KERALA

Moli P. Koshy

Professor, School of Management Studies, Cochin University of Science and Technology, Kochi, Kerala, INDIA, PIN 682 022.

[email protected]

Sabira N

Ph. D Research Scholar, School of Management Studies, Cochin University of Science and Technology, Kerala.

Baby Salini P. V

Ph. D Research Scholar, Department of Applied Economics, Cochin University of Science and Technology, Kerala.

15

16


benefit of 'poor' people in developing countries.It boosts the positive impacts of tourism on localeconomy while reducing the negatives impactson the natives. These impacts are different indifferent destinations. It enhances the linkagesbetween tourism businesses and poor people, sothat tourism's contribution to poverty reduction isincreased and the 'poor' people are able toparticipate more effectively in productdevelopment (www.propoortourism.org.uk). Theterm 'Poor', the targeted group, need to beproperly redefined. It can be either skilled orsemiskilled or micro entrepreneurs or ruralhouseholds or womenfolk which depend onstakeholders of tourism, specific to eachdestination area or type of industry (Ashley C,2006). Unemployment, low economiccircumstances, widowhood and desertion of thefamily by the husband compel or become pushfactors for women to seek work outside home.Even though some of them may succeed infinding remunerative occupations many are notable to find jobs. In such circumstances helpfrom Self Help Groups (SHG), NonGovernmental Organizations (NGO),government agencies, and private or publicorganizations in the hospitality/tourism sector actas motivating or facilitating agents for women tofind occupation for themselves.

In Kerala, Kudumbasree (a SHG) has initiatedvarious projects by collaborating tourismdepartments with local community to reap theeconomic benefits of tourism. Gopal et al (2008)have found that community support for tourismdevelopment and the attitudes and hospitality oflocal tourism workers are important forsuccessful tourism. The Implementation ofResponsible Tourism is such kind of initiative inwhich Department of Tourism and Kudumbasreeare supposed to work together to bring localcommunity to get involved in rural tourism.

Thus this study tries to find out the role of socialagents which initiate and persuade the women toinvolve in rural tourism for their livelihood andto identify opportunities for women in ruralKerala.

Objectives

i) To study the kind of rural tourism activitieswomen get involved

ii) To find out the agents/organizations whichinitiate and promote women participation inrural tourism

iii) To identify the kind of activities that canpromote involvement of women in ruraltourism.

Methodology

The study covered 18 village tourismdestinations of Kerala. Both primary andsecondary data are used. Secondary data arecollected from NGOs’ report and primary dataare collected through structured questionnaire.Native women of 18 and above years who areengaged directly or indirectly in tourism industryfor not less than one year are taken as samples.Since the village tourism activities belong toinformal sector, official data is unavailable onworker participation in tourism activities likeworking/assisting at home stay, indirectparticipation of laundry service, working atresorts as laborers etc. So it was estimated tocollect household data from maximum of 65women respondents involved in tourismactivities from each village. Snowball samplingis followed in which sample emerges through aprocess of reference from one person to theother. However some villages failed to fill thequota of 65 where female participation in thevillage tourism activity was very low.Accordingly the sample size of womenrespondents became 660.

Profile of the Respondents

Average age of respondents is calculated as 41years (table 1) and there is no retirement in ruraltourism related jobs as youngest of the cohort isof 18 years and oldest is of 77 years of age. It isvery clear that the kind of tourism jobs whichhave been formulated are in such a way that anyage group can participate in such kind of

17

Employment Generation for

Women through Rural Tourism in Kerala

activities like demo performer, home staybusiness etc. (see table 3). Regarding maritalstatus, 81 percent are married and 10 percent arewidows. Respondents with elementary educationand secondary education constitute 24 percentand 54 percent respectively and only 20 percenthave education above secondary level. Here‘secondary education’ denotes education aboveelementary level and need not be a pass in 10thstandard. As a whole majority of womenengaged in occupations related to village tourismare middle aged, not highly educated, and belongto lower-middle or middle-middle class.

Village Tourism and Employment

Generation

Before going to discuss the employmentgenerated through tourism in each village it isquite appropriate to examine the previousemployment status of respondents, to knowwhich segment of people have been attracted tothe tourism industry. Table 2 shows the type ofemployment that women were having beforeentering tourism industry. It indicates that nearlyhalf of the respondents (48%) were housewivesbefore entering tourism. Another group, largelyattracted to tourism comprised of laborerscalled as ‘coolies’ (24%). Here ‘Coolie’ refersto any manual /unskilled worker not related totourism, e.g., work on farm, construction worketc. Another interesting finding is that 6.5percent of our respondents were working asnurses before entering tourism. The low wagerate prompted them to look for another jobwithout compromising the status that they hadbeen receiving until that time, which finally endsup in running handicraft shop, home stayadministration or even work as sales girl. Otherrespondents were housemaids (3%), fish cleaningand prawn peeling workers (3%), fish sellers(0.3%), sales girls (2.3%), petty shop owners(4.2%), workers who come under MahatmaGandhi National Rural Employment GuaranteeAct -MGNREGA- (0.6%), clerks/office workers(2.3%), artisans (3.0%), teachers (1.4%) andtailors (1.4%) before involving in tourism.

Tourism Jobs

In all villages, tourism jobs were formedaccording to the nature and activity of thedestination. Six kinds of tourism jobs wereclassified based on the uniqueness and attractionsthat each village destination possess such asnature tourism, eco tourism, back water tourism,art/activity based tourism, pilgrimage tourismand beach/aqua tourism. Accordingly fifteenkinds of tourism job were recognized in whichwomen are involved. The table 5 shows sort oftourism related jobs, the percentage ofrespondents involved in each job, description ofeach job and finally how each job is distributedin different kinds of tourism. To illustrate‘catering service’ is feasible in all kinds oftourism, but doing laundry service is feasiblewhere back water tourism exists. Even thoughthe laundry service have a huge prospects invillage tourism as long as resorts/hotels run inthe locality, it has been less preferred due to oflack of professionalism on part of native womenwhile doing laundry work in terms of hygiene,punctuality and caring. Yet it is profitable andconvenient for house boat owners to engage localwomen for laundry work as their boats halt infront of their house during idle time.

Thus, tourism related works can be broadlycategorized in to three: destination inclusiveworks, destination exclusive works, interveningagent motivated works. Inclusive works refers toall kind of jobs and services related to tourismwhich are originated through direct demand, andthus can be performed/operated at all kind oftourism destinations irrespective of uniqueness/attractions/characteristics of the destination. Forexample, jobs/services like running cateringservice, petty shops, home stay and working as asale girl can be performed at any villagedestination regardless of particular characteristicsof the destination. On the other hand, there aresome other works/services which are exclusive innature or location specific. These jobs/servicesare derived because of the peculiarity/uniquenessof that particular destination. Normally theseworks are originated through derived demand

18


like laundry services, resort worker etc. Invillages like Aranmula (one of the EndogenousTourism Project destinations of Government ofIndia) and Thrikkaipetta, where women makeuse of possibilities of tourism only through theircraftsmanship of metal mirror and bambooproducts respectively. There are some otherworks which were formed through theintervention of Self Help Groups (SHGs) or NonGovernmental Organisation (NGOs). In Keralathe intervening agents who facilitate and promoterural tourism are SHGs like Kudumbasree units,other private SHGs, VSS, EDCs, NGOs likeUravu and some people are employed throughMGNREGA.

Channels of Tourism Employment

Six channels namely, self employment, SHGs,resorts/hotels, NGOs, tourism authorities andMGNREGA were identified through whichemployment was generated in various villages.Figure 1 depicts the kinds and proportion ofemployment generated / provided through the sixdifferent channels. The employments createdthrough tourism are mostly self-employed innature as 27.3 percent of respondents are self-employed. If we put aside the self employment,the major channels of employment generation areNGOs (25.3%) and Self Help Group (20.8%).Only 15.9 percent are employed throughgovernment tourism authorities.

Self employment

Self employment is defined as a kind ofemployment where worker earns income throughtrade or business and he is working for himselfinstead of working for an employer who payssalary or wage. Running home stay, doinglaundry work from home, running petty shopand driver (of own vehicle) were the recognizedself employment jobs in village tourism industryof Kerala in which about 17 percent ofrespondents are running petty shops at thetourism spots. Self Help Groups help inemployment generation process by providingassistance to avail micro credit facilities, givingmarketing insights and access to market for the

products. Catering services, sale of vegetables tonearby hotels and resorts are such kind ofemployments generated through SHGs.

Self Help Groups (SHGs)

Self Help Group (SHG) is a voluntaryassociation of poor women and othermarginalized people of 10-20 members who canbe either only women or only men, with similarsocial and financial backgrounds. The groupneed not be registered and women’s groups aregenerally found to perform better. SHG standsfor enabling livelihood opportunities for themembers, enterprising them through microcreditfacilities with bank linkages and strengtheningcollective ability of community both socially andeconomically. SHGs are operated usually byNGOs or by Government agencies likeKudumbasree. Vana Samrakshana Samithi (VSS)is a basic organization instrumental for theimplementation of Joint Forest Management(JFM) in territorial forest division of Kerala.These Samithies are registered under CharitableSocieties Act and recognized by the forestdepartment. VSS work with forest department tomaintain and sustain forest, sensitize and educatepeople. In tourism sector VSS members performas tour guides in reserved forest areas of thestate. Eco Development Committee is also aJFM institution created in protected areas of thestate. Major activities are eco tourism andserving as guides for visitors apart from theactivities undertaken by VSS.

Resorts/hotels

Resorts/hotels at the tourism spots provideemployment mainly in three ways; by employinglocal people especially women for cleaningkitchen and lawns of resorts, by keeping a groupof artists of local community who can performlocal art forms and by keeping a group of demoperformers of local community who candemonstrate local village activities like weaving,coir making etc.

19



Non Governmental Organisations (NGOs

NGOs are the prime promoters of the handicraftindustry and thus the artisans in village tourismcome to 25 percent of total respondents. Uravuis a NGO at Thrikkaipetta tourism villageworking with people, governments andbusinesses to implement programs forsustainable employment and income generationin rural areas. They are generating employmentfor more than 1000 people directly and indirectlyin Wayanad district. Uravu promotes socialenterprises based on value addition of local,natural resources, especially bamboo. Itimplements integrated, end-to-end programs inthe bamboo sector, which include providing skilltraining in bamboo processing establishing microenterprises, marketing of bamboo handicraft,cultivation of bamboo and promotion of eco-tourism. Uravu strives for empoweringmarginalized social groups, especially thetraditional artisans, women and the indigenouspeople. Thus they promote village tourism byproviding training in establishing microenterprises, bamboo products making, marketingof the products etc.

Tourism authorities

Tourism authorities like District TourismPromotion Council (DTPC) run eco shops atsome destinations in which they appoint localpeople for sales. DTPC appoints tour guides inassociation with Vana Samrakshana Samithi(VSS) in some destinations like Kuruva (NorthKerala) and Palaruvi (South Kerala) which comeunder the forest department. Cleaning/housekeeping of the tourism spots is another areawhere DTPC helps local people to be employed.

Mahatma Gandhi National Rural EmploymentGuarantee Act (MGNREGA)

The Mahatma Gandhi National RuralEmployment Guarantee Act (MGNREGA) of theGovernment of India aims at enhancing thelivelihood security of people in rural areas byguaranteeing hundred days of wage-employmentin a financial year to a rural household who

volunteer to do unskilled manual work. In somedestinations, people who joined MGNREGA takethe responsibility of cleaning and maintenance ofthe tourism places.

Suggestions and Conclusion

The following suggestions are made after thisstudy and the authors feel that, if initiatives aretaken and implemented in the right perspective,village tourism can better the lives of ruralpeople.

Location Specific Activities/ Products

As the demands are location/tourists specific,care may be taken to meet the specific demands.We find that most of the tourists in locationssuch as Kulathupuzha, Palaruvi, Thenmala,Puthuvypin and Kuruva are domestic; whereasforeigners prefer places such as Munroe,Thrikkaipetta, Kumarakom and Punnamada. Theproducts/services for sale may be refined to meetthe demands of local, other Indian states, andforeign tourists. So it is suggested to design andinitiate specific strategies to develop, sustain andshow case the ‘offerings’ of a particular village.

Women Centric Activities

Women have lot of opportunities in runningbusinesses in food services, home stay, demoperformance (coir making, pottery, weaving etc.),local art forms, petty shops/wayside vending,laundry service etc. This will enable them todevelop their innate capabilities leading toempowerment and enhanced economic andsocial development.

Training for Women

Women involved in rural tourism areunorganized workers. Thus steps have to betaken to integrate their activities to utilizeopportunities and resources in such a way as tocross fertilize each activity with others. This canbe done either through government bodiesinvolved in tourism development or throughNGOs.

20


Institutional Support

As NGOs and SHGs play a major role in ruraltourism, the collaboration of government tourismdevelopment bodies or panchayaths with NGOsand SHGs could become very effective forimplementation of grass root level initiatives intourism industry. Financial and technicalinitiatives have to be taken to encourage theiractivities.

Governmental Initiatives

Government, through the tourism departmentshould take up active promotion of villagetourism to enhance the employment opportunitiesof rural people and lessen migration to urbanareas in search of employment.

Rural tourism offers affordable tourism to alarger group of people and can reap benefits ifproperly promoted and maintained. Theinvolvement of rural folk, NGOs, SHGs andgovernment agencies can promote employmentparticularly for women when rural activities,local cuisines, art forms etc are promoted. Suchcoordinated efforts enhance the sustainabledevelopment of village tourism destinations.

References

Ashley, C. (2006), How Can Governments Boostthe Local Economic Impacts of Tourism,Overseas Development Institute, Availablefrom http://www.odi.org.uk/sites/odi.org.uk/f i l e s /od i - a s se t s /pub l i ca t ions -op in ionfiles/50.pdf, (27 January 2010 )

George P.O. (2006), Prospects of village tourism,Available from http://www.old.kerala.gov.in/kercaljuly06/pg26-29.pdf (4 December, 2010)

Gopal R, Shilpa Varma and Ms. RashmiGopinathan, Rural Tourism Development:Constraints and Possibilities with a specialreference to Agri Tourism : A Case Study onAgri Tourism Destination – Malegoan Village,Taluka Baramati, District Pune, Maharashtra,Paper presented at the Conference on Tourismin India – Challenges Ahead, IIMK , Availablefrom http://dspace.iimk.ac.in/ bitstream/ 2259/596/1/512-523.pdf , (13 January 2010 )

Jubin Joy John, Hari Sundar. G, Anoop Das andRavikrishnan, Study on Improving thePotential for Village tourism in Kerala – WithEmphasis on Developing the Potential ofLocal Areas of Tourism Importance withSpecial Reference to PonnumthuruthuAnchuthengu and Kumbalangi, Paperpresented at the Conference on Tourism inIndia – Challenges Ahead, IIMK , Availablefrom http://dspace.iimk.ac.in/ bitstream/2259/596/1/512-523.pdf , (13 January 2010 )

21



Tables and Figures

Table 1: Demographic and Socio-Economic Profile of the Respondents

1 Average age (years) 18 77 41.39 10.05

2 Marital Status - - - -

Single - - 5 % -

Married - - 81.5 % -

Separated/Divorcee - - 3.7 % -

Widow - - 9.7 % -

4 Educational status - - - -

Illiterate - - 2 % -

Literate without

formal education - - 0.6 % -

Elementary - - 23.9 % -

Secondary - - 54.1 % -

Above secondary - - 19.4 % -

5 Occupational Status

Government Service 1.8 %

Private Sector Service 34.4 %

Self Employed 63.8 %

7 Average monthly

Household Income INR1000 INR80000 11300.99 7585

8 Average no. of

family members 1 13 3.2 1.28

9 Average No. of years

of participation in

village tourism 1 14 4.27 3.16

Source: Primary data

S.No. Variable Minimum Maximum Value S.D

22


Table 2: Prior Occupation of the Women

1 Coolie (labourer) 24.4

2 Housemaid 2.9

3 Fish sales 0.3

4 Fish peeling 3.0

5 Sales girl* 2.3

6 Petty shops 4.2

7 MGNREGA 0.6

8 Administration/clerical 2.3

9 Artisan** 3.0

10 Teacher 1.7

11 Nurse 6.5

12 Jobless/Housewives 47.7

13 Tailoring 1.4

Total 100

S.No Kind of work Percentage

Source: Primary data, * Sales not related to tourism industry; ** Artisan who does not involved in any activity related to tourism

1

2

3

4

5

Fish/vegetables sales(to house boat/ resort/hotel)

Resort worker

Sales girl

Laundry service

Catering service(Food)

7 (1)

27(4)

62 (9.4)

42(6)

110(17)

collects vegetables and fish fromthe locality and keep supplyingto all the resorts/ hotels andhouseboats

Unskilled daily wage workersemployed in resorts or hotels forcleaning the kitchen andsurroundings.

