Determinants of leverage , case study of vietnam seafood processing and exporting industry both for listed and unlisted firms

The results finding from the thesis is that firm growth opportunities andfirm size have positive related to leverage, profitability, liquidity have a negativeeffect to leverage whereas t

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES

VIETNAM- THE NETHERLANDSPROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

DETERMINANTS OF LEVERAGE: case study of VIETNAM SEAFOOD PROCESSING AND EXPORTING INDUSTRY both for listed and unlisted

firms

By PHAM THI TRUC LAM

A thesis submitted in partial fulfillment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

Academic Supervisor:

Assoc Prof NGUYEN TRONG HOAI

HO CHI MINH CITY, DECEMBER 2012

helped improve my knowledge and fulfill the programme

Finally, I am greatly indebted to my family for their love for and support of me,keeping me in good condition for learning I am also grateful to my close friends fortheir warm encouragement

Although there are many prior studies about leverage as well as capitalstructure on the world in general and Vietnam in particular, results of theseresearches are still inconsistent Moreover, in context of Vietnam, most of studiesabout leverage are of listed firms So, this study works with data of both listed andunlisted firms The purpose of this study is to examine the relative importance of' some factors to the leverage of Vietnam seafood processing and exporting

enterprises A sample of 20 listed and 15 unlisted enterprises for a period of 3 yearsfrom 2009 to 20 11 was chosen and tested base on pecking order theory and tradeoff theory The results finding from the thesis is that firm growth opportunities andfirm size have positive related to leverage, profitability, liquidity have a negativeeffect to leverage whereas tangibility assets and non-debt tax shield have no impact

on leverage of Vietnam seafood processing and exporting enterprises

Key words: leverage, pecking order theory, trade off theory

Certification i

Acknowledgement ii

Abstract iii

Contents iv

List of Tables vii

List of Abbreviations vii

CHAPTER I: INTRODUCTION FOR THE STUDY 1

1.1 Problem statement: 1

1.2 Research objectives: 3

1.3 Research questions: 3

1.4 Research Methodology: 3

1.5 Structure of the thesis: 4

CHAPTER II: LITERATURE REVIEW FOR DETERMINANTS OF LEVERAGE 5

2.1 Key Concepts: 5

2.2 Related theories: 6

2.2.1 The Modigliani-Miller theorem: 6

2.2.2 Agency cost theory: 7

2.2.3 Trade-off theory: 8

2.2.4 The pecking-order theory: 8

2.3 Determinants of leverage: 10

2.3.1 Leverage and Profitability: 12

2.3.2 Leverage and firm size: 13

2.3 3 Leverage and Firm Growth: 15

2.3.4 Leverage and Non-debt tax shield : 16

2.3.5 Leverage and tangible assets: 17

2.3.6 Leverage and liquidity: 19

2.4 Empirical review on determinants of leverage: 20

2.4.1 Empirical evidences finding from the world: 20

2.4.2 Empirical evidences finding from Vietnam perspective: 24

2.5 Chapter remarks: 25

CHAPTER III: RESEARCH METHODOLOGY 27

3.1 Data and variables description: 27

3.1.1 Data: 27

3 1.2 Description of variables: 27

3.2 Methods of estimation: 29

3.2.1 Ordinary Least Squares (OLS) estimator: 30

3 2.2 Fixed Effects (FE) estimator: 31

3.2.3 Random Effects Estimator (RE): 32

3.2.4 Hausman specification test: 32

3.3 Hypotheses 33

3.4 Chapter remarks: 33

CHAPTER IV: OVERVIEW OF SEAFOOD PROCESSING AND EXPORTING ENTERPRISES 35 4.1 Introduction of the seafood export industry: 35

4.2 Analysis of capital structure of the seafood enterprises: 38

4.2.1 Debt to equity ratio: 38

4.2.2 The equity growth rate: 41

4.3 Chapter remarks: 42

CHAPTER V: EMPIRICAL RESULTS 43

5.1 Descriptive statistics 43

5.2 Analysis of the correlation between all variables: 44

• 5.3 Test multicollinearity: 45

5.4 Empirical results: 46

5.4.1 OLS estimator: 46

5.4.2 Hausman specification test: 48

5.4.3 Fixed effects estimator: 49

•

5 5 Chapter remarks: 53CHAPTER VI: CONCLUSION, POLICY RECOMMENDATION AND

LIMITATION 556.1 Conclusion 55

6.2 Policy recommendation 576.3 Limitation and suggestion for further study: 61Reference 62Appendix 68

"

•

LIST OF TABLES

Table 2.1: Summary of leverage determinants

Table 3.1: Variable, description and its expected sign

Table 4.1: The equity growth rate

Table 5.1: Descriptive statistics of sample variables

Table 5.2: Correlation coefficient matrix

Table 5.3: Result of auxiliary regression models of each dependent variableTable 5.4: Result ofOLS estimator

Table 5.5: Result ofHausman test

Table 5.6: Result of Fixed effects estimator

LIST OF ABBREVIATIONS

NDTS Non debt tax shield

SEAs Seafood processing and exporting enterprisesSMEs Small and medium enterprises

GDP Gross Domestic Product

GSO General Statistics Office of Vietnam

OLS Ordinary Least Square

STD Short term debt

.

CHAPTER 1: INTRODUCTION FOR THE STUDY

This chapter will explain the reason for choosing the thesis, its objectivesand research questions In addition, this part also presents a brief of methodologyand finally abbreviates the structure of the thesis

1.1 Problem statement:

One of the most important decisions confronting a firm in corporate finance

is the design of its capital structure In order to finance for investment, the firmscan choose either internal or external funds Internal fund is the firm retainedearnings whereas external fund can be raised from issuing equity or taking debt.Sometimes, firms may take too much or without taking debt An effective debt ratio

or leverage should not only help reducing the weighted average cost of capital ofthe firm but also it still can increase firm value To decide how much leverage thefirm should take on, the firms need to know which factors affect leverage.However, over 50 years since the first research of Modigliani and Miller (1958)about leverage, the question of them about what factors affect leverage are stillunresolved until now This seems to be the most enigmatic and interested issue ofmany researchers However, until now, the effort of researchers to solve the debtequity choice also does not bring results that satisfy all of them The results are stillinconsistent between researches at different countries and different industries.Moreover, Myers ( 1984) also suggests that "the average debt ratio will vary fromindustry to industry because asset risk, asset type and requirements for externalfund also vary by industry" (p.578) In case of Vietnam, there are also some studiestesting for determinants of the leverage of Vietnam companies, such as Nguyen andRamachandran (2006), Dzung et al (2012) and the empirical results of thesestudies also contain conflicts as results from many other studies in the world

