Determinants affect the access to formal informal credit and its impact on sales growth

By employing bivariate probit model and instrumental variable method to solve endogeneity problem, this study finds that while formal finance plays significantly positive role in improvi

IMPACT ON SALES GROWTH

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

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

By

HOANG QUYNH TRANG

Academic Supervisor:

DR TRUONG DANG THUY

HO CHI MINH CITY, Nov 2016

DECLARATION

I declare that this thesis is entitled “Determinants affect the probability of access to formal- informal finance and its impact on sales growth” is submitted by me in fulfillment of the requirements for the degree of Master of Arts in Development Economics to the Vietnam – The Netherlands Programme (VNP)

This thesis is my original work and under the guidance of my supervision and acknowledgement has been made in the text to all materials used

Hoang Quynh Trang

ACKNOWLEDGMENTS

First of all, from the bottom of my heart, I would like to say thanks to my supervisor, Dr Truong Dang Thuy, who always listens to my ideas, discusses and gives me helpful advices for my thesis He is also the first teacher who taught me how to process and analyze the data with econometric tools Although his time is quite limited, he always reminds and encourages

me to do this thesis harder Again, I deeply say thanks to Dr Truong Dang Thuy, without you

it will take very long time to complete my dissertation

I would like to express my special thanks to Professor Nguyen Trong Hoai, Dr Pham Khanh Nam and Dr Le Van Chon and all lecturers in Viet Nam- Netherland Programme where I received useful knowledge throughout two years, which will positively support for

ABBREVIATIONS

CIEM Central Institute Economic Management

GDP Gross Domestic Product

GSO General Statistic Office

MPI Ministry of Planning and Investment

IV Instrumental variable

OLS Ordinary Least Square

SMEs Small and Medium Sized Enterprises

ABSTRACT

Do different sources of finance make different sales growth? In this study, I used a dataset from Small and Medium sized enterprises (SMEs) in Vietnam in 2013 to explore the determinants affect the probability of access to formal and informal finance and its impact on growth rate By employing bivariate probit model and instrumental variable method to solve endogeneity problem, this study finds that while formal finance plays significantly positive role in improving firm performance, informal finance goes on opposite direction on growth rate Besides, in term of accessibility to official finance, firm size, receiving government assistance and good connection with banks have more advantageous than other factors Social networks such as network with banks and others also are found significantly positive effect on the probability of obtaining non-official finance More interestingly, this study also points out that young entrepreneurs are more likely to select informal sector to finance their business operation than the old ones

Keywords: formal, informal, access to financing, sales growth, Vietnam.

TABLE OF CONTENTS

CHAPTER ONE: INTRODUCTION 1

1.1 PROBLEM STATEMENT 1

1.2 RESEARCH OBJECTIVE 3

1.3 RESEARCH QUESTIONS 3

1.4 SCOPE OF THE STUDY 4

1.5 THE STRUCTURE OF STUDY 4

CHAPTER TWO: LITERATURE REVIEW 5

2.1 FIRM’S ACCESS TO FINANCE 5

2.1.1 Theoretical studies 5

2.1.2 Empirical studies 8

2.2 FINANCING CHOICES AND SALES GROWTH 14

2.2.1 Theoretical studies 14

2.2.2 Empirical studies 16

CHAPTER THREE: DATA AND METHODOLOGY 23

3.1 DATA 23

3.2 METHODOLOGY 23

3.2.1 Determinants affect firm’s access to finance 23

3.2.2 Financing choices and firm’s growth 26

CHAPTER FOUR: RESULTS 30

4.1 DESCRIPTIVE RESULTS 30

4.2 REGRESSION RESULTS 37

CHAPTER FIVE: CONCLUSION AND IMPLICATION 51

5.1 KEY FINDINGS 51

5.2 POLICY IMPLICATIONS 52

5.3 LIMITATIONS AND SUGGESTIONS FOR FURTHER STUDY 53

5.3.1 Limitations 53

5.3.2 Suggestions for further research 53

REFERENCES 54

APPENDICES 58

LIST OF TABLES

Table 2.1 Summary of theoretical studies 16

Table 3.1 Definition of variables 24

Table 3.2 Definition of variables 26

Table 4.1 Summary statistics of the sample 30

Table 4.2 Sales growth by gender and education 31

Table 4.3 Sales growth by the type of ownership 32

Table 4.4 Sales growth by region 32

Table 4.5 Sales growth by industry 33

Table 4.6 Sales growth by financing choices 34

Table 4.7 Sales growth by government assistance 35

Table 4.8 Determinants of formal/ informal finance accessibility 37

Table 4.9 How different sources of finance affect sales growth 45

LIST OF FIGURES Figure 1.1 Framework for capital structure categorization 3

Figure 2.1 The expected rate of return of lenders 5

Figure 4.1 The relationship between sales growth and some continuous variables 36