Person who works at shopswhere sales are meant only fortourists, e.g. handicrafts shops,spice shop, fancy shop at thetourism spot etc.

Washing and ironing of clothes(towels, bed spreads etc.) whichthey receive from houseboatsafter each trip. They do thiswork manually at home withoutany additional investment.

Operate this business in twoways; one, tourists are givenopportunity to customize theirmenu. The other way is toprovide village’s unique or ownfood to the tourists.

Kudumbasree andRegionalAgricultural ResarchStation (RARS),

-Nil

-Nil

Nil

Kudumbasere,

Back Water Tourism (Kumarakom, Punnmada)

Beach and Back watertourism(Cherai,Alappuzha,Punnamada)

Nature tourism (Pookode,Pookode, PalaruviThenmala)

Back water toursim(Punnamada,Kumarakom)

All kind of tourism(Nature, Eco, Back water,art/activity,pilgrimage,and beach tourism)

S.I

No

Kind of

work

Number

(%)

Job Description Support of SHGs/NGO/Organization/

Institution

Kind of tourism

(villages)

Table 3: Type of Work Women Involved in After Entering Village Tourism

23



6

7

8

9

10

11

12

13

14

15

Petty shops

Drivers

Artisan

Tour guide

Housekeeping/cleaning

Handicraft/spice shop

Home stay administration

Eco-Tourism shopsales

Demo performer

Artist/Performer

Total

114 (17.2)

2(0.3)

76(11.5)

26 (4)

80 (12)

62 (9.4)

22 (3.3)

6 (1)

9(1.3)

15(2.2)

660 (100)

Shops which sell locallyavailable food and nonfooditems

Drivers refer to only autorickshaw drivers who are mostlyfound where the majority of thetourists are backpackers whowidely rely on auto rickshawsfor their transportation.

Skilled worker who makecrafts(handicrafts). Eg..makingof bamboo products, metalmirror,

Person who provides assistance,information and cultural,historical and contemporaryheritage interpretation totourists.

Job of cleaning and maintenanceof property and areas within thetourism spots.

Shop which sells handicrafts,spices imported from differentparts of the country.

Management and service ofhome stay

Shop which sells village’s ownproducts (bamboo products,honey, pickles etc.)

Person who demonstrates villageown activities (like coir making,coconut leaf weaving etc) attourist wish.

Person who performs any artform (Classical/folk dance, folksong) for entertaining tourists.Normally Resorts/Hotels utilizethis facility to entertain theirguests

Nil

Nil

Urvavu,Kudumbasree andprivate SHGs

VSS and EDC withthe support oftourism authoritie

Kudumbasree,NREGA

VSS and EDC withthe support oftourism authorities

Nil

EDC

Nil

Kudumbasree

All kinds of tourism(Nature, Eco, Back water,art/activity,pilgrimage,and beach tourism)

Beach Tourism

Atrt/activity Basedtourism (Thrikkaipetta,Aranmula)

Nature and Eco tourism(Kuruva, Thenmala)

Nature, Eco, pilgrimage(Thirunelly, Kuruva,Pookode, Thenamala)

Nature and ECo

All kind of tourism(Nature, Eco, Back water,art/activity,pilgrimage,and beach tourism)

Nature and ECo

Art/activity, beach

Beach and Back water

S.I

No

Kind of

work

Number

(%)

Job Description Support of SHGs/NGO/Organization/

Institution

Kind of tourism

(villages)

Source: Primary data

24


Figure 1: Channels of employment generation and type of employment- shares

Source: primary data

25

ABSTRACT

The Sri Lankan economy has been undergoingstructural changes for the last few decades. Theexistence of dynamic linkages among the threemajor sectors- agriculture, industry and servicesof the Sri Lankan economy are examined in thispaper for the period of 1960-2011. We employedgraphical analysis including scatter plot, linegraph, Confidence Ellipse and Nearest Neighborfit to identify the basic features and therelationship between selected variables. Thedynamic sectoral growth linkage resultssuggested the existence of inter-linkages acrossdifferent sectors of the economy. ADF test andPP test were used to test the unit rootcharacteristics of the time series variables. Intertemporal correlation results show that thereexists a high positive statistically significantcorrelation between all sectors. Engel-Granger(EG) co-integration test provides furtherevidence for this. Error correction modelestimates indicate that short run changes inIndustry GDP have a positive impact on shortrun changes in Agriculture and services GDP.The results of Granger-causality test suggestedthat services growth causes agricultural andindustrial growth significantly. The findingsimplied that by developing the services sectors,agricultural and industrial growth can bestimulated.

Key Words: Growth linkages, Sectoral GDPdynamics, Cointegration, Causality

Introduction

Sri Lankan economy has been experiencingstructural changes in sectoral composition overthe last few decades. The contribution of the

agricultural sector to national GDP has been fastdeclining and service sector is dominating andhas been showing remarkable improvements.The high dependency of the Sri Lankan economyon agriculture during 1950s constituting 46.3%of GDP in 1950 indicated a downward trend inlater years. The emergence of Sri Lanka fromagricultural to service-driven economy during themid-1990s is an important milestone, changingproduction linkages across sectors in contributingto economic growth.

Problem Statement

Most early development strategies, advocated byRosenstein-Rodan, Nurkse, and Hirshman amongothers, emphasized industrial development as themain source of economic growth. (Schiff andValdez, 1998). The role of agriculture ingenerating economic growth was shown to beminimal with the experience from the newlyindustrialized countries and others. The servicessector was identified as an emerging sector.Based on this evidence, the present paperidentifies the importance of investigating thedynamic relationships between these sectors. Wetry to identify uni-directional and bi directionalrelationships between major economic sectorsand investigate whether the sectors are able togenerate growth in other sectors. We select SriLanka as the context for identifying sectoralgrowth linkages.

There is a significant gap in the growth literaturein Sri Lanka. Most of the studies on inter-sectoral linakges were conducted for Africancountries and India. So far there were no in-depth analyses on sectoral dynamic growthlinkages in Sri Lanka. This study fills this gap inthe existing literature by providing acomprehensive analysis of inter-sectoral linkagesin the Sri Lankan economy. An in-depth

INTER-SECTORAL DYNAMIC GROWTH LINKAGES:

EMPIRICAL EVIDENCE FROM SRI LANKA

Selliah Sivarajasingham

University of Peradeniya

SaNa

Typewriter

Thamarasi Kularatne and

SaNa

Typewriter

26


understanding of sectoral growth dynamicsbecomes more important for policy formulationin designing structured and balanced growth inthe economy. The use of studying these dynamiclinkages between sectors is found to be veryimportant in development planning to achieve abalanced sectoral development within thecountry.

As mentioned before, proper understanding ofthe linkages between sectors and identifying thekey growth-boosting sectors is crucial for theformulation of long term policy in Sri Lanka toachieve sustained economic growth of around6% which is projected by Mahinda Chinthana.

The motivation of this study is to answer thefollowing set of questions; (i) Are there positivesignificant linkages between all three sectors?(ii) Are these linkages unidirectional orbidirectional? The present study incorporatesvisual inspection and time series analysis toanswer the above research questions.

Objectives

The main objective of this paper is to study thedynamic growth linkages between agricultural,industrial and service GDP’s for Sri Lanka. Theobjective is achieved by studying the long runrelationship and the causal relationship betweenthe three major sectors on the economy for theperiod of 1960-2012. Specific objectives of thisstudy are:

1. to investigate the dynamics growthrelationships among sectors during thesample period

2. to investigate the direction of causalrelationships between growth of sectors

The paper will be structured as follows. Section2 gives an overview of the theoretical andempirical background to the study, section 3explains the methodological framework, section4 presents the empirical results, sections 5discusses the conclusions and recommendationsand section 6 provides the limitations of thestudy.

Theoretical and Empirical

Background

Literature Review

As a result of industrialization, resulting in theemergence of the contribution of non-agriculturalsectors to national GDP, many economistsdeveloped the interest in studying the linkagesbetween the major economic sectors: agriculture,industry and services. ‘The concept of sectorallinkage, which evolved from Hirschman's theoryof ‘unbalanced growth’, has been recognized asplaying a crucial role and providing substantialcontributions towards guiding the appropriatestrategies for future economic development’(Saikia, 2011).

The sectoral linkages literature begins with thediscussion of structural transformation ineconomies. Lewis (1954), one of the influentialcontributors in this regard, models thedevelopment process in terms of a structuraltransformation from agriculture to industrialactivities. This provides evidence of a interrelationship between agriculture and industrialsectors as Dorwrick and Gemmell (1991)investigate in their study that high productivity inthe industrial sector has positive externalityeffects on the agricultural sector.

Hwa (1988) has recognized that agriculturalgrowth is strongly linked to industrial growth; heidentifies this relationship as one ofinterdependence and complementarity. Althoughthe literature stresses the agriculture-to-industrylinks, there could be also a possibility andpotential for reverse linkages. Gemmell et al.(1998) points out that “a domestic source ofindustrial input to the agricultural sector canrelease bottlenecks: rising industrial wages canfoster growing agricultural demand”. Moreover,Bhagwati (1984) explains how services sectorlinks with the production of goods throughsplintering process (production of goods emergefrom services and vice versa). “The contributionsof intermediate services such as distribution andretailing to both agriculture and manufacturing

27

Inter-Sectoral Dynamic Growth Linkages:

Empirical Evidence from Sri Lanka

are obvious and frequently observed (from input-output tables) to increase overtime in LDCs”(Gemmell et al, 1998).

Authors (Saikia 2011, Rashid 2004, Fiess andVerner 2001 etc.) have come up with interestingfindings regarding the linkages between sectorsfor different economies, amongst few are listedbelow. In India, ‘agriculture-industry’ linkage hasundergone directional changes as both theproduction and demand linkages, which wereprimarily from industry to agriculture in the pre-reform period, transformed to agriculture toindustry in the post-reform period (Saikia 2011).In the same study, the author has found no anysignificant interdependence between agricultureand service sectors, while there is stronginterdependence between industry and servicesectors and it has improved in the post-reformperiod.

Fiess and Verner (2001) found that agriculturalsector as a major driving force in sectoral growthin Ecuador. Study by Rashid (2004) investigatedthat industrial sector is the leading sector inPakistan economy because the industrial sectorgranger causes aggregate GDP growth,agriculture growth as well as service sectorgrowth. Study by Kaur et al. (2009) for theIndian economy, found that the agriculturalsector exhibits strong association with theindustrial sector, while the converse connectionin terms of demand linkages of industry with theagricultural sector have weakened in the last twodecades. Demand linkages of the services sectorwere observed to have strengthened in relation tothe industrial sector overtime. Similarly, study byGemmel et al. (1998) suggests that productivitytechniques in manufacturing tend to spill over toagriculture, so encouraging convergenttendencies in sectoral productivity levels. Theresults for Romania indicated that the industrialsector is detrimental to agriculture however, theservice sector contributes positively(Subramaniam and Reed, 2009). In the samestudy,the long-run relationship of industrial,service and trade sectors to agricultural sector

were established, and the results show that theindustrial sector in Poland contributes positivelyto the agricultural sector while the growingservice sector shows mixed results. Craigwell etal. (2008) found that in Barbados in recent years,an expansion in services output was found to bethe only determinant of industrial output in boththe short and long run, as agricultural output didnot appear to have any statistically significant.

Theoretical Background

The idea of the importance of sectoral linkagesin economic growth is clearly advocated byHirshman(1958) in his theory of unbalancedgrowth. This theory stresses on the need ofinvestment in strategic sectors of the economyinstead of all the sectors simultaneously.According to this theory the other sectors wouldautomatically develop themselves through whatis known as linkages effect. Creating unbalancesare a pre-requisite of economic growth,according to Hirschman. However the questionarises, how to identify the activities with whichto create imbalances in the system. Thisnecessitates the knowledge of inter linkagesacross different sectors of the economy.Hirshman classified these linkages as forwardand backward linkages.

Thirlwall (2006) defines forward and backwardlinkages as follows. “Backward linkages measurethe proportion of an activity’s output thatrepresents purchases from other domesticactivities. Forward linkages measure theproportion of an activity’s output that does notgo to meet final demand but is used as inputsinto other activities.”

This study investigates the existence of linkagesbetween major sectors of the economy andquantifies their extent. This measurement aids inidentifying what sort of linkages exist betweensectors in the Sri Lankan economy which isfurther explained in sections 4 and 5.

28


Methodological Framework

Data:

The variables used in this study are Agriculture,Industry and Services Gross Domestic Product(GDP) at current market prices (given in Rs.Million). The data for the study period 1960-2011 are collected from the Central Bank AnnualReport 2011. All variables are transformed intonatural logarithmics.

Analytical tools:

In this study, the graphical analysis (scatter plot,line graph, Confidence Ellipse, Nearest Neighborfit) is used to identify the basic features of thevariables and to identify the relationship betweenselected variables. ADF test and PP test are usedto test the unit root characteristics of the timeseries variables. Moreover, co-integrationtechnique and error correction model areemployed to study the long run equilibrium andshort run inter-sectoral equilibrium relationship.In addition, Granger causality method is used toidentify the direction of causal relationshipbetween sectoral growth rates.

Empirical Results

We employ visual inspection to identify thebehavior of variables and their relationships.Firstly,it would be useful to review the changesin sectoral composition of Gross DomesticProduct before identifying the sectoral linkagesin the Sri Lankan economy. Figure-1 (a) showsAgriculture, Industry and Services GDPdynamics from 1960-2011.

The pick-up in GDP growth for Sri Lanka wassupported by all the sectors with a markedacceleration. Sectoral GDP of all three sectorsare rising although the importance of each sectorstarted to vary from late 1970s.

In respect of comparison of sectoral shares inGDP since the 1960s, the relative share ofagriculture is declining overtime. In 1960, thecontribution of agriculture to GDP is 37.8% andit continuously decrease with slight fluctuationsto 12.1% of GDP in 2011, become the relativelyleast contributor to GDP. The share of industrialsector to GDP increases from the lowestcontribution of 16.8% in 1960 to 29.9% by 2011becoming the second largest contributor.However, the share of industry remains more orless unchanged (between 26% to 29%) sinceearly 1980s to present. Services sector on theother hand, remains the highest contributor toGDP throughout the period whose contributionlargely accelerates after the economicliberalization in 1977. Services sector share riseup to 58% by 2011.

Figure-1 Sectoral GDP dynamics

29



Next, the growth rate of the three sectors areplotted to visualize the dynamics of each sectorgrowth throughout the study time period.According to Figure-3 and summary statisticsgiven in appendix 1, the growth of agriculturalsector exhibits the highest variability of thegrowth rate than industry or services growthrates.

Volatility in growth was measured by coefficientof variation (Table 1). Services sector remainedthe least volatile sector as compared toagriculture and industry. Agricultural growthdisplays the highest volatility during the studyperiod.

The following part of the analysis focuses ontherelationships between variables using theconfidence ellipse plot. Firstly, the confidenceellipse was plotted for the sector GDPs. Figure 4illustrates how the sector GDPsare related witheach other from 1960 to 2011.