Seafood is considered to be one of major industries and one of the keyexport sectors of Vietnam Each year, it contributes largely to GDP Particularly,

in year 2011, total value of seafood product is 99.432 billion VND, contributes about 4% to GDP (Source: GSO 2011 ) Vietnam has many outstanding advantages

- in the industry to develop this field as high position in most leading markets,

diversified channel of distribution, large farming area with lot of species ofseafood However, with many changing about interest rate through past theseyears, many seafood enterprises have been affected strongly as most of theirsources of capital are loan from bank In year 2009-2010, in order to support forfirms in time of difficulties of the economy in Vietnam as well as the worldeconomy crisis, the Government loosed monetary policy However, this forcedcredit growth become too high, leaving far growth of GDP This created unbalanceand pushed up inflation rate Therefore, by the year 2011, to control inflation, theGovernment has employed tightening monetary policy through controlling creditstrictly and limiting in supplying money for the economy Commercial banksbecome short of fund to finance for trade According this context, loan interest rateincrease and conditions to be approved for lending from bank are also moreserious This makes a lot of enterprises hard to reach of loan capital Together withmany other difficulties of seafood firm as lacking of source of material, highercost of input, decreasing of export market, lacking of capital lead many seafoodfirms to a more difficult situation, even more a lot of firms must go bankrupt Only

at the beginning of the year 2012, 470/800 exporting seafood firms must stopworking, not exclude large companies The problem is these enterprises dependtoo much on debt while the way of using debt is ineffectively So, nowrestructuring of capital structure become the most important problem for Vietnamseafood enterprises To improve financial ability and avoid risk of financial,seafood enterprises need to have a suitable leverage In order to do that, it's veryimportant for firms to know whether leverage of firms may be impacted by whichfactors Finding factors affecting leverage will help enterprises to control itsleverage through impacting on these factors

1.2 Research objectives:

The main objective of this thesis is to examine the impact of some factors asprofitability, firm growth opportunities, firm size, tangibility assets, non debt taxshield and liquidity on leverage in context of Vietnam seafood processing andexporting enterprises It is to consider whether these factors are significant, if yes,

it is positive or negative related to leverage and which theory help to explainleverage of the firms better: pecking order or trade off theory The specificobjectives are as follows:

- To examine the significant impact of factors on leverage

- To give some policy recommendations to the firms to find a better way of financing debt for Vietnam seafood processing and exporting enterprises

1.3 Research questions:

The study focuses on answering the questions:

- What factors affect leverage of Vietnam seafood processmg and exporting enterprises?

- What policy recommendations for Vietnam seafood processing and exporting enterprises to raise effect of using leverage?

Basing on the list of factors affecting on leverage from the prior literatures,this study will do the examination on some selected factors in case of Vietnamseafood processing and exporting enterprises

1.4 Research Methodology:

This paper uses annual panel data of 3 5 seafood processing and exportingenterprise in Vietnam for the period 2009-2011 include both listed and unlisted

To estimate the impact of independent variables as profitability, non-debt tax

shield, firm size, firm growth opportunities, tangibility assets and liquidity on

leverage of the firms, two methods will be used: Ordinary Least Square and Fixed

effects (FE) or Random effects (RE) which is decided by Hausman specificationtest (Dougherty (20 11) ).

1.5 Structure of the thesis:

This thesis is organized in six chapters as follows:

Chapter 1: explains the reason for choosing the thesis, its objectives and

research questions and briefly about methodology which is used to test forhypotheses

Chapter II: demonstrates the literature review It starts with the concept ofleverage and capital structure Following is some theories and prediction oftheories about determinants of leverage Empirical studies are also presented inthis part

Chapter III: descriptions data and variables which be used in the model Thischapter also presents about different estimation techniques together withhypothesis testes

Chapter IV: presents turnover of seafood firms through some years, its exportmarket as well as major seafood product This chapter also analyzes capitalstructure of these firms

Chapter V: indicates empirical results corresponding to each estimationtechnique as well as results of hypothesis testes

Chapter VI: gives conclusion of empirical results, suggest recommendationsand limitation of the thesis

CHAPTER II: LITERATURE REVIEW FOR DETERMINANTS OF LEVERAGE

This chapter firstly presents about concept of capital structure as well asleverage and optimal capital structure Next are some leverage theories includingModigliani - Miller theorem, agency cost theory, pecking order theory and tradeoff theory It also includes prediction of these theories about several factorsaffecting leverage Finally, empirical studies concerning these factors will bementioned

2.1 Key Concepts:

+Capital structure: refers to the way that a firm finances its assets through

mixing of long term debt, short term debt, common equity and preferred equity It'smeasured by some targets as ratio of debt per total assets (also called leverage),ratio of equity per total assets or ratio of debt per equity

+Leverage: this study will use ratio of debt per total assets (or leverage) to

define for capital structure to consider the proportion of debt used to finance theassets of the company This ratio still shows the capacity of repayment of the firms

in case firms do not work effectively with debt A higher this ratio represents forthe more debts compared with assets the companies have, the more leveraged aswell as the riskier it is considered to be

Depending on the objective of analysis, leverage is estimated by someways First, it is the ratio of total liabilities to total assets This is a measure of

"what is left for shareholders in case of liquidation" (Rajan and Zingales, 1995,p8) Second, it is the ratio of debt (include both short term and long term debt) tototal assets Many researchers have chosen ratio of debt to present for leverage asBevan and Danbolt (2002), Huang and Song (2002) with short term and long termdebt to be examined separately, Cuong & Canh (2012) combine both short term andlong term debt This thesis will use the ratio of liabilities to total assets to measurefor leverage include both short term, long term debt and trade credit

+ Optimal capital structure: the best debt to equity ratio that a firm can

have Modigliani and Miller ( 1963) state firms will select the mix of debt andequity that minimizes their weighted average cost of capital Blank (2000) claimsthat the optimal capital structure refers to such a ratio of using debt and equity thatmaximizing the firm' market value

+ Bankruptcy cost: is understood as the increased costs when the firms

finance with debt instead of equity that result from a higher probability ofbankruptcy Bankruptcy cost occurs when the fixed obligations to creditor can not

be met (Robert and Lemma (1978))