CHAPTER ONE: INTRODUCTION

1.1 PROBLEM STATEMENT

Small and medium scaled enterprises (SMEs) play a significant role in economic development Data collected from Ministry of Planning and Investment (MPI) showed that SMEs contributed about 40% to Gross Domestic Product (GDP), 30% to total export turnover, and 15% to government revenue’s contribution Furthermore, SMEs also play critical role in term of employment generation and make a job for over 60% employees

However, according to the report of MPI in 2012, return on revenue of SMEs stayed at 2.8% in 2007 and decreased to 2.34% in 2009 In addition, based on the dataset collected from General Statistics Office (GSO) of Vietnam in 2011 showed that SMEs represented just over 43% aggregate gross income of all endeavors and just under 14% aggregate profit before duty These numbers are quite modest compared to SMEs’ contribution to economy

Capital is one of the most important input factors of any type of manufacturing and business operation Furthermore, data about SMEs collected by Central Institute Economic Management (CIEM) in 2013 stated that 741 of more than 2500 enterprises accessed to financing sources, in which firms obtained formal loan accounted for roughly 72%, while this number was approximately recorded at 20.24% and 7.83% for loan from informal sector and combination of two sources, respectively In addition, out of 662 firms applied for bank loan, there was only 158 firms reported that they experienced problem when getting a loan This difficulty of SMEs mainly resulted from administrative procedures in obtaining clearance from bank officials and lack of collateral

The positive effect of formal finance on firm performance is proved in many empirical studies For example, Ayyagari, Demirgüç-Kunt, and Maksimovic (2010) utilized cross-section data from Chinese firms in 2003 to explore whether there exist the positive role of informal finance on firm performance These authors claimed that whereas financing from official sector supports higher firm performance, non-official finance does not Similar finding

can be found in the study of Saeed (2009) where the author pointed out that formal financing inserts a positive impact on firm performance while informal abates the outcome In practice, enterprises could not get money from standard mechanism at the level of 100%, sometimes they have to seek finance from another sector such as their family, partners, relatives or even

in the black market to have enough money to finance their business operation

Along with official source of fund, non-official finance in some empirical studies is also considered as an alternative channel to fill in the demand for finance of firms, especially in developing countries where the weakness of financial and legal system exist This can be seen clearly through Allen, Chakrabarti, De, and Qian (2012)’s study when they do research for small and medium sized enterprises in India The authors claimed that non-legal governance mechanisms dominate legal mechanism in solving disputes, overcoming bureaucracies and fostering firm’s performance In another study, such as Degryse, Lu, and Ongena (2013), the author suggest that the combination of formal and informal fund is an optimal choice for small firms, especially in emerging countries, where asymmetric information is pretty severe

There are many sources to finance firm’s business operation such as from retained earnings, issuing stocks, borrowing from financial institutions or even combination two-three that of sources Due to an important role of capital, many empirical studies worldwide investigate the relationship between so-called capital structure and firm performance The structure of capital describes the way that firms raised their needs to establish or expand their business activities In other words, capital structure is defined as “the relative amount of debt and equity that firms need to finance” (T D K Nguyen & Ramachandran, 2006) Figure 1.1 presents the categorization of capital structure for further information

Figure 1.1 Framework for capital structure categorization

While there are many empirical studies explore the connection between internal- external finance or bank credit and credit constraint on firm performance, very few papers evaluate the impact of formal and informal finance on sales growth

1.4 SCOPE OF THE STUDY

This study will investigate the determinants affect the access to formal and informal finance and its impact on SMEs’ sales growth by using small and medium sized enterprises in Vietnam in 2013 There is a lot of information in this type of dataset, however I just collect the data regarding to the access to formal and informal finance