We visualize the highest degree of positiveassociation between services sector andindustrial sector, in which the degree isassociation compatible with inter temporalcorrelation results; r = 0.9998 (refer appendix 2).The confidence ellipse of agriculture andservices GDP (r = 0.9775) and agriculture andindustry GDP (r= 0.97728) also display a strongpositive association. All correlation results arestatistically significant different from zero at 5percent level.

The same visual analysis was carried out in logforms as shown in figure 5. A similar but a morestrong relationship was observed between thesectors when confidence ellipse was plottedusing the log transformation.

Sectors Coefficient of

variation

(1960-2011)

Agriculture 1.0393831

Industry 0.5270767

Services 0.4320362

Table- 1: Growth rates volatility

30


A more positive association between all sectorsis displayed with a greater magnitude.Correlation results support the above graphicalanalysis indicating pairwise correlation valuesabove r = 0.99 in all (see appendix 2).

Next, confidence ellipse was plotted for thegrowth rates of the three sectors. Although theassociation between sectors is not so strong whencompared with figure 4 and 5, the growth rateassociation still remains positive and statisticallysignificant, but weak. A relatively more positiverelationship could be identified betweenindustrial sector and services sector (r =0.66)when considering the growth rates. Agriculturaland industrial sectors display a positiveassociation of r=0.386 whereas agricultural andservices sectors association is given as r=0.287(appendix 2). This preliminary analysisdemonstrates that these sectors are positivelyrelated and provides evidence that all growthlinkages are statistically significant.

Unit root test:

Prior to testing for co-integration, unit root testsare performed on each of the sectoralGDPgrowth series. In order to check the

stationarity, this study analyses the order ofintegration of sectoral GDP growth ratesbyusingunit root test based on Augmented Dickey Fuller(ADF) in table 3. The variables under study arechecked for stationary with an equation withintercept on the basis of ADF. All the variablesarenon-stationary, integrated order one, I(1) atlevels. The ADF test statistics are found notsignificant at 1% with comparison MacKinnoncritical values.

As all the variables are found stationary at firstdifference so it is now feasible to apply cointegration test to verify the long run relationbetween variables.

Table 2: Unit root test

Engel-Granger (EG) co-integration test:

Engel-Granger (EG) co-integration test resultsshow that there exists a strong long runequilibrium positive relationship amongst eachsector GDP growth rates in a bivariateframework. Residual series of each cointegrationregression equation is tested for unit root. Thesetest results confirm all residual of cointegrationregression equation is stationary. Unit root testresults of each residual series are in appendix 3.Therefore, sets of concerned variables arecointegrated and have long run equilibriumrelationship.

The results of estimated cointegration regressionequation are given below.

GROWIND = 0.1156 + 0.253943GROWAGR(1)

P value (0.0000) (0.0047)

31



GROWSERV = 0.122061 + 0.149415GROWAGR(2)

P value (0.0389) (0.0000)

GROWSERV = 0.062889 + 0.526845GROWIND(3)

P value (0.0000) (0.0000)

GROWIND = 0.026862 + 0.843164GROWSERV(4)

P value (0.1877) (0.0000)

GROWAGR = 0.034208 + 0.552608GROWSERV(5)

P value (0.3879) (0.0389)

GROWAGR = 0.026455 + 0.586853GROWIND(6)

P value (0.4146) (0.0047)

Equation 4 shows that growth of services sectorinfluences growth of industrial sector by 0.84percent which is considerably higher. Similarly,equations 3, 5 and 6 indicate high long runequilibrium relationships. Lowest long runrelationships could be identified in growth ofagricultural sector influencing growth ofindustrial sector and growth of services sector.

Table 3: Long run and Short run responses

and adjustment speeds

Highest long run statistical significant effectcome from service sector to industrial sector.Least long run significant effect comes fromagriculture to service sector. The highest shortrun statistically significant marginal effect comesfrom service to industry. Service and agriculturehave no significant short run effect in bothdirections.

Error Correction Model

The Error Correction Model results of agricultureand services growth equation indicate that errorcorrection coefficient is -0.906 and is statisticallysignificant. The short run marginal impact (0.19)of service sector growth on agriculture growth isnot statistically significant (p=0.46). Thenegative sign of error correction coefficientindicates that agriculture growth moves towardsthe long run time path. 90 percent of thedisequilibrium is corrected each year.

Short run impact of industrial sector growth hasa statistically significant positive impact (0.468)on growth of agricultural sector. Error correctioncoefficient estimate is -0.89 (0.000) statisticallysignificant. The negative sign of error correctioncoefficient indicates that agriculture growthmoves towards the long run time path in theindustrial and agricultural growth linkages.

Short run impact of industrialsector growth havea statistically significant positive impact of0.404(0.000) on growth of service sector. Errorcorrection coefficient estimate is -0.677(0.000)and it is statistically significant. It shows thespeed of the services growth towards theequilibrium state. It indicates that 67 percent ofthe deviation from long run equilibrium path iscorrected in each year.

Granger Causality test

The Granger causality test exhibits the pair wisecausal relationship between the variables underconsideration. It may be unilateral or bilateraleither way. So, this study also uses the test tofind the causal relationship between growth ratesof each sector separately. Table 4 illustrates thepair wise Granger Causality estimation. The firstcolumn shows theNull hypotheses for possiblerejection at different significance level whilethird and fourth columnsindicate F statistic andprobability respectively.

The Granger Causality estimation wasundertaken for lags 4, 5 and 7 to identify thecausal relationship between growth rates at

32


different levels. At 7 lags, significantbidirectionalcausality is observed between agriculture andservices growth rates whereas only unidirectionalcausality exists between agriculture-industry andindustry-services. Agriculture granger causesindustry and industry granger cause servicessignificantly based on the growth rates. At 5 lags,bidirectional causality was observed between thesectors agriculture-services and industry-services.But only agriculture Granger cause industrygrowth rates and not vice versa. At 4 lags, weobtained similar results as in 5 lags. The resultsof the Granger causality tests imply that thegrowth of Agriculture and services sectors areable to generate growth in each other stronglywhich proves the existence of inter linkages in-between. Growth in Industrial sector has longterm impacts on the growth of services but thetwo sectors are highly causal in the short termthus showing inter sectoral linkages. Growth ofagricultural sector at all times is only able togenerate growth in industrial sector, hence thereexists bidirectional linkages.


The overall analysis of the empirical resultsestablishes evidence for inter sectoral linkages atdifferent extents. The preliminary analysis(figures 3,4 & 5) graphically demonstrates thestrength of the relationship between sectors. Theconfidence ellipses substantiate a strongerrelationship between industry and servicessectors than the rest which is backed by thecorrelation and causality results at level moredepth and concrete.

The inter relationship between industry andservices sectors provides evidence for theexistence of forward and backward linkagesbetween the two sectors. As stated by Thirlwall(2006), one sectors output is used as inputs inanother sector creates forward linkages and thatproduction one sectors’ output requires thepurchase of inputs form another sector createsbackward linkages as further explained in section

2.2. This is bidirectional according to the resultsof above two sectors in Sri Lanka.

Although agriculture is able to generate growthin the industrial and services sectors, industrialsector is not capable of generating growth in theagricultural sectors based on the results. Theabsence of backward linkages could be a reasonin the sense that agricultural sector productiondoes not require inputs form the industrial sector.Growth of agriculture can help the emergence ofmany industrial sector activities but the oppositeis absent for the Sri Lankan scenario.

Services sector appears to be very significant ingenerating growth in both sectors thus, identifiedas a strategic sector of the Sri Lankan economy.This highlights the importance and the need ofgovernment policies to be focused towardsdeveloping this so called strategic sector so thatit would enable the growth and development ofthe other two sectors.

Limitations

The analysis in this study in only limited toidentifying whether there exists linkages betweenmajor economic sectors of the Sri Lankaneconomy and the degree of the relationship. Itdoes not go to the extent of identifying what arethese forward and backward linkages betweeneach sector, which leaves space for furtherdevelopments in research in this regard.

33



References

Bhagwati, J.N. (1984), Splintering anddisembodiment of services and developingnations, World Economy, Vol. 7, pp. 133-43.

Craigwell, R. Downes, D. Greenidge, K. Steadman,K. (2008), Sectoral Output, Growth andEconomic Linkages in the Barbados EconomyOver the Past Five Decades. AppliedEconometrics and International Development.Vol. 8-2.

Dowrick S.J. and Gemmell N. (1991),Industrialization, Catching up and EconomicGrowth: A ComparativecStudy Across TheWorld’s Capitalist Economies. EconomicJournal,Vol. 101, pp. 263-275.

Gemmell, Norman, Lloyd, T. and Mathew M.(1998), Dynamic sectoral linkages andstructural change in a developing economy.CREDIT research paper.Centre for Research inEconomic Development and Internet 98/3.

Hirshman A.O. (1958), Strategy of EconomicDevelopment. New Haven, Conn. YaleUniversity Press. pp. xiii, 217

Hwa, E.C (1988), The Contribution of Agricultureto Economic Growth: Some EmpiricalEvidence.World Development, Vol. 16, No. 11,pp. 1329-1339.

Kaur, G. Bordoloi, S. and Rajesh, R. (2009), Anempirical investigation on the inter-sectorallinkages in India. Reserve Bank of IndiaOccasional Papers. Vol. 30, No. 1

Lewis W.A., (1954), Economic Development withUnlimited Supplies of Labor. ManchesterSchool. Vol. 22, pp. 139-191.

Linden, M. and Mahmood, T. (2007), Long runrelationships between sector shares andeconomic growth A panel data analysis of theSchengen region. DP-50, Department ofEconomics, University of Joensuu.

Rashid, Abdul (2004), Sectoral Linkages;Identifying the Key Growth Stimulating Sectorof the Pakistan Economy. Pakistan BusinessReview. Vol. 6, No. 1, pp. 10-39.

Saikia, D. (2011), Analyzing inter-sectoral linkagesin India. African Journal of AgriculturalResearch Vol. 6 (33), pp. 6766-6775.

Sastry, D.V.S. Singh, B. Bhattacharya, K. andUnnikrishnan, N.K. (2003), Sectoral Linkagesand Growth: Prospects Reflection on the IndianEconomy.Economic and Political Weekly. June14, Vol. XXXVIII (24), 2390-97.

Schiff, M., Valdez, A., (1998), Agriculture and themacroeconomy. In: Gardner, B., Rausser, G.(Eds.), Handbook of Agricultural Economics.Elsevier Science, Amsterdam .

Subramaniam, V. and M. Reed. (2009),Agricultural Inter-Sectoral Linkages and ItsContribution to Economic Growth in theTransition Countries. Contributed Paperprepared for presentation at the InternationalAssociation of Agricultural EconomistsConference, Beijing, China, August 16-22,2009.

Thirlwall, A.P. (2006), Growth and Development,Eighth Edition: With Special Reference toDeveloping Economies. 8th Edition. PalgraveMacmillan.

34

ABSTRACT

The main purpose of this research study is toinvestigate the relationship between intrinsic jobsatisfaction factors such as: Ability Utilization,Achievement, Advancement, Creativity andMoral Values and demographic factors: Gender,Ethnicity, Age, Civil Status, Educationalqualifications and Years of Experience ofgovernment school teachers working atKalmunai zone, Ampara district in Sri Lanka.For this research study twelve governmentschools were selected randomly from Kalmunaieducational zone. In order to collect requireddata for the study, the Minnesota SatisfactionQuestionnaire (MSQ) standard method was used.

Reliability Analysis, Principal componentAnalysis, t-test and One-way ANOVA were usedto analyze the data. This research studysuggested that there is no significant differencewas found in the intrinsic satisfaction betweenmale & female teachers and married &unmarried teachers. Age and Years ofexperiences didn’t explore the intrinsicsatisfaction difference in teachers. But, theteacher’s educational qualifications significantlyimpact on intrinsic job satisfactions.

Key Words: Job Satisfaction, GovernmentSchools, Intrinsic Factor, Kalmunai Zone.

Introduction

A study of job satisfaction is a major researchactivity throughout world in all walks oforganizational life including education. More ofthe work is done in business and industry,however, a reasonable number of studies havealso been conducted in the field of educationthroughout the world. Every individual needs jobto fulfill basic needs. It shares in strengtheningthe financial basis for individuals’ life style.Therefore the job satisfaction is a mostinteresting field for many researchers to studywork attitude in workers (Koustelios, 2001).

Due to better performance shown by satisfiedworkers, it is the top priority of all organizationsto achieve the desired goals by increasing theirjob satisfaction (Chambers, 1999). It is alsoimportant due to its significance with thephysical and mental wellbeing of workers. It isclosely related to behaviors, such as productivity,absenteeism and turnover. Besides itshumanitarian value it makes the economic basisto get maximum financial remuneration.

When teachers are satisfied with their job theycan perform responsibilities with moreconcentration and devotion (Rajkatoch 2012).

In this research study the demographic factors ofgovernment school teachers in Kalmunai zone inSri Lanka is compared with their intrinsic job

ANALYZING THE IMPACT OF INTRINSIC JOB

SATISFACTION OF GOVERNMENT SCHOOL TEACHERS

SPECIAL REFERENCE TO KALMUNAI

EDUCATIONAL ZONE, SRI LANKA

Aboobacker Jahufer

Senior Lecturer in Statistics, Department of Mathematical Sciences, Faculty of Applied Sciences, South Eastern University of Sri Lanka.

[email protected]

M.H.M. Sarjoon

Faculty of Applied Sciences, South Eastern University of Sri [email protected]

35

Analyzing the Impact of Intrinsic Job Satisfaction of Government School

Teachers Special Reference to Kalmunai Educational Zone, Sri LAnka

satisfaction. For this purpose MSQ standardquestionnaire was administrated to governmentschool teachers in Kalmunai zone to collect data.Questionnaire is prepared with five point likert -scaling system (1-Highly not satisfied, 2-Notsatisfied, 3-Satisfied, 4-Very satisfied and 5-Extremely satisfied). The questionnaire wasdivided into two sections. Part-A consistingquestions relating to personal characteristics ofthe respondents known as demographic factors,part-B consisting questions relating to jobsatisfactions.

Intrinsic job satisfaction can be measuredthrough the variables: Ability utilization,Achievement, Advancement, Creativity andMoral values. Each variable consists 5 sub-questions to measures the main variable.

The study comprised 64 government schools inKalmunai zone in Ampara district Sri Lanka.Twelve Schools were randomly selected assample to achieve the goal of this research study.There are 635 government school teachers ofwhich 271 males and 364 female teachersworking in the selected schools. Out of 635teachers 170 (27%) teachers were randomlyselected from twelve schools to collect data forthis research of which 73 male and 97 femaleteachers and also equal number of educationalqualification (G.C.E.(A/L), Diploma & Degree)teachers.

To achieve the goal of this research study thisresearch paper is composed into five sections.Section 2 derives the literature review; section 3gives research methodology; section 4 describesdata analysis and discussions. And in the lastsection conclusions and recommendations aregiven.

Literature Review

With respect to the two-factor theory of jobsatisfaction both components, intrinsic andextrinsic, are essential for dentists but thepresence of intrinsic motivating factors like theopportunity to use abilities has most positive

impact on job satisfaction according to Goetz(2012).

Herzberg (1959) two-factor theory (also knownas motivator-hygiene theory) attempts to explainsatisfaction and motivation in the workplace.Motivating factors are those aspects of the jobthat make people want to perform and providepeople with satisfaction, for exampleachievement in work, recognition, promotionopportunities. These motivating factors areconsidered to be intrinsic to the job, or the workcarried out.

Herzberg (1959) revealed that intrinsic factorsare related to job satisfaction, in other wordswhen people felt satisfied and happy at work theconditions present were directly affecting theirinner feelings and self esteem, further intrinsicelements of the job are related to the actualcontent of work, such as achievement,recognition, the work itself, responsibilities andadvancement. These were referred to as‘Motivational’ factors and are significantelements in job satisfaction.