+ Agency cost: a cost of the agency relation ship which is "a contract under

which one or more persons (the principal(s) engage another person (the agent) toperform some service on their behalf which involves delegating some decisionmaking authority to the agent" (Jensen and Meckling (1986))

In short, "leverage" is the main concept in the thesis It's used as dependentvariable and the thesis will find factors affecting it Bankruptcy cost and agencycost are used in explaining about effect of factors to leverage

2.2 Related theories:

2.2.1 The Modigliani-Miller theorem:

Whenever mentioning about leverage, almost studies base on theory ofModigliani and Miller proposition This is considered the pioneer and the mostimportant theory about leverage in modem sense It's usually still called MMtheory

As Murray and Vidhan (2007, p.6) stated: "the Modigliani-Miller theoremdoes not provide a realistic description of how firms finance their operations, it

provides a means of finding reasons why financing may matter" This theorem isalso the pattern for many later theories about leverage of the firm

Modigliani and Miller (1958) consider linking between cost of capital anddebt of enterprise Every model must base on certain assumptions, Modigliani andMiller theory was not also out of this rule Under assumptions about perfect

markets: no transaction cost, no tax, no bankruptcy cost or no unequal access toinformation, the theory concludes that firm value do not base on capital structure,leverage have no effect on firm's value However, assumptions about perfectmarkets do not exist in reality Until 1963, there is a progress on this theory byadding corporate income tax on cost of capital This leads to the results thatleverage concern with firm value Due to tax deductibility of interest payment ondebt, the higher debt ratio the firm use, the higher value the firm get So, thistheory supports that the firm should use debt as much as possible to inherit taxdeduction The result of this theory has become an interest for many economistslater.

Since this pioneer theory, many other theories have been suggested todecide determinants of leverage of the firms Three fundamental theories which beusually used by many researchers will be presented in next following parts

2.2.2 Agency cost theory:

Research concerning this theory was initiated by Jensen and Meckling(1976) Jensen and Meckling (1976, p.308) defined the agency relationship as "acontract under which one or more persons (the principal) engage another person(the agent), to perform some service on their behalf which involves delegatingsome decision making authority to the agent" According to Jensen and Meckling(1976), there are two types of conflict, the first is between shareholders andmanagers, the second is between shareholder and debt holder Benefit ofshareholders is related to firm's profitability, so shareholders wish for management

to run the company in a way that increases shareholder value While benefit ofmanagement connects closely with their income, so they may wish to grow thecompany in ways that maximize their personal power and wealth that may not be

in the best interests of shareholders

Due to conflict between various group, Jensen and Meckling (1976) argue that agency cost play an important role in financing decisions and the optimal

capital structure will be determined by minimizing the costs arising from these

conflicts

2.2.3 Trade-of/theory (Myers, 1984) (or Tax based theory):

Theories about leverage are divided into two groups - theory which proposesthe optimal debt-equity ratio and theory which define no well-defined target debt-equity ratio (Patrik (2004)) Trade off theory relating to group of theories proposingthe optimal debt-equity ratio is basing on theorem of Modigliani and Miller (1963),this proposes that corporate income tax was added to cost of capital and postulate theexisting of an optimal capital structure

It refers to the idea that a company should choose how much debt finance andhow much equity finance to balance the costs and benefits Another way, it requires atrade off between tax advantages of borrowed and cost of financial distress Both debtand equity used policy also have its own advantage and disadvantage While companyheavily relies on debt have to utilize largely part of its income for paying interestpayment, a debt free company can use all of its net income to refinancing for newinvestment (Zehra (2008)) However, to compensate for loss income from payinginterest payment, these companies can inherit tax shield - is understood as thedifferences between taxes of a company having debt and company without having debt

on its capital structure (Wrightsman (1978))- from its interest payment

The optimal capital structure is at which interest tax shield balance withbankruptcy costs and agency costs In another way, trade off theory says that optimalcapital structure will be obtained by determining to trade off between benefits of debtwith the costs

2.2.4 The pecking-order theory (also being called as the information asymmetry theory):

Talking about pecking order theory, almost researches mentioned about Myers and Majluf while Harris and Raviv (1991) raised that this theory first is

resulted from the work of Donaldson ( 1961) and then developed by Myers and Majluf(1984).

This theory mentions about three sources of funds as retained earnings, debtand equity It begins with the basic assumptions that managers have moreinformation about value of the firm's assets than potential investors So this theory

is still called asymmetry theory and to be considered as contestant with "trade-offtheory Both manager and investor know about this fact According to Myers andMajluf (1984 ), due to lacking of information about the firm, outside investors willtend to misprize equity of the firm Even if NPV of the project is positive, it alsomay not be accepted (Harris and Raviv, (1991))

In order to minimize the problem of information asymmetry betweenmanagers (insiders) and investors (outsiders) and avoid underinvestment, thistheory encourage using other ways of collecting fund instead of issuing equity.Myers and Majluf (1984, p.46) went to the result that "firms should go to bondmarket for external capital, but raise equity by retention if possible That is,external financing using debt is better than financing by equity" Particularly, thistheory proposes that when the firms collect capital for new investment, it shouldfollow a hierarchy financing Firstly, it will choose internal funds (retainedearnings) and only when this source of fund is exhaust, it will acquire externalfunds Using external funds, firms also give priority to debt instead of equity.Issuing equity is considered the most expensive way to collect fund for newinvestment, so in order to avoid transaction costs, firms will choose this way as thelast channel of collecting fund (Fama and French (2004)) when it does not haveany other selection The decision rule of Myers and Majluf (1984, p.3) is "takeevery positive- NPV project, regardless of whether internal or external funds areused to pay for it" This theory also argues that there is no well-defined targetleverage and leverage depends on active business and investment need Due toraising the role of internal funds, according to this theory, profitability of the firm

is one of the most important factors affecting leverage of the firm

Deciding to choose external financing to finance new investment is not onlybecause of existing asymmetric information problem but also relating to position

of organizational sales and structure of the firms (Mehmet and Eda (2008)) Firmswith stable sales will create belief in lenders about collecting loan due to stableprofitability, so they will easily to borrow than other firms Similarly with firmswhich have larger size and structure since they will have more collateral assets toensure for their loan

Murray and Vidhan (2007, p.1) recognize that "private firms seem to useretained earnings and bank debt heavily Small public firms make active use ofequity financing Large public firms primarily use retained earnings and corporatebonds"