1.5 THE STRUCTURE OF STUDY

There are five chapters in this study Chapter 1 describes problem statement, research objective, and scope of the study Chapter 2 discusses the related literature and previous empirical studies Data and methodology are illustrated in chapter 3 For the empirical results

we can find in chapter 4, and chapter 5 concludes the research

CHAPTER TWO: LITERATURE REVIEW

Chapter 2 reviews some related literatures as well as empirical results derived from previous studies in the similar considered problems The first section presents theories explaining access to finance of firms, as well as a review of empirical studies about this issue The review on theories and empirical studies on the impacts of financing choice and sales growth are presented in the next section

2.1 FIRM’S ACCESS TO FINANCE

2.1.1 Theoretical studies

2.1.1.1 Lenders’ behavior in asymmetric information

Asymmetric information is construed as one has more advantageous information than the others in the same transaction The behavior of lender in asymmetric information should be explored firstly to understand the mechanism why some applicants receive loan, but some do not

According to Jaffee and Stiglitz (1990), the expected rate of return of lenders is represented as a function of quoted interest rate, and it has concave shape as in Figure 2.1:

r1 Quoted Interest rate

Expected rate of return

r*

Figure 2.1 The expected rate of return of lenders

where vertical and horizontal axis are represented by expected rate of return of lenders and interest rate, respectively r* is the optimal interest rate

There exists an excess of demand for credit, it is undoubted that lenders will increase their loan interest rate, which is seen as the price has to be paid for borrowing money According to the law of demand and supply, an increase in price will lead to a decrease in quantity demanded and a rise in quantity supplied, until market clearing is restored However, at the optimum quoted interest rate (r*), there is no incentive for lenders to raise the rate of interest This is because of two reasons: adverse selection and moral hazard, which are discussed hereafter

♦ Adverse selection effect (before the transaction)

If the rate of interest increases, safer applicants, who need fund to finance projects with low risk and low return, will be unable to pay at high interest rate, and dropped out of the market In contrast, riskier applicants, who have high risk and high return, are only remaining

in the market This will lead to an increase in risk of default on loan (a promise of repayment

is broken), so a decline in lenders’ expected rate of return (Stiglitz & Weiss, 1981)

♦ Moral hazard effect (after the transaction)

The moral hazard represents the situation in which individual or cooperation (borrowers) does not have an incentive to do as what they committed in the contract after the transaction had been carried out Accordingly, in order to offset the high interest rate, the borrowers are more likely to undertake risky projects, this will lead to a higher risk of default on credit, and a lower lenders’ expected rate of return

As those problems mentioned above, although facing the excess demand for credit, lenders would not respond to market by increasing the rate of interest for borrowers And in this situation, although there are people who are willing to pay at a higher level of interest, lenders remain unchanged at the level of optimal interest rate, and limit loan for borrowers In other words, firms are constricted credit in asymmetric information

2.1.1.2 Firm’s financing decision

● Trade-off theory

According to this theory, small firms will choose how much debt finance and equity finance that they utilize to balance between benefits and costs Using debt financing has more advantageous than equity financing This is simply because interest payment is tax-deductible while income from equity is suffered from corporate tax While using more debt, along with benefit from tax-shield, firms have to face an increase in financial risk that makes debt become not cheaper, as compared to equity financing An increase in debt finance will lead to an increase in marginal cost and a decline in marginal benefit In order to optimize firms’ overall value, firms should balance between the benefit of tax-shield and the cost of financing distress

In other words, they have to trade-off and choose how much debt and equity finance to get balance

Besides, follow trade-off theory, firms have more tangible assets and more profitable are predicted that are likely to obtain high debt ratio, while equity financing is anticipated for those firms with high intangible assets

● Pecking order theory

In this theory, borrowers, especially small firms, tend to utilize internal finance firstly, followed by debt, and finally external equity is considered (Myers & Majluf, 1984) These priorities reflect cost of various financing sources where external equity’s cost is higher than the others As a result, external equity may not the optimal choice towards small enterprises,

this is because of some reasons as follows: i/ Stock market flotation is very expensive and the

initial public offerings are underpricing, these things create disadvantages for small

enterprises ii/ For a given level of risk, higher cost of equity is subjected by small firms as

compared to large counterparts due to small firm effect when they obtain stock market

flotation iii/ A wider share ownership is one of the prerequisites for stock market flotation