Sitizawaiah and Zahari (2006) highlighted thatthe significant influence of age, experience andmarital status on job satisfaction. Further,environmental factors, especially thesurroundings, context dependence and thebuilding’s function also had a significant impacton job satisfaction.

According to Sharma and Jeevan (2006) thedegree of job satisfaction secured by teachers isnot high and the reason lies in insufficient pay.Secondary level teachers are more satisfied thanprimary level teachers. Contrary to expectation,private school teachers are more satisfied thangovernment school teachers despite the poor paypackage, due to the congenial atmosphere in theprivate schools. Female teachers are moresatisfied due to the nature of the job and thesocio-cultural value of the profession. The levelof education inversely affects the pay satisfactionof the employees working at the same level.Satisfaction with teaching as a career, not merelyas a job, is an important policy issue since it is

36


associated with teacher effectiveness, whichultimately affects student achievement.

Satisfaction has been generally viewed asfunction of worker’s rewards and expectations bymany researchers. That is why workers whopossess better rewarding jobs have highersatisfaction than with little intrinsic and extrinsicvalues (Kalleberg, 1977).

Teachers are arguably the most important groupof professionals for our nation’s future.Therefore, it is disturbing to find that many oftoday’s teachers are dissatisfied with their jobs(Andre Bishay, 1996).

Research Methodology

Objective of the Research Study

This research study intends to achieve thefollowing objectives:

1. To Study the relationship betweendemographic factor Gender and the intrinsicjob satisfaction of government schools inKalmunai zone, Sri Lanka.

2. To Study the relationship betweendemographic factor Ethnicity and theintrinsic job satisfaction of governmentschools in Kalmunai zone, Sri Lanka.

3. To Study the relationship betweendemographic factor Age and the intrinsic jobsatisfaction of government schools inKalmunai zone, Sri Lanka.

4. To Study the relationship betweendemographic factor Civil Status and theintrinsic job satisfaction of governmentschools in Kalmunai zone, Sri Lanka.

5. To Study the relationship betweendemographic factor EducationalQualifications and the intrinsic jobsatisfaction of government schools inKalmunai zone, Sri Lanka.

6. To Study the relationship betweendemographic factor years of Experiences andthe intrinsic job satisfaction of governmentschools in Kalmunai zone, Sri Lanka.

Theoretical Frame work

Independent Variables Dependent Variables

37



Data Analysis and Discussion

Frequency Distribution Analysis for

Demography Factors

The frequency distribution table for demographyfactors is given in Table 4.1.

Reliability Analysis

Before applying statistical tests and analyses,testing of the reliability of the scale is very muchimportant as its shows the extent to which a scale

produces consistent result if measurements aremade respectively. This is done by determiningthe association in between scores obtained fromdifferent administrations of the scales. If theassociation is high, the scale yields consistentresult, thus is reliable. Cronbach’s alpha is mostwidely used method to check the consistency ofvariables. It may be mentioned its value variesfrom 0 to 1. But satisfactory value is required tobe more than 0.6 for the scale to be reliable (Hairet al., 2008). The cronbach’s alpha estimatedvalues for intrinsic job satisfaction factors are

Table 4.1: Frequency distribution table for demography factors

given in table 4.2. As the cronbach’s alpha valuesin this research study were all much higher than0.6, the constructs were therefore deemed tohave adequate reliability.

Principle Component Analysis (PCA)

The PCA is very useful analysis to reduce thelarge number of correlated variables intouncorrelated variables. Further first one or twoprincipal components are enough to explain thetotal variation of the original variables. In thisstudy, intrinsic job satisfaction measured byabove 5 factors. Each factor has 5 indicators orsub variables. These sub variables were reducedas one variable using principal componentanalysis that contains large amount ofinformation. Generally the reduced variables arecontaining more than 70% of information isacceptable to explain the original variables

(Jahufer, 2005). Accordingly, the reducedvariables were contained the followingpercentage of information: Ability Utilization,Achievement, Advancement, Creativity and

Moral values explained by 73.5%, 74.1%,73.3%, 81.4% and 83.4% of original informationrespectively.

t-Test for Two Levels Demographic Factor

Variables: Gender, Ethnicity and Civil

Status

Results of independent sample test that wasimplemented with the purpose of testing whetherthere is meaning full differences between male &female, Muslim & Tamil and unmarried &married teachers in terms of intrinsic jobsatisfaction and the results are given in Table 4.3.

38


Table 4.3: The Independent Sample t-Test Results for Gender,

Ethnicity and Civil Status Variables

Table 4.2: Cronbach’s Alpha Values for Intrinsic Job Satisfaction Factors.

From the above table 4.3 the probability valuesare: P = 0.130, P = 0.018 and P = 0.563 forvariables Gender, Ethnicity and Civil Statusrespectively. Since intrinsic job satisfaction isstatistically not significant at 5% level forvariables Gender (Male & Female) as well asCivil Status (Unmarried & Married) these meanmale & female and unmarried & marriedteacher’s intrinsic job satisfaction not different.Whereas, intrinsic job satisfaction is statisticallysignificant at 5% level for Ethnicity variable(Muslim and Tamil), this means Muslim & Tamilteacher’s intrinsic job satisfaction are different.Moreover, according to the mean value, Tamilteacher’s intrinsic job satisfaction is higher thanMuslim teacher’s.

ANOVA Test for More Than Two Levels

Demographic Factors: Age, Educational

Qualification and Years of Experience

Results of ANOVA that was implemented withthe purpose of testing whether there are meaningfull differences between teachers’ Age,

Educational Qualifications, and Years ofExperience in terms of intrinsic job satisfaction.The ANOVA results are given in Table 4.4.

According to the table 4.4, the probability valuesfor variables age and years of experiences onintrinsic job satisfaction of teachers are 0.412and 0.354 respectively. And the P-values aremore than 0.05 this indicates that there is nosignificant impact of variables age and years ofexperience on the intrinsic job satisfaction ofteachers that is, each categories of variables ageand years of experience have same level intrinsicjob satisfaction. However, the probability valuefor Educational Qualification is P=0.019 thisindicates that there is a significant impact ofteacher’s Educational Qualifications on theintrinsic job satisfaction of government schoolteachers at 5% significance level. That is theteacher’s educational qualification categorieshave different levels of intrinsic satisfaction.Hence, multiple mean comparison test wascarried out and the results are shown in Table 4.5.

39



Table 4.4: ANOVA results by Age, Educational Qualifications

and Years of Experiences

From the multiple comparisons results in abovetable 4.5, it can be concluded that G.C.E.(A/L) &Degree, Diploma & Degree are significant at10% level. According to Tukey mean comparisontest G.C.E. (A/L) qualification teacher’s intrinsicjob satisfaction is different from Degreequalification teacher’s and also Diplomaqualification teacher’s intrinsic job satisfaction isdifferent from Degree qualification teacher’s.Further, Degree level teachers were more jobsatisfied than the G.C.E. (A/L) and Diplomalevel teachers.

Correlation among Intrinsic Job Satisfaction

Variables

Correlation among intrinsic job satisfactionvariables are given in table 4.6. According to thecorrelation probability values there is a positivecorrelation among intrinsic job satisfactionvariables at 1% significance level. This isconfirmed that intrinsic job satisfaction variablesmeasure the teacher’s job satisfaction.

40


Table 4.5: Correlation among intrinsic job satisfaction variables

Ability

Utilization

Achievement Advancement Creativity

Achievement 0.707

(0.000)

Advancement 0.496

(0.000)

0.628

(0.000)

Creativity 0.655

(0.000)

0.759

(0.000)

0.689

(0.000)

Moral Values 0.517

(0.000)

0.501

(0.000)

0.365

(0.000)

0.486

(0.000)

Table 4.5: Multiple Comparisons results for

Variable Educational Qualification

Conclusions

This study examines the personal characteristicsdemographic factors of teachers how related withtheir intrinsic job satisfaction. Results of thestudy indicated that gender does not havesignificant impact on intrinsic satisfaction. Themale & female teachers have same level ofintrinsic satisfaction and civil status does nothave significant impact on intrinsic satisfaction.The unmarried and married teachers have samelevel of intrinsic satisfaction. But, ethnicity hasthe significant impact on intrinsic jobsatisfaction. That is, Tamil teachers have moreintrinsic satisfaction than Muslim teachers. Eachcategories of age have same level of intrinsic jobsatisfaction and each category of years ofexperiences also have same level of intrinsic jobsatisfaction. But, the teacher’s educationalqualification categories have different levels ofintrinsic satisfaction.

References

Andre, B. (1996), Teacher motivation and JobSatisfaction: A study employing the experiencesampling method. Journal of UndergraduateScience, Vol.3, pp. 147-154.

Chambers, J. (1999). The Job Satisfaction ofManagerial and executive Women: Revisitingassumptions. Journal of Education forBusiness, Vol. 75, No. 2, pp. 69-74.

Goetz, K. (2012), The impact of intrinsic andextrinsic factors on the job satisfaction ondentists. John Wiley & sons A/S.

Hair, J.F., Anderson, R.E., Tatham, R.L. and Black,W.C. (2008), Multivariate Data Analysis, 6thEdition, Low Price Edition, Pearson Education.

Herzberg, F. (1959), The motivation to work. NewYork: John Wiley & sons Inc.

Jahufer, A. (2005), Growth Performance of ShoreaSeedlings at Kalutara District in Sri Lanka.Journal of Management. Vol. 3, No. 1, pp. 58-62.

Kalleberg, A.L. (1977), Work values and jobrewards theory of job satisfaction. AmericanSociological Review, Vol. 42, pp. 124–143.

Koustelious, A.D. (2001), Personal Characteristicsand Job Satisfaction of Greek teachers. TheInternational Journal of EducationalManagement, Vol. 15, No. 7, pp. 354-358.

Rajkatoch, O.M. (2012), Job Satisfaction amongcollege teachers: A study on Governmentcolleges in Jammu (J&K). Asian Journal ofResearch in Social Science & Humanities, Vol. 2, isssue. 4, pp. 164-180.

Sharma, R.D. and Jeevan, J. (2006), JobSatisfaction among school teachers, IIMBManagement Review, pp. 349-363.

Sitizawaiah Md.dawal and Zahari T. (2006), Theeffect of job and environmental factors on jobsatisfaction in automotive industries.International journal of occupational safety andErgonomics, Vol. 12, No. 3, pp. 267-280.

41



42

ABSTRACT

Predictability of asset returns in share markethas been an immense interest over the pastdecades and Statistical Modeling has beenplaying a vital role in it. This paper reviewsstatistical modeling in Technical Analysis offinancial markets. Linear and non linearregression models, Vector Auto Regression (VAR)Models and Spectral analysis found tested onshare return and trading volume. Some commonweaknesses were identified in reviewed articles.Authors have not reported the results ofmodeling assumptions; independence, normalityand homoscedasticity of errors. Modelverification criteria and results also not reported.Hence findings of their studies were not reliable.Majority of studies were focused on developedmarkets and very few attempts on emergingmarkets. Only two studies were found in SriLankan context and their results werecontradictory. It is recommended to test GARCH/ARCH models and Spectral Analysis in SriLankan context.

Key Words: Statistical Modeling, TechnicalAnalysis,

Introduction

Predictability of asset returns in share market hasbeen an immense interest over the past decades;as such Statistical Modeling has been playing avital role in financial markets. The methods usedto analyze securities and make investmentdecisions fall into two broad categories. Theywere Fundamental analysis and Technicalanalysis. Fundamental analysis involves

analyzing the economic factors of a companywhile Technical analysis interested in the pricemovements and trading volume in the market.

Fundamental Analysis based model of assetpricing; which is known as Capital Asset PricingModel (CAPM) has been subjected to extensiveempirical testing in past few decades and showedconsiderable evidence that not all the marketstake the behavior of CAPM. Nimal (1997),Samarakoon (1997) and Konarasinghe &Abeynayake (2014) showed that CAPM does notvalid for Sri Lankan share market. But SriLankan stock market still depends on CAPM forforecasting returns of listed companies ofColombo Stock Exchange (CSE).

Technical analysis based studies were also verypopular all over the world and has been tested onlarge number of stock markets. But most of thestudies were applied researches. It means most ofthe researchers have attempted to apply existingmathematical / statistical techniques inforecasting returns. They have not tried toimprove the existing models or find newknowledge in forecasting. Also technical analysisbased studies were very limited in Sri Lankancontext.

In order to improve existing methods or find newmethods for forecasting returns, it is essential tounderstand the existing methods and models. Assuch current study was focused on mathematicalperspective of previous research in technicalanalysis in financial markets. Objectives of thestudy were; understanding the variousmathematical / statistical models used intechnical analysis of financial markets andcritiques some of the previous studies from themathematical point of view.

REVIEW OF STATISTICAL MODELING IN TECHNICAL

ANALYSIS OF FINANCIAL MARKETS

WGS. Konarasinghe

Postgraduate Institute of Agriculture, University of Peradeniya, Sri [email protected]

43

Review of Statistical Modeling in

Technical Analysis of Financial Markets

Significance of the study

Share trading is an important part of theeconomy of a country. Whenever a companywants to raise funds for further expansion orsettling up a new business venture, instead oftaking loans it can issue shares of the company.On the other hand an investor can get the partownership of the company through buyingshares. This gives him/ her vote at annualshareholder meetings, and a right to a share offuture profits. Investors have the ability toquickly and easily sell securities. This is anattractive feature of investing in stocks,compared to other less liquid investments such asreal estate. In a stock trading system, forecastingis the most important activity that helps to judgethe market risk and grab scarce opportunities. Assuch predictability of asset returns in sharemarket has given an immense interest over thepast decades.

Literature shows considerable evidence thatCAPM is unable to explain market returns ofmany share markets of the world, including SriLankan share market. Technical analysisapproach also has been widely used and wassuccessful in many developed markets, but thesame was not true for emerging markets.Technical analysis based studies were verylimited in Sri Lankan context. Konarasinghe &Pathirawasam (2013) and Konarasinghe &Abeynayake (2014) have done technical analysisbased studies for Sri Lankan share market, butwere unable to find suitable technique forforecasting. As such there exists a knowledgegap in forecasting share returns in Sri Lankanshare market as well as some other markets. Thisstudy will pave the path for finding newknowledge in forecasting returns and help to fillthe knowledge gap.

Methodology

Scientific forecasting in any field of study isbased on mathematical modeling. Amathematical model is a simplification of a realworld situation into an equation or a set of

equations. Process of designing a mathematicalmodel is split into several stages. They are; a realworld problem is observed, a mathematicalmodel is devised, real world experimental data iscollected, real world expected behavior ispredicted by mathematical model, predicted andobserved outcomes are compared and themathematical model is refined (if necessary).

Mathematical models have many classifications;“Deterministic models Vs Stochastic models” isone of them. A deterministic model is one inwhich every set of variable states is uniquelydetermined by parameters in the model and bysets of previous states of these variables.Deterministic models are not associated with anyrandomness, therefore less realistic. A modelwhich randomness is present and variable statesare described by associated probabilitydistributions is called stochastic model. Ingeneral stochastic models are known as statisticalmodels.

Scientific forecasting in share markets has ahistory going back to 1950’s. Study of Osborne(1959) was the first recorded study in technicalanalysis of financial markets. Followed byOsborne (1959), large number of studies wasdone on price / return- volume relationship usingvarious statistical techniques. This studyreviewed number of research articles publishedbetween year 1959 and 2014. From themathematical point of view, those studies werecategorized in to several parts as;

i. Studies based on Fourier analysis.

ii. Studies based on Regression analysis.

iii. Studies based on Auto Regression models.

Model fitting procedure and model validationprocedure of those studies were considered incritique.