From the above theories, some important factors affect leverage of the firms

as profitability, growth rate, firm size, tangibility assets, liquidity and non-debt taxshield which will be explained clearly in next part

2.3 Determinants of leverage:

Determinants of leverage can be divided into two groups as firm specificfactors which are considered as internal factors and industry or country specificfactors which are considered as external factors (Antoniou et al (2002) andGurcharan (20 10) ) External factors including macro elements are out of control

of the firms Due to limitation of data, this study only bases on investigating aboutfirm characteristics that can be controlled by the manager of the firm Moreover, itseems that firm specific factors affect leverage more strongly than industry specificfactors (Mac (2010)) DeJong et al (2008) considering both firm specific factorsand country specific variables also conclude that firm specific factors havedominant role to determinants of leverage

There are many factors having means to leverage, however as Harris andRaviv (1991, p.299) claims in their research: "the models surveyed have identified

a large number of potential determinants of capital structure The empirical work

so far has not,however, sorted out which of these are important in various

contexts"

Harris and Raviv (1991, p.334) has summarized results from many studiesthat "leverage increases with fixed assets, non-debt tax shields, growthopportunities, and firm size and decreases with volatility, advertising expenditures,research and development expenditures, bankruptcy probability, profitability anduniqueness of the product" Due to lacking of data, this study will base on some offactors which are found significant in research of Harris and Raviv ( 1991) such asprofitability, firm size, growth rate, non-debt tax shield and tangibility assets plus

of liquidity ratio as Zehra (2008) uses to check for Vietnam seafood processing andexporting enterprises Three factors which the study investigates as profitability,tangibility and firm size are also three of six core factors that make changing up to27% of the leverage in many studies before (Murray and Vidhan (2007))

Pecking order theory and trade off theory are considered as two mamtheories in explaining determinants of leverage of the firms in almost previousstudies Both of these theories can explain leverage although explanation of eachtheory about effect of factors is different Some researches support that peckingorder theory explain better choice of leverage (DeJong et at (2008), some othersincline to follow trade off theory (Murray and Vidhan (2007) or Jason and Eric(2009)) whereas there are also researches argue leverage is consistent withprediction of both pecking order and trade off theory (Deesomsak et al (2004 ),Antoniou et al (2002) In context of Vietnam firms, Dzung et al (2012) states thatpecking order theory explains better leverage of firms

This study also focuses on both pecking order theory and trade off theory tofind whether leverage of Vietnam's firm is affected by which theory The followingsection will explain impact of each factor on leverage according to point of view ofpecking order theory and trade off theory one by one Pecking order theory seems

to be better than trade off theory in explaining leverage of the firms due toaccording trade off theory, firms should take more debt to inherit tax shield

deduction, however, in reality, there are some firms with high profit take less debt.This theory can not explain for that With pecking order theory, this is explainedthrough firms' hierarchy of finance: retained earnings, debt and equity.

2.3.1 Leverage and Profitability:

Theoretical prediction about effect of this factor on leverage is inconsistent

• According to trade-off theory, profitability of a firm clearly has positive

relationship with leverage and it's also a key point of this theory (Fama andFrench (2002), Mehdi et al (2011)) Profitable firms seldom meet problem withfinancial distress and they also find interest tax shield more valuable (Murray andVidhan (2007)) Therefore, higher profitability encourage firm to take more debt

to finance investment in order to inherit tax benefit of debt Moreover, a firm withhigh profitability may create belief in lender, so it's easy for them to borrow thanfirms with bad profitability (Philippe et al (2003))

On the contrary, the pecking order theory supports a negative relationshipbetween them Because of asymmetric information situation, cost of collectingoutside capital will be always higher than inside capital So manager likely to useretained earning for self supporting if any Especially, firms with higherprofitability will certainly have more retained earnings Therefore, the firm willuse this internal source of fund to finance its investment, without depending onexternal finance to avoid problem of changing of interest rate or pressure ofrepayment at due date, so level of ratio of debt will be at low degree

Most empirical studies are in line with pecking order theory that

• profitability have negatively correlated with leverage Murray and Vidhan (2007,

p.32) supports that "firms that have more profits tend to have less leverage".Akinlo (2011) uses data of 66 Nigerian listed firms over the period 1999-2007argues that good profitability firms tend to reduce their need for external debt Li-

Ju and Shun-Yu (2010) find the most significant variable impact on leverage isprofitability and its coefficient is negative This result found base on data of 305Taiwan electronic companies, one of most important industry in Taiwan, during

the year 2009 Titman & Wessels ( 1988), Schoubben and Van Hulle (2004) alsohave the same results While supporting for trade off theory, Mehdi et al (2011)find that profitability is the most influential factor on leverage and it has positiveimpact on leverage.

In context of Vietnam, enterprises feel difficult to borrow from bank due tohigh transaction cost plus of so many requests as tangibility assets, aim of usingfund or checking after supplying loan so enterprises with more retained earninglikely to use this source for financing and avoid taking debt from bank It seemsthat pecking order theory is more suitable to explain for leverage of Vietnam firms

So, a first hypothesis is conducted for the relationship between profitability andleverage as follows:

Hl: There will be negative relationship between profitability and leverage

2.3.2 Leverage and Firm size:

One of main conditions to be considered whenever lenders approve offinancing for a firm is related to its size From the point of view of pecking orderand trade offtheory, the effect of firm size on leverage is mix

Pecking order theory predicts a negative relationship between size andleverage of a company Large firms seldom meet problem with informationasymmetries, so the costs of issuing equity for large firms are lower than smallfirms Moreover, due to information about large firms is more popular with outsideinvestors, they will likely to get equity of these firms (Huang and Song (2002)).With low cost of issuing and preference of outside investors, certainly large firmswill give priority to issuing equity to finance for new investment project toeconomize the cost instead of taking debt

With trade off theory, the relationship is positive Due to the larger the firmsare, the more diversified they are, so size is still considered as an inverse proxy ofprobability of bankruptcy (Titman and Wessels, (1988)) Large firms will less beimpacted by bankruptcy, hence they are easily to borrow with lower interest rate(Pinches and Mingo, (1973)), so the leverage will be higher Besides,

lenders also prefer to supply debt to larger firm than smaller firm due to it's easierfor them to get repaid (Joshua, (2008)).