However this may lead to loss of control of initial managers or the problem of takeover possibility

● Agency theory

Agency theory claimed that there is moral hazard and adverse selection problem exists in contractual arrangements between small firms and lenders These problems are likely to be more severe and the costs to solve them by using monitoring and bonding are very expensive for small enterprises Another way to solve these problems is using more collateral in lending

to small firms to avoid agency problems and help banks to align interests (Chittenden et al., 1996)

Above is three important theories relating to financing decisions, and throughout many empirical evidences, pecking order and agency theories seem to be supported over trade-off theory (Chittenden et al., 1996)

2.1.2 Empirical studies

In order to investigate the determinants affect finance accessibility, most of papers primarily focus on the access to formal loan rather than informal loan, and very few papers examined both financing sources simultaneously In their regression models, the dependent variables such as formal and informal choices are usually binary variable, which is a function

of firm age, firm size, firm ownership, owner’s education, owner’s gender, the age of manager, collateral, and networks Using firm-level data, and mostly employed logit model (there is one paper used multivariable probit model) to find down factors affect the probability

of access to these financing sources

Regarding to determinants of formal finance accessibility, most of studies claimed that firm age, firm size significantly positive effect on getting official loan, while these variables are found mixed results in the informal choice’s equation Especially, a paper indicated that the older firm is, the higher access to informal finance (Essien & Arene, 2014) However, there also exists perspective claimed that young firms whose owners are female do not accumulate enough social capital, and assets (Akoten, Sawada, & Otsuka, 2006) So, it is quite hard for them to obtain bank loan, and non-official loan is an alternative channel for these types of enterprise in this situation With regard to the size of firm, Essien and Arene (2014)

stated that firm size does not affect informal credit accessibility The type of ownership also affects formal credit accessibility In detail, private firms are less likely to access to official credit, as compared to state owned firms (Kumar & Francisco, 2005)

With regard to owner’s characteristics, many empirical studies found that better educated owners could be expected to access credit more easily (Nikaido, Pais, & Sarma, 2012) However, according to Rand (2007), this relationship goes on the opposite direction The nexus between gender and formal credit accessibility is still ambiguous For example, while many empirical studies stated that male-headed firms are more advantageous in getting bank loan, relative to their counterparts, Yaldiz, Altunbas, and Bazzana (2011) suggest that formal credit is more available for female-owned enterprises Some papers suggest that there is no difference between man and woman in access to official credit (Fatoki & Asah, 2011; Harrison & Mason, 2007) With regard to gender and informal credit, due to lack of assets, female-owned firms are more likely to borrow from non-official sector, as compared to their counterparts (Akoten et al., 2006) Another variable is owner’s age and the effect of this variable is ambiguous too Many studies point out that young owners are usually considered as inexperienced, lower level of social capital, more likely to borrow from friends and family rather than the other institutions On the other hand, following Vos, Yeh, Carter, and Tagg (2007)’s discussion, young managers tend to use bank loan, overdraft rather than older ones Besides, collateral plays significantly positive role in improving the probability of formal credit accessibility throughout many papers The last is networks dummy variable The effect

of this variable on credit accessibility is opaque as well Most of papers indicated that the positive relationship between good connection with lenders and credit accessibility However, the network with government officials is found negatively affects bank finance (Le, Venkatesh, & Nguyen, 2006)

To sum up, based on theoretical literature and many other empirical studies, factors affect credit accessibility can be divided into 3 sub-groups including owner/manager’s characteristics, firm’s characteristics and networks

Owner’s age, education and gender can be added in the regression as proxy variables for owner’s characteristics

Regarding to the age of entrepreneur, Akoten et al (2006) claimed that junior managers

are young and usually considered as inexperienced in social interaction leading to lower level

of social capital, so they have disadvantage in accessing to bank loan The authors also point out that young owners, the age from 21 to 35, are more likely to borrow from friends and family rather than the other institutions This is also consistent with Omboi and Wangai (2011)’s finding They argued that senior managers are able to accumulate assets utilized as collateral by banks, so they are easier to access to formal financial institution than junior managers Nevertheless, the opposite relationship is found in Vos et al (2007) The authors argued that young owners are more likely to use external finance, such as bank loan, overdraft, than old entrepreneurs