Findings

Forecasting stock returns by Technical analysisgoes back to 1950’s, findings of Osborne (1959).Osborne’s study was based on Brownian motion

44


which is known as a particle theory tooBrownian motion, found by Biologist RobertBrown in 1827 is among the simplest of thecontinuous-time processes, and it is a limit ofboth simpler and more complicated stochasticprocesses. Osborne (1959) showed thatlogarithms of common stock price changes alsohave a probability distribution similar to aparticle in Brownian motion. According to him,

if where P (t+δt) and P0 (t)are the price of the same random choice stock atrandom times (t+δt) and t, then the steady statedistribution function of Y is,

He also showed that the expected value of shareprice of a common stock (P) increases withincreasing time interval δt, at a rate of 3% to 5%per year and the variance of P is increasing whiledepend on number of transactions. Osborne hastried to address the price- volume relationship byassuming number of transactions in a Uniformdistribution, but was unable to address the issue.

Followed by Osborne (1959), large number ofstudies was done on price / return- volumerelationship using various statistical techniques.Accordingly, this article is organized as;

i. Review of studies based on Fourier analysis.

ii. Review of studies based on Regressionanalysis.

iii. Review of studies based on Auto Regressionmodels.

Review of Studies based on Fourier

Analysis

TFourier analysis, also known as SpectralAnalysis was originated in the field of electricalengineering. Spectral analysis is a frequencydomain type analysis. Fourier transformation wasthe first transformation between time domain(time series) and frequency domain series.

Granger and Morgenstern (1963) was the firstapplication of Spectral analysis for findingreturn-volume relationship in a share market.

Authors have used weekly data of New Yorkstock market for the period 1939-1961. FirstGranger and Morgenstern (1963) have tested theperiodic function;

Where Rt is the return on day t and Vt is thetrading volume on day t. They found that thisperiodic function is not suitable and then theyhave used the Fourier transformation on it.However they could not find any correlationbetween returns and trading volume.

It was difficult to find any other application ofSpectral Analysis in Technical Analysis. It maybe due to the complicated nature of thetechniques involved. A sound knowledge inMathematics; Trigonometry, Complex numbers,Calculus etc. are essential in understanding andapplication of those transformations.

Review of Studies based on Regression

Analysis

Regression analysis investigates and models therelationship between a response variable and oneor more predictors. Regression models can becategorized as; simple regression models,multiple regression models and logisticregression models. These can be either linearmodels or non-linear models. Simple regressionand multiple regression models the relationshipbetween numerical variables while logisticregression model the relationship betweencategorical response variable and numerical orcategorical predictors. Ordinary Least Squaremethod is used in parameter estimation ofsimple/ multiple regressions and maximumlikelihood procedure is used in logisticregression.

Price- volume relationships were tested onsimple regression and multiple regressionmodels;

45



Where, (3) is the simple linear regression modeland (4) is the multiple regression model. Xi’s arethe predictor variables and ε is the random error.In regression analysis it is mandatory to testseveral assumptions about errors. They are;independence of errors (errors are not seriallycorrelated), normality of errors andhomoscedasticity (constant variance) of errors. Ifany of these assumptions is violated then theforecasts and economic insights yielded by aregression model may be inefficient ormisleading.

Study of Crouch (1970) has tested Regressionmodels on daily share price changes and tradingvolume data. His study based on New York stockexchange, data collection period was sevenmonths, from December 1966 to March 2007.Results of the Crouch (1970) have givenevidence for positive linear relationship betweenabsolute price change and trading volume. ButR2 of the models were below 50% for daily data,therefore author had used hourly share price dataand trading volume to improve the model. Thisis clearly a disadvantage of his method, becausestock market forecasting is practically not usefulon hourly basis. Further his data collectionperiod, which is seven months also not sufficient.It is mandatory test model assumptions;normality of residuals, serial auto correlation ofresidual and homoscedasticity of residuals, butauthor has not reported the results ofcorresponding tests. Also it is essential to domodel verification in model fitting procedure, butauthor has not done model verification too.

Clark (1973) has applied subordinate stochasticprocess for speculative price changes. In thestudy he has tested following linear and non-linear models;

Where share price change and V is tradingvolume. His study evidenced for relationshipbetween share price change and trading volume.However Clark (1973) has not performed testson residuals, not done model verification. Assuch validity of the fitted models is doubtful. Atthe time of Clark (1973), most of the academicsand economists believed that share price changesand share returns are normally distributed. ButClark (1973) found distribution of returns andtrading volumes follow Log- Normal distribution.

Study of Timothy (1994) is based on daily AllOrdinaries Index (AOI) values and tradingvolume statistics of Australian stock market fromApril 1989 to December 1993. He has testedlinear and non linear regression models betweentrading volume and magnitude of return;

Where, Rt: return on day t, Vt: trading volume ofday t, Dt = 1 if Rt < 0, and Dt = 0 if Rt ≥0. Hisfindings support the relationship between pricechange and trading volume, irrespective of thedirection of the price change. Also Timothy(1994) tested trading volume in the context ofconditional volatility using Generalize AutoRegressive Conditional Heteroscedasticity(GARCH) framework and showed that GARCHmodel is suitable for volatility explanations.

GARCH model is an improvement of AutoRegressive Conditional Heteroscedasticity(ARCH) models of Engle (1982). Traditionaltime series models assume a constant one-periodforecast variance. Engle (1982) generalize thisimplausible assumption, introducing a new classof stochastic processes called Auto-RegressiveConditional Heteroscedasticity (ARCH)processes. ARCH is a mean zero, seriallyuncorrelated processes with non constantvariances conditional on the past, but constantunconditional variances. For such processes, therecent past gives information about the one-period forecast variance.

46


Review of Studies Based on Vector

Auto Regression (VAR) Models

Vector Auto Regression models were introducedby Sims (1980), as a technique that could beused to characterize the joint behavior ofcollection of variables. The structure of VARmodels is that each variable is a linear functionof past lags of itself and past lags of the othervariables. VAR models have been used in manyfields including financial management.

Timothy (1992) used VAR models on sharereturns and trading volumes. Weekly data ofNASDAQ stock market (an American stockexchange.) from 1972 to 1986 used in modeltesting. In the study, Timothy (1992) testedfollowing univariate causal models;

And following multivariate causal models;

Where Rt is the return of week t, Rt-i is the returnof i lag behind, Vt is the trading volume of weekt, Vt-j is the trading volume of j lag behind and Dt

is the dummy variable. He could find noevidence for multivariate causal relationshipbetween returns and trading volume but foundevidence for model type (11), univariatecausality of returns. Author has been testedregression coefficients, but not tested modelingassumptions and not validated the selectedmodel.

Saatcioglu and Starks (1998) have examined thestock price-volume relation in a set of LatinAmerican emerging markets. They havecollected data from six emerging Latin American

stock markets with at least $5 billion in marketcapitalization; they are Argentina, Brazil, Chile,Colombia, Mexico, and Venezuela. In theirempirical tests, they have employed the monthlyvalue-weighted total return index in both U.S.dollars and local currency for all six marketsfrom January 1986 to April 1995. They havetested VAR models;

Where the dependent variable (Vt) is volumemeasured by monthly turnover, the percentage ofmarket capitalization traded in a given month,and the independent variable is the naturallogarithm of the price relative (or its absolutevalue) for a given month. They found evidencefor return- volume relationship for four of the sixmarkets, but not for all. They also have notperformed tests for residuals and not done modelvalidation.

Chordia and Swaminathan (2000) formed set ofportfolios based on USA stock databases in orderto test trading volume and cross auto-correlationsin stock returns. Models tested in the study were;

Where, RA,t: return on the lowest trading volumeportfolio of A on day t, RB,t: return on thehighest trading volume portfolio B on day t, R0,t:return of zero net investment portfolio on day t.Data collection period was 1963 to 1996. Dailyand weekly equally weighted portfolio returnswere modeled with corresponding trade volumes.Authors have concluded that trade volume is asignificant determinant of the cross- autocorrelation patterns (lead –lag patterns) in stock

47



returns. But R2 of all the tested models were low.They also have not performed model verificationsand test for errors. As such their selected modelswould not have been suitable for forecasting.

Study of Wen-Hsiu, Hsinan, and Chwan-Yi(2004) was similar to the study of Chordia andSwaminathan (2000). Wen-Hsiu et. al.(2004),have used data from the Taiwan stock marketfrom January 1991 through December 2002.Results of their study were different and they didnot find a causal relationship between returnsand trading volume.

Guillrmo, Roni, Gideon and Jiang (2002) havestudied the dynamic relation between return andvolume of individual stocks listed on New YorkStock Exchange and American Stock Exchange.They have used daily data from 1993 to 1998and tested the auto regression model withinteractions;

Where Ri,t: return of ith company on day t, Vi,t :trading volume of ith company on day t. Resultsof the study supported the return –volumerelationship with interactions.

Gong-Meng, Michael and Oliver (2001) studiedthe dynamic relation between stock returns,trading volume, and volatility based on dailymarket price index and trading volume seriesfrom 1973 to 2000 for nine largest stockexchanges New York, Tokyo, London, Paris,Toronto, Milan, Zurich, Amsterdam, and HongKong. They have tested following bivariateauto regressions;

Where Rt: return on day t, Vt : trading volumeof day t,

According to their results, returns cause volumeand volume causes returns for some countries,

but not for all. Their findings suggest that morecan be learned about the stock market throughstudying the joint dynamics of stock prices andtrading volume.

Jianping, Olesya and Lubomir (2002) examinethe dynamic relation between return and volumeof individual stocks in Russia and otheremerging markets. Their study concentrates on28 large Russian stocks, which constitute about93% of the market capitalization of allcompanies traded on the Russian Trading System(RTS). Daily closing prices daily trading volumefrom 1995 to 2001 used to test the followingauto regression models with interactions;

Where Ri,t+1: return of ith company on day t+1,Rm,t+1: total market return on day t+1

Ri,t: return of ith company on day t, Vi,t :trading volume of ith company on day t

Vm,t : trading volume of the market onday t, εi,t+1, : error on day t+1

They have found strong evidence of returncontinuation following high volume days,suggesting the presence of private informationtrading in emerging markets.

Ciner (2003) attempted to find the linkagebetween trading volume and price of small-capitalization firms in the US and France. Thedata set consists of daily closing price values andaggregate trading volume for the S&P 600 andthe NM stock indices from 1995 to 2002. Theyhave tested the Vector Auto Regressive modelsincluding a dummy variable Di to account forthe day of the week and month of the yeareffects in stock returns;

Ciner (2003) also confirmed the return- volumerelationship for both US and France stockmarkets.

48


Xiangmei, Nicolaas and Yanrui (2003) examinethe relation between trading volume and stockreturns for two Chinese A-share markets and tenindividual stocks in the energy sector. The dataset comprises daily data on Shanghai A andShenzhen A share price indexes and volume(turnover) as well as prices and volume data forfrom 1997 to 2002 and they have tested linearregression models and auto regression modelsbetween returns and trading volume. They alsohave found strong evidence for causalrelationship between returns and trading volume.

Kamath (2007)’s empirical investigationexamines the causal relations between daily pricechanges and trading volume changes on theNascent stock exchange of Istanbul, Turkey. Thestudy has utilized the daily data of the IstanbulStock Exchange from 2003 to 2006 in order totest the causality between daily index returns anddaily volume. The long held view that risingmarkets tend to be accompanied by risingvolume and declining markets tend to beaccompanied by falling volume is robustlysupported by the evidence uncovered for theIstanbul Stock Exchange. Findings of this studyalso support the notion that it takes tradingvolume to make the market index move.

Malabika, Srinivasan and Devanadhen (2008)have examined the empirical relationshipbetween stock price changes and trading volumefor selected Asia-Pacific Stock Market. The dataset has comprised of seven national stockmarkets for the period 2004 to 2008. Results ofthe study have evidenced for significantrelationship between trading volume and theabsolute value of price changes for most of theselected markets, but not for all.

Sarika and Balwinder (2009), has examined theempirical relationship between return, volumeand volatility dynamics of stock market by usingdaily data of the Sensitive Index (SENSEX)during the period from October 1996 to March2006. The empirical analysis has providedevidence for causal relationship between volumeand return.

The study of Naliniprava (2011) has investigatedthe dynamic relationship between stock returnand trading volume of Indian stock Market andevidenced for bi-directional causality betweentrading volume and stock return volatility.

Habib (2011) has investigated the joint dynamicsof stock returns and trading volume in a smallemerging financial market, i.e., the EgyptianSecurities Exchange (ESE). His analysissuggested that there is no relation betweenvolume and stock returns.

Ong Sheue and Ho Chong (2011) have examinedthe short-run linear and nonlinear Grangercausality between stock return and tradingvolume in Malaysia and Singapore cases basedon the Vector Autoregression (VAR) model andTaylor expansion of the nonlinear model. Theyhave found evidence for significant bidirectionalnonlinear causality between returns and tradingvolume in Malaysia case while unidirectionalnonlinear causality from trading volume to stockreturn in Singapore.

Marwan (2012) also examined the causalrelationship between return and trading volumein the Palestine Exchange using weekly tradingvolume and returns over the period from October2000 to August 2010. They have found that therelationship preserves after taking heteros-kedasticity into account. Moreover, the results ofcausality tests show that there is bidirectionalcausality between returns and trading volume,regardless of the measures of trading volume.

Konarasinghe & Pathirawasam (2013) havetested causal relationship between returns andtrading volumes in Sri Lankan share market.Their study was somewhat similar to Chordiaand Swaminathan (2000). Monthly total marketreturns and trading volumes, monthly sectorreturns and trading volumes from 2005 to 2011were used for model testing. Results ofmultivariate tests revealed that there is no causalrelationship between market returns and tradingvolumes. Further they have found that stockreturns are auto-correlated and stationary whiletrading volumes are auto-correlated but notstationary.

49




Predictability of share returns in secondarymarkets is an immense important to the investorsas well as the regulators of the market. As suchStatistical Modeling has been playing a vital rolein financial markets over decades. One of themain strands of forecasting returns is TechnicalAnalysis, which has been in practice from1950’s. This paper reviewd more than twentyfive articles based on return- volumerelationship on view of identifying variousstatistical models used in them.

Regression models and Vector Auto Regression(VAR) models were the mainly tested on return/price –volume relationship all over the world and most of the studies have given evidence forsuccess in forecasting returns. But somecommon weaknesses were identified in them.Most of the authors have not reported the resultsof modeling assumptions and modelverifications. For examples; test results ofindependence of errors, normality of errors,homoscedasticity of errors etc were not reported.Model verification criteria and results also notreported. Hence findings of their studies cannotbe considered as reliable.

Spectral analysis has been successfully appliedin fields, Engineering, Medicine, Physics andmany others. But, except Granger andMorgenstem (1963), no other attempt was foundin applying Spectral analysis for forecastingshare returns.

According to literature, most of the studies werefocused on developed markets and very fewattempts were on emerging markets. Only twostudies could find in Sri Lankan context and theirresults were contradictory.

It is recommended to test GARCH/ ARCHmodels and Spectral Analysis in Sri Lankanshare market as well as other emergingmarkets.

References

Ciner, C., (2003), Dynamic linkages betweentrading volume and price movements:Evidence for small firm stocks. Journal ofEntrepreneurial Finance. Vol. 8, No. 1, pp.1-15.

Chordia, T., Swaminathan, B., (2000), TradingVolume and Cross Auto-correlations in StockReturns. Journal of Finance, Volume. LV (2).

Clark, P.K.A., (1973), Subordinated StochasticProcess Model with Finite Variance forSpeculative Prices. Econometrica. Vol. 41, No. 1, pp. 135-155.

Crouch, R.L., (1970), The Volume of Transactionsand Price Changes on the New York StockExchange. Financial Analyst Journal.

Engle, R.F., (1982), Autoregressive ConditionalHeteroscedasticity with Estimates of theVariance of United Kingdom Inflation.Econometrica, Vol. 50, No. 4, pp. 987-1007.

Gong- Meng, C., F. Michael, F., Oliver, M., R.,(2001), The Dynamic Relation Between StocksReturns, Trading Volume, and Volatility. TheFinancial Review. Vol. 38, pp. 153-174.