Empirical studies support for the trade off theory and find the positiverelationship as Patrik (2004) with data of Czech Murray and Vidhan (2007) withdata of publicly traded American firms conclude that "larger firms tend to havehigh leverage" (p.32) Research of Amarjit (2011) about 166 Canadian listed firmsfor the period of 3 years from 2008 to 2010 also find that leverage positive related

to firm size Rajan and Zingales (1995) find positive relationship for almostcountries in G-7 except for Germany Pervez (2009) supports that large firm tend

to borrow more debt than smaller do With data of Vietnam, Anh (20 10) also findthe positive relationship between firm size and leverage In a research aboutdemand of debt of six African countries, Bigsten et al (2000, p 7) supports that

"small and medium sized firms are less likely to get a loan" In this research, thewriter argues that smaller firms usually meet problem with credit So, althoughthese firms have high demand of debt, their request usually are rejected than largerfirms While ratio of demand but rejected of micro firms is 64%, small firms is42% and medium firms is 21%, this ratio is only 10% for large firms

In opposite with those results, Abel (2008) follows pecking order theory and

do not agree with positive relationship between firm size and leverage in case size

is explained as a inverse proxy for bankruptcy cost The result of the researchabout 71 quoted firms in Nigerian stock market during the period 17 years from

1990 to 2006 supports for a negative related due to "larger firms that have builtenough reserves may choose to finance their operations through their respectiveinternal markets, rather than passing through the difficulties inherent in accessingthe external financial markets" (p.9)

In context of Vietnam, larges firms usually receive favors of many banks.These banks competes each other to attract them through reducing interest rate andcost

So, a second hypothesis is conducted for the relationship between firm sizeand leverage as follows:

H2: There will be positive relationship between firm size and leverage

2.3.3 Leverage and Firm Growth:

The relationship between growth and leverage of the firm is still unclearwith different theories While the pecking order theory argues a positive

relationship, trade off theory states a negative one

With trade off theory, growing of the firms is only intangible assets, so itcan't be used as collateral to borrow Moreover, firms with more investmentchances tend to have lower leverage since they have stronger incentives to avoidunder-investment and asset substitution that can arise from stockholder-bondholderagency conflicts (Drobetz and Fix (2003))

Researches of Raj an and Zingales ( 1995) plus of Titman and Wessels(1988) supports for this negative relationship Huang and Song (2002) also findthat in China, firms with high growth rate in the past tend to have higher leveragewhile firms with good growth opportunity at present have lower leverage Patrik(2004) states that firms with higher growth opportunities tend to give priority tousing equity Gurcharan (2010) agrees that firm with high growth have many otherdeveloping chances, so they do not need to take much debt to avoid the restriction

of lenders Akinlo (20 11) also supposes that growing firms do not depend toomuch on debt financing for Nigerian case

From the perspective of pecking order theory, the requirement of finance ofthe firms will increase together with its growing The internal fund cannot fill theirneed And if only depending on this source of fund, the growth of firms may be

restricted So the firms tend to take more second source of finance - debt - to

finance for their investment Researches encourage for this theory as Baskin(1989), Bevan and Danbolt (2002) Most of researches for Vietnam contextindicate that firms with high growth will finance with more debt (Nguyen andRamachandran (2006) and Biger et al (2008)) Therefore, a third hypothesis isconducted for the relationship between firm growth rate and leverage as follows:

H3: There will be positive relationship between growth rate andleverage

2.3.4 Leverage and Non-debt tax shield:

Non-debt tax shield is a characteristic that fits in with only the trade off theory According to this theory, firms should take more debt to inherit deducting

corporate tax However, besides this way, firms also can reduce corporate tax by

using non-debt tax shield as depreciation Therefore, this theory supports for anegatively correlated between non-debt tax shield and leverage The moredepreciation the firms have, the less potential debt they will use De Angelo andMasulis ( 1980), one of the persons who care and raise the role of non-debt taxshield argue that non-debt tax shields are substitutes for a debt-related tax shield,

so firms with large non-debt tax shield can also inherit tax deduction and have lessdemand of debt on their capital structure However, Scott (1977) proposes acontradiction Scott ( 1977) connects debt capacity of the firm with non debt taxshield to explain for positive relationship between leverage and non-debt taxshield According to his opinion, firm with more collateral assets also have moredepreciation Besides getting tax deduction from depreciation, these firms willmake use of having lower interest rate from secured debt to borrow more So,together with increasing of non-debt tax shield, leverage of these firms alsoincrease

In reality, most of empirical studies support for negative relationship oftrade off theory and De Angelo and Masulis (1980) although there are also someexceptions depending on country characteristics Linda and George (200 1) withDutch data, Patrik (2004) with Czech data support for this negative relationshipthat non-debt tax shield as substitutes to debt-related tax shield whereas Bradley et

al (1984) reports a positive relationship Gurcharan (20 10) with study about someselected countries in Asean region claims that due to being inherited benefit fromnon-debt tax shield, firms likely to reduce using debt Liaqat (20 11) used data of

170 Indian companies in the textile industry for the period 2005-2010 find a

positive result inconsistent with theory that increasing in non-debt tax shield alsoleads to increasing in leverage ClarifYing for this contradiction, the researcherclaimed that "expected income streams of textile firms in India, against whichinterest expenses and non debt tax shield (depreciation), can be deducted are veryhigh as compared to the total of debt and non-debt tax deductions Therefore,depreciation does not work as a "substitute to the tax benefits of debt financing inthe Indian textile firms" (p.58).

With research about the role of tax in capital structure of Arab tax and nontax economy, Mounther and Ramesh (2004, p.ll) assume that in taxed countries, ifnon-debt tax shield proxies for its "intended effect", it should have a adverserelated with leverage" whereas if it proxies for collateral effect, it expect "either apositive, negative or insignificant coefficient for non debt tax shield in taxableregimes depending on the relative strength of the substitution vs collateraleffects" So, a fourth hypothesis is conducted for the relationship between non-debttax shield and leverage as follows:

H4: There will be negative relationship between non-debt tax shield andleverage

2.3.5 Leverage and tangible assets:

While profitability is mentioned as key rejection of leverage, tangible asset

is considered as the major factor which determines increasing of leverage of thefirm (Titman and Wessels (1988) and Harris and Raviv (1991)) Most of researchesshow that there is positive relationship between tangible assets and leverage Thatalso complies with pecking-order theory and trade off theory, the more tangibleassets the companies have, the more easily for them to take loan because theseassets will be used as collaterals to assure for their loan Debt being assured withtangible assets also make smaller lost for lender in case the firm go into distress, solender also like to supply debt for firms with high tangible assets (Murray andVidhan (2007)) Bigsten et al (2000) also find that one of the reasons of not takingdebt of firms is that the collateral is inadequate with their demand

Some other empirical studies support this positive relationship as Rajan andZingales ( 1995), Murray and Vidhan (2007), Dilek et al (2009), Mehdi et al (2011) with data of 72 Iranian listed companies find the result complying withpecking order theory and trade off theory that leverage has positive associatedwith tangibility Nguyen and Ramachandran (2006) give result that Vietnam firmswith more tangibility assets also have more debt.

Research of Jason and Eric about leverage of firms in a group of countriesincluding American, Asian and Europe also states that tangibility has positiveimpact on leverage Explaining about positive relationship between leverage andtangibility assets, Farooqi (2006) link with growth options of the firms due tofirms with more tangibility assets usually are "mature firms"

Mounther and Ramesh (2004) suggest that collateral asset will increases thedebt repayment capacity of the firm and therefore leverage will increase at thesame direction with tangibility assets

Although most of researches supports that firms with more tangibility assetswill get more debt, there are also several researches find the opposite result.Example Daskalakis and Psillaki (2009) find an inverse impact between leverageand fixed assets in small and medium firms Joever (2006) also has the same resultfor less developed economies Florinita (2012) with data of 100 firms listed in

2010 in Romania find that "companies with large portion of fixed assets discourage

to employ debt capital" (p.529) Explaining for this, the researcher proposes maybethese firms have already found "a stable source of return which provides themmore internally generated funds and discourage them from turning to externalfinancing" (p.529) Researching about determinants of leverage of Europeancountries of Antoniou (2002), the result about relationship between leverage andtangibility assets is completely different between countries While the result foundfor Germany is positive related between leverage and tangibility assets, the resultfor the UK is negative and insignificant in France Explaining for insignificantvalue in French, the researcher argues that bankruptcy laws in France is "stronglyfavor the rehabilitation of firm hence reducing the need to recover the

loan by selling tangible assets" (p.13) So, a fifth hypothesis is conducted for therelationship between tangible assets and leverage as follows:

HS: there will be positive relationship between tangible asset and leverage

2.3.6 Leverage and liquidity:

This is the proxy which attracts researchers less than other variables.Empirical studies about effect of this factor on leverage are not as much as abovefactors This factor is represented for the capacity of returning debt of a firm

According to pecking order theory, to finance for new investment project,firms prefer to use internal finance to external finance Firms with more liquidityasset will have more cash flow to finance for its investment instead of depending

on debt So it supports an inversely related between leverage and liquidity Thatleads to its leverage will be at low degree Supporting for this idea, Mehdi et al.(20 11) with research about leverage of Iranian listed firms by using data of 72firms from 2003 to 2009 find the result that "companies with more liquid assetsuse lesser amount of debt compare to companies with lower liquidity ratio"(p.168) With data of four countries in Asia Pacific region include Thailand,Malaysia, Singapore and Australia over the period 1993-2001, Deesomsak et al.(2004) conclude that liquidity has a significant negative relationship with leveragefor all countries they estimate They state that "firms tend to use their liquid assets

to finance their investment in preference to raising external debt" (p.15)

On the other hand, sometimes, firms with higher liquidity ratio tend to havehigher leverage Since their capacity to match with short term obligations whenthey fall due, they may be easily to borrow than firms with low liquidity ratio Thisencourages a favorable relationship between liquidity and leverage Research ofZehra (2008) about UK listed companies from 1998-2007 supports for this idea.Ntogwa (2012) also argues a positive related in case of Tanzanian non-financialcompanies in the sense of "high liquidity eases the availability of debt" (p.418)

Due to the main element of Vietnam' firms is current liabilities (Vuong and Tran, 201 0), liquidity is a factor that should be attended more in determinants

of leverage So, a sixth hypothesis Is conducted for the relationship betweenliquidity and leverage as follows:

H6: there will be an adversely related between liquidity and leverage

Table 2.1: Summary of leverage determinants

-Source: by the author after literature review done

2.4 Empirical review on determinants of leverage:

Through many past years, a lot of researches about leverage of a specificcountry, comparison between two countries or a region of area have been done tofind out which factors affect leverage of the firm However, the finding results arenot the same Even more, some studies show contradiction with theories Thissection will be divided into two parts include researches concerning countries onthe world and researches concerning Vietnam context

2.4.1 Empirical evidences finding from the world:

Gurcharan (20 10) looks at leverage of some selected countries in Aseanarea as Malaysia, Indonesia, Philippine and Thailand On one hand, the researcheruses firm specific factors which be used by many other researches as profitability,growth opportunities, non-debt tax shields and firm size On the other hand, he stillobserves country-specific factor as size of banking industry and stock market,gross domestic product growth rate and inflation

Data which is selected to test for relationship between leverage and specific factors are of 155 main listed firms on stock exchange (excluded financialcompanies) during the period 2003 to 2007 while data for country-specific arechosen mostly from World Bank and Euromonitor International The research findsthat profitability, growth and non-debt tax shield have negative relationship withleverage but insignificant with Malaysia for indicator profitability, with Indonesiafor indicator growth and significant with only Malaysia for indicator non-debtshield About firm size, the results are significant with Indonesia and Philippine forpositive effect whereas the results for two rest of countries Malaysia and Thailandare insignificant with negative correlation The study also shows that countryfactors as stock market and GDP growth rate have negative impact on leveragewhereas bank size and inflation have no correlation with leverage This resultimplies that factors affect leverage will vary from country to country.

firm-Huang and Song (2002) also do a research with data of over 1000companies in China - the largest developing and transition economy of the world-

up to the year 2000 They divide leverage into short term debt and long term debtand the result they find that China tend to have much lower long term debt Theychecks for the role of some proxies as profitability, firm size, non-debt tax shield,fixed assets, corporate tax, growth opportunities, volatility, ownership structure andmanagerial shareholdings They uses both book value and market value of equity,

so they uses until 6 measures of leverage, within total liabilities is main measureand others as long-term debt ratio ands short-term plus long term debt ratio areused to check for robustness The results of effect of these factors to leverage of thefirm are consistent with other previous researches They conclude that profitability

is "strongly negatively related" with leverage Growth rate and non-debt tax shieldsalso have adverse impact on leverage while size, tangibility, volatility andownership of institutes have favorably influence Remain factors as tax andmanagement share holdings have no impact on leverage Especially, in China,leverage ratio of listed companies is much lower They tend to use more equityrather than debt in their capital structure It's hardly for firms to take debt

due to the credit market in China contains many regulation and mostly beingdecided by central bank Moreover, bond market is still at early stage ofdevelopment The situation of China is contrary to the prediction of pecking ordertheory that gives priority to debt instead of equity.