Granger, C.W.J, Morgenstem, O., (1963), SpectralAnalysis of New York Stock Market Prices.Kyklos. Vol. 16, pp. 1-27.

Guillrmo, L., Roni, M., Gideon, S., and Jiang, W.,(2002), Dynamic Volume- Return of IndividualStocks. The Society of financial studies. Vol. 15, No. 4, pp. 1005-1047.

Habib, N.M.,(2011), Trade Volume and Returns InEmerging Stock Markets An Empirical Study:The Egyptian Market. International Journal ofHumanities and Social Science. Vol. 1, No.19

Jianping, M., Olesya V.G., Lubomir.P. L., (2002),Measuring Private Information Trading inEmerging Markets. New York University.

Kamath, R.R, (2007), Investigating CausalRelations between Price Changes and Trading

50


Volume Changes in the Turkish Market.ASBBS E- Journal. 3 (1).

Konarasinghe, W.G.S. & Pathirawasam, C. (2013),Modeling Stock Returns and Trading Volumeof Colombo Stock Exchange. Sri LankanJournal of Management. Vol. 18, Nos. 3 & 4.

Konarasinghe, W.G.S. & Abeynayake, N.R.(2014),Modeling Stock Returns of IndividualCompanies of Colombo Stock Exchange.Conference Proceedings of the 1st InternationalForum for Mathematical Modeling 2014,Department of Mathematics, University ofColombo, Sri Lanka, 111.

Malabika, D., Srinivasan K.K., Devanadhen, K.,(2008), The Empirical Relationship betweenStocks Returns, Trading Volume and Volatility:Evidence from Select Asia-Pacific StockMarket. European Journal of Economics,Finance and Administrative Sciences. ISSN1450-2275(12).

Marwan, D., (2012), Testing the Contemporaneousand Causal Relationship between TradingVolume and Return in the Palestine Exchange.International Journal of Economics andFinance.4 (4).

Nimal, P.D., (1997), Relationship between StockReturns and Selected Fundamental Variables;Evidence from Sri Lanka. Sri Lankan Journalof Management.

Naliniprava.T., (2011), The Relation between PriceChanges and Trading Volume: A Study inIndian Stock Market. Interdisciplinary Journalof Research in Business. Vol.1 No.7, pp.81-95.

Osborne, M.F.M, (1958), Brownian motion in theStock Market. U.S. Naval ResearchLaboratory, Washington.

Ong Sheue, L, Ho Chong, M., (2011), Testing ForLinear and Nonlinear Granger Causality In TheStock Return And Stock Trading VolumeRelation: Malaysia And Singapore Cases.Labuan Bullllettiin of International Business &Finance.9.

Samarakoon, M.P., (1997), The Cross Section ofExpected Stock Returns of Sri Lanka. SriLankan Journal of Management.

Saatcioglu, K., Starks L.T., (1998), The StockPrice-Volume Relationship in Emerging StockMarkets: The Case of Latin America.International Journal of Forecasting. Vol.14,No. 2, pp. 215-225.

Sarika, M., Balwinder, S., (2009), The EmpiricalInvestigation of Relationship between Return,Volume and Volatility Dynamics in IndianStock Market. Eurasian Journal of Businessand Economics 2009. Vol. 2, No. 4, pp. 113-137, 2009.

Sims, C. (1980), “Macroeconomics and Reality”,Econometrica Vol. 48, pp.1-48.

Timothy, S.M. (1992), Portfolio returnsautocorrelation. Journal of FinancialEconomics, Vol. 34, pp. 307-344

Timothy, J.B. (1994), The Empirical Relationshipbetween Trading Volume, Returns andVolatility. Asia-Pacific Finance AssociationConference.

Wen-Hsiu, K, Hsinan, H, Chwan-Yi, C., (2004),Trading Volume and Cross-Autocorrelations ofStock Returns in Emerging Markets: Evidencefrom the Taiwan Stock Market. Review ofPacific Basin Financial Markets and Policies.Vol. 7, No. 4, pp. 509–524 .

Xiangmei, F, Nicolaas, G, Yanrui, W., (2003), TheStock Return-volume Relation and PolicyEffects: The Case of the Chinese EnergySector. Proceedings of the 15th AnnualConference of the Association for ChineseEconomics Studies Australia (ACESA).

51

ABSTRACT

Among the working-age population, one of themost damaging individual experiences isunemployment. Mostly it is a serious issue inthird world countries like Srilanka. The main aimof this study is to identifies the nature of theunemployment and its impact on individual’swell-being among the working age group withrespect to their vocational activities such asincome, savings housing ,health, education andother social welfare in Jaffna District inSrilanka. The primary data were collected from100 unemployed households in Jaffna District byusing structured questionnaire. The techniquessuch as descriptive statistics such as descriptiveand regression model were used in this study.The results of the regression model indicate thatthe unemployment negatively affecting in alleconomic and social conditions of the people inJaffna District. Finally the overall results revealthat among unemployed people, non-pecuniaryfactor such as job prospects, health and socialrelations – show significant effects on individualwell-being, along with their income. Policymakers have to take necessary actionsincorporating strategic approaches which canreduce the unemployment and improve the well-being of the people in Jaffna District, Srilanka.

Key Words: Unemployment, well-being,education, income, saving, health, education

Introduction

Unemployment is a severe problems prevailingin developing as well as developed countries.These problems have a serious effect not only onthe living standards of people and socio –economic status of the nation but also increasingthe magnitude of corruption effecting the Theseproblems have a serious effect not only on theliving standards of people and the socio-economic status of a nation, but also elevate themagnitude of corruption, poverty, crime andsuicidal rates in a society(Neeleman and Lewis1999; Asghar 2002; Blakely et al. 2003).Unemployment has negative effects on thephysical well-being of the suffering peoples.

Individual well-being (or happiness) depends onmany things, including income, labour marketstatus, job characteristics, health, leisure, family,social relationships, security, liberty, moralvalues and many others. Although unemployedworkers usually suffer a reduction of income, itsextent varies depending on other income sources,such as savings and income-generating assetholdings, unemployment insurance and privatetransfers. Non-pecuniary consequences such asthe loss of identity and self-esteem, stress anddepression also depend on the individual, familyand social circumstances surroundingunemployed workers. On the other hand,unemployed workers gain time for activities suchas leisure, training, physical exercise anddomestic activities (Ahn et al., 2004). Therefore,in evaluating the effect of unemployment on

THE EFFECT OF UNEMPLOYMENT ON

SOCIO ECONOMIC STATUS OF THE PEOPLE IN

JAFFNA DISTRICT, SRI LANKA

Paulina Mary Godwin Phillip

Senior Lecturer, Dept. of Economics & Management, Faculty of Business Studies, Vavuniya Campus

Thayaparan Aruppillai

Senior Lecturer, Dept. of Economics & Management, Faculty of Business Studies, Vavuniya Campus

52


individual well-being, we should consider allthese relevant factors as well.

Sri Lankan Unemployment rate is increasingtrend in 2012. Sri Lankan basic quality characterrelated to the unemployment, these peoplemainly focus the government job rather than theother job; they waiting for getting governmentjob that also lead the Unemployment.unemployment amount is increased year to year.But, job opportunities aren’t created to equalincrease ratio of unemployment. This problem isfound in Jaffna peninsula as well.Unemployment is significance in any macro-economic decision making. These variable issubject of social and economic life of everycountry. Thus, Unemployment is termed ascontinuous and unpleasant situation that describethe endemic nature of a country’s economy. Thisanalysis or research is employed for analyzingseveral economical, social, cultural effectsamong unemployment target groups in JaffnaDistrict.

Objectives of the Study

l To identify the nature of unemployment inJaffna district.

l To identify the impact of unemployment onindividual’s well – being

Literature Review

The International Labour Organization (ILO)defines the unemployed as numbers of theeconomically active population who are withoutwork but available for and seeking work,including people who have lost their jobs andthose who have voluntarily left work (WorldBank, 1998:63). Examples include housewives,full time students, invalids, those below the legalage for work, old and retired persons. Theunemployment rate is expressed as a percentageof the total number of persons available foremployment at any time.

Types of Unemployment

Structural Unemployment – Occurs when thereis a change in the structure of an industry or theeconomic activities of the country. As aneconomy develops over time the type ofindustries may well change. This may be becausepeople's tastes have changed or it may bebecause technology has moved on and theproduct or service is no longer in demand.

Frictional Unemployment – This type ofunemployment is caused by industrial friction,such as, immobility of labor, ignorance of jobopportunities, shortage of raw materials andbreakdown of machinery, etc. Jobs may exist, yetthe workers may be unable to fill them eitherbecause they do not possess the necessary skill,or because they are not aware of the existence ofsuch jobs. They may remain unemployed onaccount of the shortage of raw materials, ormechanical defects in the working of plants. Onaverage it will take an individual a reasonableperiod of time for him or her to search for theright job.

Seasonal Unemployment - This is due toseasonal variations in the activities of particularindustries caused by climatic changes, changes infashions or by the inherent nature of suchindustries. The rain coat factories are closeddown in dry season throwing the workers out oftheir jobs because there is no demand for raincoat during dry season. Likewise, the sugarindustry is seasonal in the sense that the crushingof sugar-cane is done only in a particular season.Such seasonal industries are bound to give rise toseasonal unemployment.

Cyclical Unemployment - This type ofunemployment (also known as Keynesianunemployment or the demand deficientunemployment) is due to the operation of thebusiness cycle. This arises at a time when theaggregate effective demand of the communitybecomes deficient in relation to the productivecapacity of the country. In other words, when theaggregate demand falls below the full

53

The Effect of Unemployment on Socio-Economic

Status of the People in Jaffna District, Sri Lanka

employment level, it is not sufficient to purchasethe full employment level of output.

Disguised Unemployment - This refers to themass unemployment and underemploymentwhich prevail in the agricultural sector of anunderdeveloped and overpopulated country. Forexample, if there are four persons trying tocultivate an area of land that could be cultivatedas well by three persons, then only three of thesepersons are really fully employed and theremaining fourth person represents disguisedunemployment. The people in underdevelopedcountries are outwardly employed but actuallythey are unemployed, the reason being thatagricultural production would suffer no reductionif a certain number of them are actuallywithdrawn from agriculture.

A review of the literature reveals views regardingimpact of unemployment on socioeconomicstatus in the country. Meaning of unemployment“unemployment is defined as a state of affairswhen in a country there are a large number ofable bodied person of working age who arewilling to work but cannot fine work at thecurrent wage levels. People who are either unfitfor work for physical or mental reason, or don’twant to work are excluded from the category ofthe unemployed. There are three types ofunemployment frictional unemployment,structural unemployment and cyclicalunemployment. Employment is the major sourceof income for a great majority of the people, afall in employment signifies a fall in theirincome also.(Ahuja.H.L, 2007)

Lawanson (2007) opined that economicrecession has significant negative implication forthe utilization of country’s human resources,leading to high level of unemployment.According to him, this problem has aggravatedin the nation to the extent that many universitygraduates could not secure jobs, let alone schoolleavers. Furthermore, Lawanson (2007) said theproblem is twofold showing both demand andsupply side. On the demand side not only arethere inadequate jobs for youths. But also the

increasing decline in quality of education andtraining, thus making many youths unemployed.On the supply side, the inability of thegovernment to adequately finance the nation’seducational enterprise has led to deterioratinginfrastructural facilities and discouragingpersonal emoluments for teachers, it wasdiscovered that despite various governmentpolicies and programmes aimed at reducingunemployment among youths and adults, theproblem of unemployment remains unabated. Onthis note, Lawanson (2007) concluded thatUnemployment has been found to reducenational wealth, increase in crime waves andsocial political violence can also be attributed tothe high level of unemployment especiallyamong youths in Nigeria.

F. Nazir, M. A. Cheema, M.I. Zafar, and Z.

Batool (2009) identified the unemploymentnegatively affecting the socio-economic status offamily in Urban Faisalabad, Pakistan by usingdescriptive analysis. He has also found that theunemployment leads to poor mental health andincreases the magnitude of corruption, drugaddiction, crimes and suicide in a society. A largepercentage of the respondents of age limit 24-30+ believed that the lower rate of education isresponsible for the present situation ofjoblessness.

Syed Haider Rasa Investigated that the socio-economic impacts of unemployment on citizenof North Nazimabad, Karachi with regard to thecharacteristics, magnitude and direction ofchanges in social relationships, and the attitudeof unemployed people towards the society.

Mel Bartley (1994) found in his study, tounderstand the relationship betweenunemployment and health and mortality, fourmechanisms need to be considered: the role ofrelative poverty; social isolation and loss of selfesteem; health related behavior (including thatassociated with membership of certain types of"subculture"); and the effect that a spell ofunemployment has on subsequent employmentpatterns.

54


Auwal Abubakar Muhammad; Bello Malam

Sa’idu; Obinna John Nwokobia & Kabiru

Musa Yakubu(1999) investigated implies thatunemployment significantly influence wage ratewhereas inflation is positive but has nosignificant effect on wage rate. Therefore, thereis a need for strong institutional collaboration fordealing with these triple macroeconomicvariables; unemployment, inflation and wages inNigeria.

Methodology

Conceptual model

Conceptual model can be developed as fallows.

and interview are the main primary form at datacollection tool used in this study. Data analysismade based on collected questionnaire,correlation and regression analysis. MS Excelversion 2007 and SPSS version 16 use to dataanalysis.

Questionnaire prepared by researcher to targetpeople about unemployment (workless people),income level of family, consumptionexpenditure, health, saving, malnutrition,education level, and socio economic. ThisQuestionnaire includes the personal details andsocio economic details. This questionnaireconsists of closed ended as well as open –ended

Stratified sampling method was used to the data.Primary data is the major source of datacollection method of this study. Questionnaire

question and Nominal and likert scale to measurethe impact of unemployment.

55



In addition the questionnaire was administeredamong the respondent to make tick (√) for theiroption regarding the statement in thequestionnaire. The five point likert scale is usedfor statements of the second section ranging fromstrongly disagree to strongly agree, very low tovery high. The numerical values were given forthe purpose of quantification of quantitativevariable as follows.

Hypothesis

H0 – Unemployment does not effect the socio-economic status of the people

H1 - Unemployment effects the socio-economicstatus of the peopl

Descriptive Statistic

The technique is used to find out mean medianother statistical data, through this researcher canexplain the whole data in this research.

Regression Analysis

Regression analysis is to describe the nature ofthe relationship between two variables in termsof a mathematical equation. Regression lineexplains the pattern of variation of the dependentvariable in relation to values the independentvariables. It is used for drive the line of best fit.

Yi= β0+β1Xi + Ui

Y – Socio Economic Status of the people

X – Unemployment

β0 - the line crosses the vertical y-axis orconstant term

β1 – Coefficient of Xi.

Results and Discussions

Reliability of this research questionnaire is 0.701the following shows the reliability

Descriptive Analysis

Gender Analysis

Among the total sample 39% of respondents aremale and 61% of respondents are female. 100%of respondents are Sri Lankan Tamils because ofthe area selected for study is Jaffna.

Table 3: Scale of measurement

56


Type of Unemployment

There are many type of unemployment in thewhole economy, But here only selected four typeof unemployment. According to the figure 41%of the unemployed families coming under thestructural unemployment. Remaining familiescoming under the other categories

Educational Qualification

Based on the analysis among the 100 samples,1% of the sample represent the illiteracy, 11%sample complete their primary education, 60%sample unemployed people who have completedG.C.E O/L, 19% of the sample have completedG.C.E A/L, 4.41% of sample who are presentlyfollowing Technical College studies, and the3.68% are graduated.

Age distribution shows that among the sample,33% of unemployed household were found to belying under the age category between 15-25years. 46% were found under the age categorybetween 25-35 years & 21% represents above 35 years.

Health

Out of sample population 14 people affected bycritical illness and 11 child death occurred, thisshow poor health condition. Unemployment

leads to poor health condition that affects theliving standard of people. Long termunemployment leads to mentally disappear.