Zehra (2008) analyzes the determinants of leverage of UK firms and bases

on data of 173 firms over a period of ten years from 1998 to 2007 The study usedthree dependent variables to measure for leverage as total debt, long term debt andshort term debt The explanatory variables included firm size, growthopportunities, non-debt tax shield, profitability, liquidity, uniqueness and assettangibility Two techniques of regression of panel data method as pooled OLS andfixed effect model have been employed to investigate the effect of these variables

to leverage after being checked by Breusch and Pagan Lagrangian Multiplier testbetween random effects model and pooled OLS and Hausman test between randomeffects and fixed effects model The results of the study are as follow:

- Size has positive related to all three variable - total leverage, long termand short term leverage under pooled OLS This result is suitable with trade offtheory and expected sign of the study However, with fixed effect model, size onlyhas positive effect to total and long term leverage and negative effect to short termleverage So, the study encourages that large firm should chooses long term debtinstead of short term debt to finance for its investment

- Growth opportunities has negative relationship with total leverage andshort term leverage and positive related to long term leverage under pooled OLSwhile fixed effect regression give negative result for all kind of leverages

- Non-debt tax shield is found to have positively related with total leverage,short term leverage and long term leverage under fixed effect method but it's onlysignificant with total leverage while with pooled OLS, a negative related betweennon debt tax shield and long term leverage is supposed

- Profitability: this research use two indicator for profitability as the ratio ofoperating profit to total assets and the operating profit to sales For the first

indicator, the pooled OLS gives a negative related between profitability and longterm leverage and a positive related with total and short term leverage whereasfixed effect model shows negative relationship for all three measure of leverage.The second indicator of profitability is found to have positive related with leveragefor all Only one independent variable but with different estimation way, it may beconsistent with pecking order theory or trade off theory.

- Liquidity: the result is consistent between pooled OLS and fixed effectsmodel A negative relationship between liquidity and total and short term debt isfound while long term debt is insignificant with leverage

- Tangibility: under pooled OLS method, tangibility increase at the samedirections with long term debt and increase at an opposite direction with total andshort term debt Fixed effect method gives insignificant result for all threedependent variables

In short, learning from the above researches, many factors have effect tochanging of leverage and the relations are not the same under different countries,different technique estimations as well as different measures of leverage Thisthesis will follow the way of Gurcharan (20 10) by using one measure of leverage

to test, without following the way of Huang and Song (2002) by using 6 measures

of leverage and Zehra (2008) using 3 measures of leverage Besides 4 factors asprofitability, non debt tax shield, growth and firm size which Gurcharan (2010)employs to check for Malaysia, Indonesia, Philippine and Thailand, this thesis willcheck for two more proxy as liquidity and tangibility assets as research of Zehra(2008) due to these two proxies are important in context of Vietnamese firms.Above researches are divided into two group include Asia region and Europeregion with both developed and developing countries The results of researches inAsia region seem to have consistent while the results in Europe are mix Vietnam

is also in group of developing countries in Asia, so maybe the results found will bethe same that profitability, growth opportunities of the firm, NDTS have negativerelated to leverage and firm size have positive related to leverage Although

- -~

following the research of Gurcharan (20 10), this thesis only check for firm

specific factors, leaving out country specific factors due to limitation of data

2.4.2 Empirical evidences finding from Vietnam perspective:

Nguyen and Ramachandran (2006) have done a research about leverage of

558 small and medium sized enterprises (SMEs) over the period 1998-2001.Vietnam stock market has only been established since the year 2000, so thisresearch is certainly for unlisted firm only They conclude that leverage of VietnamSMEs increases at the same direction with growth, firm size, business risk, networking and relationship with banks In contrast with theories, they state tangibilityassets have negative effect on leverage and profitability is found to have no impact

on leverage The study concludes that to finance for new investment, SMEs dependmostly on short liabilities

Dzung et al (2012) also consider factors affecting leverage of Vietnamfirms but for listed companies They employ data of 116 firms listed on both stockexchange- Ho Chi Minh stock exchange (HOSE) and Ha Noi stock exchange(HNX) for the period 2007-2010 The research uses book value data to measuretotal, short-term and long term leverage They tests determinants of leverage atboth side - firm specific factors including profitability, tangibility, size, growthopportunity and liquidity and country specific factor as state-ownership The state-ownership is checked by using dummy variable to distinguish between firm withover and less than 50% state-owned share Difference with many researches, thisstudy uses panel data regression with GMM model Completely contradicting withresearch of Nguyen and Ramachandran (2006) about insignificant relationshipbetween profitability and leverage of unlisted firms, this research finds thatprofitability has inverse relationship with leverage for all measures of leverage andit's also the most strongly factor effect to leverage This result is consistent withpecking order theory During this period, firms likely to finance its investment withinternal sources of finance About tangibility and size indicator, the results aremixed It has no impact on total leverage while impact on short term and long

term leverage are in opposite direction, negative for short term and positive forlong term leverage Firms with fewer tangible assets do not satisfy loan conditions

of bank, so they tend to rely more on short-term liabilities as trade credit Theyalso find the positive relationship between growth and total leverage, long termleverage while effect on short term leverage is not statistically significant Theresult about liquidity which they find is negative relationship with total and short-term leverage whereas an insignificant related with long term leverage It seemsthat long term lenders are not interested in liquidity ratio of the firm The mainfinding of this research is that Vietnamese enterprises still depend much more onshort term debt and "pecking order theory better explains financing decision inVietnam than trade offtheory"

Another research about determinants of leverage of Vietnam firms isresearch of Cuong (2008) With data of 22 seafood companies in Khanh Hoaprovince, the study finds that firm size is the strongest factors affect to leverage ofcompanies and it has positive related Fixed assets have inverse related to leverageshows that these firms tend to use equity to finance for long term investment whileprofitability and growth have positive related to leverage