Medical Expenses of theunemployed families

Income

Out of the sample 49% of the respondents’family income is between Rs 6000- Rs 12000 aswell as 36% of the respondents’ family income isbelow Rs 6000, 12% of the respondents’ familyincome is between Rs12000 - Rs18000, 3% ofthe respondents’ family income is aboveRs18000 . This low income level is mostlyattributed by unemployment. This shows thatwithout having the job they managing expensesthrough the other sources.

Monthly Expenses of theunemployed families

Among the sample 54 % of the family expensesis between Rs 6000 to Rs 12000 ,30 % of thefamily expenses between Rs 12000 to Rs18000,11% of the family expenses is above Rs18000 and 5% of the family expenses below Rs6000.

Among these unemployed families 59 % of therespondents borrow the money from other, itreflect the family credit situation. These people

Source: Survey Data

Figure 1:

Type of unemployment

Figure 2:

Medical Expenses

manage their needs through borrow the money.18% of respondents mortgage of gold forsatisfy their basic needs and 23% of therespondents get the help from others,

Saving status of the unemployed

familiesunemployed families

Out of sample 29 % respondents maintained thesaving and 71% respondents not maintained anysaving. Poor saving lead to poor investment, thatreflect poor standard of living. Official Povertyline at national level for February 2013 is Rs.3656 (Source: Department of Census & Statistics- Sri Lanka) According this source in survey data36 families under the national poverty line.These 36 families critically affect by theunemployment and their standard of living alsoaffect.

Housing

According the data collection respondentsanswer the 3 type of house. 49% respondentshave been living in the hut, 39% respondentshave been living in the rented house & rest of the12% respondents have been living in the ownhouse. Hut is not satisfying residence to safetylife although 49 families live in the hut, no onenot like to live in the hut even though because oftheir family income situation they live.Unemployment influences their residence, onethe basic needs affects by the unemployment.

Electricity

Electricity is fundamental needs to the humanbeings. Out of the sample 59% of people areresiding in Jaffna with electricity facility, rest ofthe 41% of people who are residing in Jaffnawithout electricity facility.

Poverty

Among the sample 65% of the unemployedfamilies face the poverty related problem.Official Poverty line at national level forFebruary 2013 is Rs. 3656 (Source: Department

of Census & Statistics - Sri Lanka).According tothis source in survey data 36 families under thenational poverty line. These 36 families sufferedby poverty because of unemployment.

Output of regression analysis

Regression economic status on unemployment

Table 4: Out put

Y – Economic Status of the people

X – Unemployment

If unemployment is X=0, economic condition isto be 49.69units .Unemployment is increased byone unit scale, the economic status of the peoplewill be decreased by 0.304 unit scale Therefore,it can be said that there is a significant negativerelationship between unemployment andeconomic condition.

Regression Social status on Unemployment

Table 5: Out put

Y – Social Status of the people

X – Unemployment

The regression equation Y= 20.93 - 0.041Xexhibits that the relationship betweenunemployment and Social Status. If theunemployment level is zero, the social status is20.93 units. Unemployment is increased by one

57



Source: Survey Data

Source: Survey Data

58


unit, Social status of the people decreased by0.041 unit. Even though it has small effect onsocial status, but it has significant effect onsocial status of the people.

Conclusion and Recommendation

The results show that the unemployment hasnegative effect on both social and economicstatus of the people. So it leads to low standardof living of the people in Jaffna District. So thatunemployment affects the socio-economic statusof the family, leads to poor mental health andincreases the magnitude of corruption, drugaddiction, crimes and suicide in a society.

In this study mostly young males and females areaffected by unemployment. Nowadays, with thedevelopment of society, youth unemployment isbecoming a global problem, which affects notonly individuals but also society in every regionof the Jaffna district as well as country. A series ofproblem create other crisis. youth unemploymentshould be alleviated as soon as possible.

Youth unemployment is an unseen unutilizedresources. Our government spends a considerablenumber of resources on education, in order tomake the youth useful for our society. To bringthe youth a bright future, our schools andgovernment as well as the youth themselves haveresponsibilities to improve the situation.

Firstly, our schools especially universities andcolleges should present courses associated withemployer expectation. Secondly, The governmentshould take necessary to reduce theunemployment among the graduates. Thecompanies have responsibilities to offer jobopportunities to the graduates. Thirdly, the youththemselves should do their utmost to becomequalified. The policies should be adapted toprovide the opportunities to create theentrepreneurs.

Government gives the employment opportunityto youth through the government project. As well

as give necessary skill training to educatedpeople in that area.

In this study structural unemployment is high insample population at Jaffna district. ReducingStructural Unemployment necessary fordevelopment of Jaffna district for that some ideasare given below,

Policy suggestions to reduce structuralunemployment include providing governmenttraining programs to the structurallyunemployed, paying subsidies to firms thatprovide training to displaced workers, helpingthe structurally unemployed to relocate to areaswhere jobs exist, and inducing prospectiveworkers to continue or resume their education.

Unemployment substantially reduces anindividual’s satisfaction levels with his or hermain activities and finance, while it substantiallyincreases his or her satisfaction level with leisuretime. With respect to health, it has a smallnegative effect. Unemployment duration, on theother hand, shows a small negative effect onindividual well-being, suggesting theunemployment has lasting and aggravatingeffects that contradict the theory of adaptation.

The solution for unemployment is, obviously, tocreate new jobs. Usually, a healthy economicgrowth rate of 2-3% is enough to create the150,000 new jobs needed to keep unemploymentfrom rising. But our country growth rate is veryless because of that can’t create the new jobs.Our government should focus on our growthrate. When unemployment rate above 6-7% andstays there; it means the economy isn't strongenough to create sufficient new jobs withouthelp. Our country unemployment rate for theThird quarter 2012 was reported as 4.1 %(Source - Sri Lanka Labour Force Survey) butthis rate excluding the Northern Province. That'swhen the government is expected to step in andprovide solutions.

Government should tack some step to reduce theunemployment through policies and regulationsuch as monetary policy and fiscal policy.

59



Expansive monetary policy is powerful, quickand usually effective. Lower interest rates allowfamilies to borrow more cheaply to buy whatthey need; this stimulates enough demand to putthe economy back on track. Low interest ratesalso allow businesses to borrow for less, givingthem the capital to hire new workers to meetrising demand. However, when monetary policydoesn't work, then fiscal policy is usuallydemanded. This means the government musteither cut taxes or increase spending to stimulatethe economy

References

World Bank (1999); African DevelopmentIndicators 1998/99, the World Bank,Washington, D. C.

Auwal Abubakar Muhammad; Bello MalamSa’idu; Obinna John Nwokobia & KabiruMusa Yakubu (2013), Journal of EmergingTrends in Economics and ManagementSciences (JETEMS) 4(2): 181-188©Scholarlink Research Institute Journals, 2013(ISSN: 2141-7024) jetems.scholarlinkresearch.org

Ahuja H.L “Macroeconomics Theory and Policy”(2007), twel fifth revised edition, published byS. Chand and Company ltd.7361, Ram Nagar,New Delhi- 110055 220 – 223

Farhat Nazir, M. Asghar Cheema, M.Iqbal Zafar,and Zahira Batool (2009), “Socio-economicImpacts of Unemployment in UrbanFaisalabad, Pakistan”http://www.J Soc Sci,18(3): 183-188 (2009)

Mel Bartley (1994), “Unemployment and ill health:understanding the relationship” http://wwwJournal of Epidemiology and CommunityHealth 1994; 48:333-337

Syed Haider Rasa “Socio –Economic ofunemployment on citizens of NorthNazimabad,Karachi”

District Planning Secretariat. (2011), StatisticalData -2008/2011, District Secretariat Jaffna Sri Lanka.

60

ABSTRACT

A business concern can go for different levels ofthe mixtures of equity and debt or other financialfacilities with equity having the emphasis onmaximizing the firm’s value. And also it’s affectsthe liquidity and profitability of a firm. Thereforethis study focused on the impact of capitalstructure on the profitability of listedmanufacturing companies in the Colombo StockExchange in Sri Lanka. The primary objectivesof this study are to investigate the significantimpact of capital structure on profitability andfind out the significant relationship betweencapital structure and profitability of listedmanufacturing companies in Sri Lanka. In thisregard, researcher has selected a sample of 20manufacturing companies listed on the CSE inSri Lanka during the year from 2009 – 2013. Forthe purpose of this study, the secondary data wasextracted from the annual reports of samplelisted manufacturing companies. Multipleregression and correlation analysis were usedwith the SPSS -16 versions.

The results indicate that the capital structure ofthe manufacturing companies listed on ColomboStock Exchange has a significant impact onprofitability. Moreover the findings revealed thatthere is a significant negative relationshipbetween LDA and profitability while there is apositive strong relationship between TDA andprofitability. But there is an insignificantnegative relationship between SDA andprofitability of manufacturing listed companies inthe CSE in Sri Lanka.

Key Words: Capital Structure, Long-termdebt, Short-term debt, Return on Asset,Profitability.

Introduction

Capital structure theory is one of the mostimportant areas of finance. It shows the firm'sfinancial frame work and refers to the variousfinancing options of the assets by a firm. Thecapital structure is how a firm finances its overalloperations and growth by using different sourcesof funds. It is most likely referring to a firm'sdebt-to equity ratio. Debt comes in the form ofbond issues or long-term notes payable, whileequity is classified as common stock, preferredstock, retained earnings and reserved fund. Ingeneral, a firm can choose among manyalternative capitals structures. It can issue a largeamount of debt or very little debt. Usually acompany more heavily financed by debt posesgreater risk. A firm can issue number of distinctsecurities in countless combinations; however, itattempts to find the particular combination thatmaximizes its overall market value (Abor, 2005).

Firm capital structure plays a determinant role infirm profitability. It is suggested that utilizationof different levels of debt and equity in the firm’scapital structure is one such firm-specificstrategy used by managers in search forimproved performance. A business can go fordifferent levels of combination of equity anddebts or other financial facilities; that may belease financing, term financing, debentures anddirect loans from bank etc with equity capital

IMPACT OF CAPITAL STRUCTURE ON PROFITABILITY:

A STUDY OF LISTED MANUFACTURING COMPANIES

IN THE COLOMBO STOCK (SEC) EXCHANGE

IN SRI LANKA

Sithy Safeena M.G. Hassan

Department of Management, South Eastern University of Sri Lanka Sri [email protected]

(Raheman, Bushra, and Mustafa, 2007).Managers who are judicious enough to identifyand set up the appropriate mix of debt and equityare fully rewarded in the market place, because,all things being equal, this appropriate mix ofdebt and equity minimizes a firm’s cost offinancing.

An optimal capital structure is usually defined asone that will minimize a firm's cost of capital,while maximizing firm value. Researcherscontinue to analyze capital structures and try todetermine whether optimal capital structuresexist. Modigliani Miller (1958) found that in aperfect competitive market; the capital structuredoesn't have influence on the market value of thecompany, which will be settled by thecomposition of its assets. And also found thatunder the perfect capital market conditions; afirm’s value depends on its operating profitabilityrather than its capital structure.

A company can finance its operations by eitherdebt or equity or different combinations of thesetwo sources. But it is a difficult decision for thefirms to determine the proportion of the equityand the debt in the optimal capital structure tomaximize the profit and minimize the risk andthe cost of capital. Nirajini and Priya (2013)explained that capital structure decisions havesignificant impact on profitability of the firm.Exactly how firms choose the amount of debtand equity in their capital structures remains anenigma and it is not an easy task to everycompany and its managers, while there are manyempirical studies emerging all over the worldand moreover still straggling to fix a suitableproportion of the capital structure. An optimumcapital structure is a critical decision making forany organization. But the capital structuredecision is important for the need to maximizereturns to various organizational constituencies,and also this decision has on an organization’sability to deal with its competitive environment(David, 2001).

Therefore it should be made clear understandingon the impact of capital structure on the firm’s

financial performance. The main problems of thisstudy is to analysis does capital structure has aimpact on firm’s financial performances and howthe capital structure negatively or positivelycorrelate on profitability of the companies in theManufacturing sector in Sri Lanka.

Literature Review

A firm's capital structure is the composition ofstructure of its liabilities (Nirajini and Priya2013). The behavior of the capital structure ofthe firm influences by many factors such ascapital intensity, tangibility, expected growth,size, profitability, non debt tax shields, liquidity,volatility, uniqueness and industry classification(Titman and Wessels, 1998; Kajananthan andAchchuthan, 2013; Samarakoon, 1999;Sangeetha and Sivathaasan, 2013).

The theories of capital structure try to justify andexplain the capital structure from time to timewhich addressed the nature of capital structurefrom different angles. The first concept of thecapital structure was introduced by Durand(1952) but Modigliani and Miller (1958) was apioneer in opening the way for wide range ofstudies, contemporary thinking and arguments oncapital structure with their “irrelevancy theory”and rejected the Durand’s theories. Identifyingthe right proportion of debt and equity of capitalstructure has been much difficult to bringfavorable or profitable results for theorganizations.

Haugen and Senbet (1978) explained that in afirm debt capital increases probability ofbankruptcy increases. Harris and Raviv (1991)confirmed that firms with a high growth ratehave a high debt to equity ratio and also pointedout bankruptcy costs were also found to be animportant effect on capital structure. Titman(1984) demonstrates the idea of indirectbankruptcy costs and he argues that stakeholdersnot represented at the bankruptcy bargainingtable, such as customers, can suffer materialcosts resulting from the bankruptcy. Given these

61

Impact of Capital Structure on Profitability: A study of Listed Manufacturing

Companies in the Colombo Stock (SEC) Exchange in Sri Lanka.

bankruptcy costs, the operating risk of the firmwould also influence the capital structure choiceof the firm because firms which have higheroperating risk would be exposed to higherbankruptcy costs, making cost of debt financinggreater for higher risk firms. When a firmfinances a project through debt, the creditorscharge an interest rate that they believe isadequate compensation for the risk they bear.Because their claim is fixed, creditors areconcerned about the extent to which firms investin excessively risky projects. The order ofpreferences reflects the relative costs of variousfinancing options. Clearly, firms would preferinternal sources to costly external finance (Myersand Majluf, 1984).

The pecking order theory does not claim a welldefined debt target. Brealey, Myers and Allen,(2006) suggested that having equity in both endsof the pecking order is one explanation for this,which is due to the existence of both internal andexternal equity. Every firm’s cumulative need forexternal finance is therefore shown by its debtratio. And also concludes that the most profitablefirms in general do not raise debt, a findingconsistent with the pecking order theory, simplysaying that profitable firms in general are not inneed of external financing.

There are few studies related with capitalstructure and profitability in Sri Lanka(Samarakoon, 1999; Nimalathasan and Brabete,2010; Pratheepkanth, 2011; Velnampy andNiresh, 2012; and Lingesiya, 2012; Kajananthanand Nimalthasan, (2013). Those finding differin time period of studies and industries,analyzing somewhat different set of variablesand indicate different degrees of results.Nimalathasan and Brabete (2010) analyzed thecapital structure and its impact on profitability astudy of listed manufacturing companies in SriLankan. The findings revealed that Debt Equityratio is positively and strongly associated withprofitability.

Manawaduge, DeZoysa and Chandrakumara(2010) investigated about Capital structure and

its implications, empirical evidence from anemerging market in South Asia. They usedsample of 171 companies. The sample representsapproximately 74 per cent of the listedcompanies in Sri Lanka. These companiesbelonged to all industrial sectors of the CSE,excluding the bank, finance and insurance sectorover the period of 2002 to 2008. The resultshowed there is a significant negativerelationship between leverage ratios andaccounting performance measures.

Pratheepkanth (2011) conducted a study on therelationship between capital structure and firm’sperformance during the period from 2005 to2009. The sample consists of 30 business listedcompanies in Sri Lanka. The results shown thatthere is no significant relationship betweencapital structure and gross profit but there was anegative significant relationship between capitalstructure and Net profit, ROE, ROI and ROA.Velnampy and Niresh (2012) studied therelationship between Capital Structure andProfitability with a sample of 10 listed banksover the period of 2002 to 2009. The findingsrevealed that there is a significant negativerelationship between the Capital Structure andProfitability.

Nirajini and Priya (2013) examines the impact ofcapital structure and financial performance of thelisted trading companies in Sri Lanka with asample of 11 trading companies in Colombostock exchange over the period of five yearsfrom 2006 to 2010. The findings revealed thereis a significant positive relationship betweencapital structure and financial performance. Andalso capital structure has a significant impact onfinancial performance. Tharmila and Arulvel(2013) also examined the relationship betweencapital structure and financial performance of thelisted companies traded in Colombo stockexchange (CSE) using the sample of thirtycompanies during the period of 5 five years from2007 to 2011. This research results showed thatthere is an insignificant negative relationshipbetween the capital structure and firm’s financialperformance

62


Significance of the Study

The capital structure of a firm concerns the mixof debt and equity, the firm uses in its operationthis has centered on whether there is an optimalcapital structure for an individual firm orwhether the proportion of debt usage is irrelevantto the individual firm’s value. The optimalcapital structures the one that maximizes themarket value of the firm’s outstanding shares.The fact that an optimal capital structure has notbeen found is an indication of some flaw in thelogic. So, the choice of capital structure is afundamental problem for each and every firm.The trade-off theory of optimal capital structureassumes that firms balance the financialdiscipline and marginal present values of interesttax shields against the costs of financial distress.Naturally, it lies in every firm’s interest to findan optimal balance between internal and externalfinancing. Byoun and Rhim (2003) found thatdifferences between that the target debt ratio andactual debt ratio is an important aspect to takeinto consideration. According to their study firmstend to adjust their debt ratios to specific targetdebt ratios. This is consistent with the trade-offtheory.

There are number of studies have beenconducted to find out the relationship betweenthe capital structure and financial performance inSri Lanka and other countries. Some researchersfound that there is a positive significantrelationship between capital structure and firm'sfinancial performance (Frank and Goyal, 2003;Abor, 2005; Berger and Bonaccorsi, 2006;Nimalathasan and Brabete, 2010; Abbadi andAbu Rub, 2012; Nirajini and Priya, 2013) whilesome studies have reported a significant negativerelationship between capital structure and firm'sfinancial performance (Myers, 1984; Fama andFrench, 1998; Frank and Goyal, 2003; Huangand Sang, 2006; Tang and Jang, 2007; Ebaid,2009; Pratheepkanth, 2011; Velnampy & Niresh,2012; Ebrati, Emadi, Balasang, Safari, 2013).But Abu-Tapanjeh (2006) found that, there is norelationship between the capital structure andprofitability.

In this scenario the researcher attempt toinvestigate the impact of capital structure onprofitability of manufacturing companies listedon the CSE in Sri Lanka. And also the finding ofthis research would be important and useful tothe managers and shareholders of the companiesto take the efficient financing decision.

Research Questions

From the above discussion, the researcherdeveloped the following research questions.

l Does the capital structure have impact onprofitability of the manufacturing companiesin the CSE in Sri Lanka?

l To what extent the capital structurecontribute to the profitability of themanufacturing companies in the CSE in SriLanka?

l Does the capital structure have a significantrelationship with profitability of themanufacturing companies in the CSE in SriLanka?

Objectives of the Study

The primary objectives of this study are;

l To investigate the significant impact ofcapital structure on profitability ofmanufacturing companies listed on the CSEin Sri Lanka.

l To analysis the significant relationshipbetween capital structure and profitability ofthe manufacturing companies in the CSE inSri Lanka?

Methodology

Data Collection and sample selection

For this study the researcher used the secondarydata which were collected from the Annualreports of the selected manufacturing companiesin Sri Lanka. Due to the availability of data, theresearcher considers only 20 manufacturingcompanies from the entire manufacturing sector,

63



which were listed on the CSE in Sri Lanka. Thestudy period is five years from year 2009 to2013.

Research Framework

In this study independent variables are the debtratios. The debt ratios include short-term debt tototal assets ratio, long-term debt to total assetsratio and total debt to total assets ratio. Thedependent variable is Return on Asset (ROA)ratio

Figure 1: Research framework

Hypotheses to be tested

In this study the researcher were developed thefollowing hypotheses

Null hypotheses;

H01: There is no significant impact of capitalstructure on profitability ofmanufacturing companies listed on theCSE in Sri Lanka.

H02: There is no significant relationshipbetween short-term debt to total assetsand profitability of manufacturingcompanies listed on the CSE in SriLanka.

H03: There is no significant relationshipbetween long-term debt to total assetsand profitability of manufacturingcompanies listed on the CSE in SriLanka.

H04: There is no significant relationshipbetween total debt to total assets andprofitability of manufacturingcompanies listed on the CSE in SriLanka.

Alternative hypotheses;

Ha1: There is a significant impact of capitalstructure on profitability ofmanufacturing companies listed on theCSE in Sri Lanka.

Ha2: There is a significant relationshipbetween short-term debt to total assetsand profitability of manufacturingcompanies listed on the CSE in SriLanka.

Ha3: There is a significant relationshipbetween long-term debt to total assetsand profitability of manufacturingcompanies listed on the CSE in SriLanka.

Ha4: There is a significant relationshipbetween total debt to total assets andprofitability of manufacturingcompanies listed on the CSE in SriLanka.

Techniques used to Analyze theData

For the analysis the researcher used multiplelinear regressions and correlation analysis usingthe SPSS 16.0 version software. Multiple linear

64


Short-term debt to Total assets (SDA)

Long-term debt to Total assets (LDA)�

Total debt to Total assets (TDA)�

Independent variable Capital Structure

Return on Asset (ROA)

Dependent Variable Profitability

regression models are formed to find out theimpact of capital structure on profitability.Further correlation coefficient analysis used tofind out the significant relationship between thedepended variable and independent variables.

Model Specification

The regression model will be formulated in thefollowing manner;

YROA = α + β1LDA + β2SDA + β3TDA+ ε

Empirical result

Table 1: Result of Multiple Regressions

(Source: SPSS output)

This multiple linear regression equation showsthat β equals to -0.003, -0.083 and 0.030 whichshows the slop of the regression line. It indicatesthat there is a negative relationship between theLDA and SDA while TDA has a significantpositive relationship with ROA (p(0.000) <0.05). The value of “α” is 0.076. It can bederived the regression equation as,

ROA= 0.076 – 0.003LDA - 0.083SDA + 0.030TDA

ANOVA table of this model indicate that theoverall model is significant since the p-value is(0.000) and R2 is 0.606 indicating the 60.6% ofthe Return on assets (ROA) is explained by thegiven independent variables; LDA, SDA and

TDA. This reveals that capital structure has asignificant impact on Profitability ofmanufacturing listed companies in the CSE in SriLanka. Therefore current study accepts thealternative hypothesis Ha1 and rejects the nullhypothesis H01.

65



Detail Dependent Variable: ROA �- value t Sig.

(Constant) .076 2.656 .009 LDA - .003 -.043 .966 SDA -.083 -1.178 .242 TDA .030 11.340 .000 R .779 R2 .606 Adj. R2 .594 Std. Error .10273 F Value 49.321 Sig.(p.Value) 0.000

66


Statement Null Alternative

Hypothesis Hypothesis

There is no significant impact of capital structure onprofitability Rejected(H01) Accepted (Ha1)

There is no significant relationship between LDA and ROA Rejected(H02) Accepted (Ha2)

There is no significant relationship between SDA and ROA Accepted(H03) Rejected (H03)

LDA SDA TDALDA Pearson Correlation 1

Sig. (2-tailed)N 100

SDA Pearson Correlation .154 1Sig. (2-tailed) .127N 100 100

TDA Pearson Correlation -.282** -.156 1Sig. (2-tailed) .004 .120N 100 100 100

ROA Pearson Correlation -.230* -.196 .775**Sig. (2-tailed) .021 .050 .000N 100 100 100

**. Correlation is significant at the 0.01 level (2-tailed).*. Correlation is significant at the 0.05 level (2-tailed).

Correlation Analysis

Table 2: Result of Correlation Analysis

The above table - 2 shows that the relationshipbetween dependent variable and independentvariable of companies in term of correlationcoefficient. Correlation coefficient of ROA andLDA is -0.230 with a p-value of 0.021 whichdescribes there is a significant negativerelationship between ROA and LDA at 5%significant level. The result of correlationanalysis of ROA and SDA shows coefficient ofcorrelation -0.196, with p-value of 0.050. Itindicates that there is no significant relationshipbetween ROA and SDA. Moreover Correlationcoefficient of ROA and TDA is 0.775 with a p-value of 0.000 which describes there is strongpositive significant relationship between ROAand TDA. From the above discussion the resultsof hypotheses testing have been summarized inthe following table.

Table 3: Result Summery of Hypothesis Testing

Conclusions

The empirical findings of this study revealed thatcapital structure has a significant impact onprofitability of manufacturing listed companiesin the CSE in Sri Lanka. Moreover the resultrevealed that there is a significant negativerelationship between LDA and ROA while foundthat there is a positive strong relationshipbetween TDA and profitability. Mean time theresult indicates that, there is an insignificantnegative relationship between SDA andprofitability of manufacturing listed companiesin the CSE in Sri Lanka.

References:

Abbadi, M., Abu-Rub, N. (2012), “ The Effectof Capital Structure on the Performance ofPalestinian Financial Institutions, BritishJournal of Economics, Finance andManagement Sciences, Vol.3 No.2, pp. 91-101.

Abor, J. (2005), “The effect of capital structureon profitability: an empirical analysis oflisted firms in Ghana”, The Journal of RiskFinance, Vol. 6 No. 5, pp. 438-447.

Abu-Tapanjeh, A.M. (2006), “An EmpiricalStudy of Firm Structure and ProfitabilityRelationship: The Case of Jordan”, Journalof Economic & Administrative Sciences, Vol.22 No.1, pp. 41 -59

Annual Reports (2009-2013) of 20 listedmanufacturing companies in the CSE in SriLanka

Berger, A. N., Bonaccorsidi, P. E. (2006),“Capital structure and firm performance: anew approach to testing agency theory andan application to the banking industry”,Journal of Banking & Finance, Vol.30,pp.1065 -1102.

Brealey, R., Myers, S. and Allen, F. (2006),“Principles of Corporate Finance” 8thedition, McGraw-Hill.

Byoun, S. and Rhim, J. (2003), “Tests of thePecking Order Theory and the TradeoffTheory of Optimal Capital Structure,”Proceedings of the Midwest BusinessEconomics Association, pp. 31-40.

David, A. (2001), “Pricing the strategic value ofput able securities in liquidity crises”,Journal of Financial Economics Vol.59, pp.63–99.

Ebaid, E. I (2009), “The impact of capital-structure choice on firm performance:empirical evidence from Egypt”, The Journalof Risk Finance, Vol.10 No.5, pp. 477-487.

Fama, E.F., and French, K.R. (1998), “Taxes,financing decisions, and firm value”, Journalof Finance, Vol. 53, pp. 819-43.

Frank, M. and Goyal, V. (2003), “Trade-off andPecking Order Theories of Debt” WorkingPaper, Forthcoming in B. Espen Eckbo (ed.),Handbook of Corporate Finance: EmpiricalCorporate Finance, Vol 2.

Harris, M., and A. Raviv. (1991); “The theory ofcapital structure”, Journal of Finance, Vol.46No.1, pp. 297–355.

Haugen, R., and Senbet, L. (1978), “Theinsignificance of bankruptcy costs to thetheory of optimal capital structure”, Journalof Finance, Vol. 33 No. 2, pp. 383-93.

Huang, S., and Song, F. (2006), “Thedeterminants of capital structure: evidencefrom China”, China Economic Review,Vol.17 No.1 pp. 14-36.

Kajananthan, R., and Achchuthan, S. (2013),“Liquidity and Capital Structure: Specialreference to Sri Lanka Telecom Plc”,Advances in Management & AppliedEconomics, Vol. 3 No.5, pp. 1792-7544

Kajananthan, R. and Nimalthasan, P. (2013),“Capital structure and study on Sri Lankanlisted manufacturing companies,” Journal ofBusiness and Management, Vol. 1 No.2,pp.037-044,

Manawaduge, A. S., De Zoysa, A., andChandrakumara, P. (2010), “Capital structureand its implications: empirical evidence froman emerging market in South Asia”, ThirdAnnual International Business ConferenceSri Lankan Journal of Management, Vol. 3,pp.35-51.

Modigliani, F. and Miller, M. (1958), “The costof capital, corporate finance and the theoryof investment”, American Economic Review,Vol. 48, pp 261-297.

Ebrati, M.R., Farzad Emadi, Balasang, R.S.,Safari, G. (2013), “The Impact of Capital

67



Structure on Firm Performance: Evidencefrom Tehran Stock Exchange”, AustralianJournal of Basic and Applied Sciences, Vol.7No.4. pp.1-8.

Myers, S.C. (1984); “The capital structurepuzzle”, Journal of Finance, Vol.39, pp 575-92.

Myers, S.C. and Majluf, N.S. (1984), “Corporatefinancing and investment decisions whenfirms have information that investors do nothave”, Journal of Financial Economics, Vol. 12, pp 187-221.

Nimalathasan, B., Valeriu Brabete (2009), “Thedeterminants of capital structure: evidence onUK property companies”, Journal ofProperty Investment & Finance, Vol. 17 No.5, pp 464-480.

Nimalathasan,B., Valeriu, Brabete ( 2010).Capital Structure and Its Impact onProfitability: A Study of ListedManufacturing Companies in Sri Lanka. The Young Economists Journal, Vol.13, pp. 55-61.

Nirajini, A., Priya, K. B. (2013), “Impact ofCapital structure on financial performance oflisted trading companies in Sri Lanka,”International Journal of Scientific andResearch Publications, Vol.3 No.5 pp.1-9.

Pratheepkanth, P. (2011), “Capital structure andfinancial performance: Evidence fromselected Business companies in CSE in SriLanka”, The International Refereed ResearchJournal V0l-2, No.2.

Raheman, A., Bushra Z., and Mustafa (2007),“Capital Structure and Profitability: Case ofIslamabad Stock Exchange” Internationalreview of Business Research Papers, Vol.3No.5, PP. 347-361.

Samarakon, L.P.(1999), “The capital structure ofSri Lankan Companies”, Sri Lankan Journalof Management Vol. 4 No,1&2 , pp. 18-30

Sangeetha, M., and Sivathaasan, N. (2013),“Factors Determining Capital Structure: ACase study of listed companies in SriLanka”, Research Journal of Finance andAccounting, Vol.4, No.6, pp. 236-247.

Tang, C.H. and Jang, S.S. ( 2007), “Revisit to thedeterminants of capital structure: acomparison between lodging firms andsoftware firms”, International Journal ofHospitality Management, Vol.26 No.1,pp.175-187.

Tharmila, K., Arulvel, K. K. (2013), “The impactof the capital structure and financialperformance: A study of the listed companiestraded in Colombo stock exchange”, Journalof Accounting, Auditing, Economics andFinance, Vol.1No.5, pp.106-117.

Titman, S. (1984), “The effect of capital structureon a firm’s liquidation decisions”, Journal ofFinancial Economics, Vil.13, pp. 137–151.

Titman, S., and Wessels, R. (1988), “TheDeterminants of Capital Structure Choice”,The Journal of Finance, Vol. 43No.1, pp.1-19.

Velnamp, T., Aloy Niresh. J. (2012), “TheRelationship between Capital Structure &Profitability”, Global Journal of Managementand Business Research, Vol.12 No.13, pp. 67-74

68


Date post:	14-Sep-2020
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

JOURNAL OF MANAGEMENT journal/UoSE V_10_BOOK-final.pdfISSN 1391-8230 Volume X No. 1 April 2014...

Documents