In short, results of some researches in Vietnam context also haveinconsistence Although research in Asia group proposes a negative relatedbetween firm growth and leverage, researches in Vietnam show a contradictionresults Most of them argue a positive related between firm growth and leverage.The main findings from the studies about Vietnam context is Vietnamese firmsmostly rely on short term debts

2.5 Chapter remarks:

This chapter presents four mam points Firstly, some main conceptsconcerning content of the thesis have been explained as capital structure, leverage,optimal capital structure, agency cost and bankruptcy cost Secondly, two maintheories which the thesis bases on are also presented as trade off theory andpecking order theory Thirdly, some factors affecting leverage which are

withdrawn from many previous researches such as profitability, firm size, NDTS,growth opportunities, tangibility assets and liquidity are forecasted about its effectaccording to point of view of theories one by one Lastly, this section also includessome empirical evidences from the world as well as Vietnam context The results

of these researches are not the same at different countries, different measures ofleverage as well as different techniques of regression These variables may bepositive or negative significant or even more it shows an insignificant result.Learning from the result on the world and Vietnam context, this thesis proposes apositive related between firm size, tangibility assets and firm growth with leverageand a negative related between profitability, NDTS and liquidity with leverage

CHAPTER III: RESEARCH METHODOLOGY

This chapter will be divided into three sections The first is the introduction

of data source as well as description of variables includes dependent andindependent variables Next, the description of different estimation methods will

be introduced including: Ordinary Least Square, Fixed Effect and Random Effects.Hausman test to choose between Fixed and Random Effect method are alsopresented in this section Finally are some hypotheses which will be tested in thestudy

3.1 Data and variables description:

3.1.1 Data:

The study consist data on balance sheet of 35 Vietnam processing andexporting enterprises during the period 2009 to 2011 including both listed andunlisted enterprises Data of 20 listed enterprises which be taken from either the

Ho Chi Minh city stock exchange (HOSE) or the HaNoi stock exchange (HNX)and 15 unlisted enterprises will be tested 3 5 enterprises for period 3 years willcreate 105 samples for the study

Due to investigating both listed and unlisted enterprises, this study will base

on book value of data only

3.1.2 Description of variables:

(a) Dependent variable: leverage (L VE)

Learning from the research of Rajan and Zingales (1995), dependentvariable "leverage" employed here is the ratio of total liabilities over total assets.According to them, trade credit should be included in leverage since it alsorepresent for debt which firms have responsibility to pay When the creditorsconsider of supplying debt for the firm, they not only give attention to long termdebt but also care about current debt and trade credit due to it will reduce debtcapacity of a firm (Huang and Song (2002)) Jason and Eric (2009) also have the

same idea in using the ratio of total liabilities to total assets to measure for

leverage

Many researches use three ways separately to measure leverage includesshort term debt, long term debt and total debt (Zehra (2008), Dzung (2012)).However, due to most of debt of Vietnam SEAs are short term debt, the ratio oflong-term debt is rather small, the study will combine both short term debt andlong term debt

(b) Independent variables for determinants ofleverage:

•!• Firm size (LGSIZE):

Size of the firm may be estimated by some indicators such as logarithm ofrevenues sale (Gurcharan (2010), DeJong et al (2008), Amarjit (2011)), logarithm

of total assets (Joshua (2008), Deesomsak (2004), Huang and Song (2002))

This study chooses natural logarithm of assets to measure for firm size

•!• Firm Growth (GRO):

Firm growth is measured by different ways by different researchers, mgeneral two scales for measurement are percent change of total assets (Liaqat (20

11), Akinlo (20 11) or percent change of total revenues sale (Pramuan et al (2003)

or Mehdi et al (2011))

In this study, growth will be measured by percent change of total assets

•!• Non-debt tax shield (NDTS): is measured by ratio of depreciation to totalassets (as Wald (1999), Titman and Wessels (1988), Zehra (2008)) Due to ratio ofdepreciation is data of whole year while total asset is data at a period, so to calculatefor this ratio, the study will use the average total assets of the beginning and the end of

a given year

•!• Profitability (PRO) is measured by the ratio of earning before interest andtaxes (EBIT) to total assets following research of Huang and Song (2002), Ntogwa(20 12) Total assets in this ratio are also average total assets of a given year

•!• Tangibility (TANG- value of tangible assets) is measured by ratio of fixedassets to total assets It has been argued by Bevan and Danbolt (2000), Husni and Ali(2010)

•!• Liquidity (LIQ) is measured by current asset over current liabilities as DeJong et al (2008)

All proxies in the model are compared with total assets to assure aboutlogical of the model

Table 3.1: Variable, description and its expected sign

Dependent variable

assets

Independent variables

-liabilities

3.2 Methods of estimation:

The relationship between leverage and other factors as profitability, size,tangibility assets, growth opportunities, non-debt tax shield and liquidity will beexamined with quantity method by using different empirical techniques of panel

regression analysis data which is a combination of times series data and cross-sectiondata Using panel data has many advantages such as it can controls for individualheterogeneity, supply more degrees of freedom as well as more variables Moreover, itstill can reduce problem of multicollinearity which is the problem time series datamethod usually meet (Baltagi (2001)) Comparing with only cross section or timesseries method, panel data method seem the better method to use for analyzing withboth cross-section and time series data and this method is also usually used by manyresearchers in term of investigating determinants of leverage.

Next is the description about Ordinary Least Squares (OLS) method and thenabout Fixed Effects (FE) and Random Effects (RE) In order to choose which model ismore suitable between FE and RE, a test which is called Hausman test will be done

3.2.1 Ordinary Least Squares (OLS) estimator:

Sample of the thesis include both data across firms and overtime, so using ofpanel data is more appropriate to study the dynamics of change First, ordinary least square estimator, the most simple method will be applied

Given a model as follows:

Yit = ao+X\t~+ ui,t (*)

Where Yit is dependent variable, Xi,t is a (K-l)xl vector of exogenousregressors and ui,t ~ N(O,cr~:2) is random disturbance The error term ui,t in OLSmethod is assumed to be independent and normally distributed with zero mean andconstant variance It means:

cr/~0

E(ui,t Uj,s) = 0, i:;t:j, tt:s,

E(Xi,t Uj,s) = 0 V i,j,t,s

Panel OLS may increase sample size, however with data contained observations

on the same cross-sectional units over several time periods, maybe there is crosssectional effects on each firm or on a set of group of firms The