In term of the second owner’s characteristic, the effect of education on ability to access

institutional credit is found mixed Educational attainment is seen as an importantly positive indicator in demand for institutional credit of small firms by Omboi and Wangai (2011) In line with this statement, Nikaido et al (2012) claimed that firms with better educated owners could be expected to access credit more easily than those with less educated This is because higher educated owners tend to be less difficult in following application procedure than the others In addition, managerial skills in finance, marketing production and international business areas are often mastered by higher educated owners and these skills and knowledge are attributable to firms’ performance (Kumar & Francisco, 2005) However, in Rand (2007)’s study, owners with better educated have significantly negative impact on the probability of credit accessibility This is because owners with higher educated will refrain from applying for bank loan if they know their bank application is potentially higher rejected

Owner’s gender is also considered as one of the factors affects the ability to access to

bank credit Female managers face many difficulty in obtaining bank loan due to lack of assets, which are required by formal institution, so they depend more on micro-finance sector (Akoten et al., 2006) Omboi and Wangai (2011) claimed that demand for institutional credit

of females owned firms are found to be less than males owned firms due to some of cultural norms Furthermore, according to Nikaido et al (2012)’s argument, due to underprivileged status, females owned firms are more likely to be excluded in credit accessibility, compared to males and this argument is in accordance with Tonin, Dieci, Ricoveri, Foresi, and Hansohm (1998) These authors also pointed out that access to bank credit of females are more difficult than their counterparts, which are largely due to lack of collateral of females-headed enterprises In contrast, there exists a positive relationship between females and bank credit, or even no difference between two genders in applying for institutional loan More specifically, obtaining bank loan becomes easier for females, so they depend less on the loan coming from non-legal mechanism (Yaldiz et al., 2011) Moreover, according to the finding of (Fatoki & Asah, 2011; Harrison & Mason, 2007), there is no significant difference between woman and man in accessing to institutional credit

One of the most important indicators that affect credit accessibility as well as firm

performance is firm size Depend on the purpose of research, and the availability of data, firm

size can be proxied by total employees (Becchetti & Trovato, 2002; Klapper, Sarria-Allende,

& Sulla, 2002), total sales (Degryse et al., 2013; Rahaman, 2011) or total assets (Allen et al., 2012; Essien & Arene, 2014; Rahaman, 2011) Bigsten et al (2003) argued that while larger firms are more likely to obtain bank loan, it goes opposite direction for smaller ones Beck, Demirgüç‐Kunt, and Maksimovic (2005) utilized firm-level database for 54 countries, pointed out that the size is really matter to credit accessibility of firms In particular, small and medium scaled enterprises face greater disadvantages, relating to financial, legal and corruption obstacles than large enterprises The authors also argued that these obstacles will adversely affect firm performance, especially small and medium sized enterprises This claims are similar to Saeed (2009)’s findings Accordingly, firms with small and medium size are more financially constraint, they could not provide necessary information as well as post collateral as bank required, as a result they depend more on non-market source of finance than their counterparts

Firm size is also a matter in Okura (2008)’s study The author showed that informal credit, internal fund and equity becomes increasingly important when firms become smaller For example, formal institutions lend more than 40 percentage of working capital to larger firms, whereas this figure is 16 percentages to smaller enterprises Accordingly, there exist a positive relationship between bank accessibility and firm size Several reasons are explained by Kumar and Francisco (2005) why small firms face greater difficulty in accessing bank credit than larger firms Firstly, due to market imperfection, in term of informational opacity, poor quality and provision of financial information are often found in small firms These obstacles make creditors hardly to believe firm’s quality or the quality of firm’s project (Binks & Ennew, 1996) In addition to informational opacity, small firms are considered riskier than larger firms

in term of unsustainable earnings This is because these type of firms have less opportunities

to diversify their output as well as client base (Klapper et al., 2002) Secondly, information asymmetry, including adverse selection and moral hazard, is another mentioned issue Accordingly, information asymmetry leads to credit rationing of commercial bank (Stiglitz & Weiss, 1981) In order to alleviate this issue, posting collateral is solution (Angelini, Di Salvo,

& Ferri, 1998) However, while larger firms often own more assets which can be provided as collateral in bank’s perspective, small firms face more difficult than their counterparts (Angelini et al., 1998)

The second factor affects credit accessibility of firms is firm age According to Levenson

and Willard (2000)’s argument, smaller and younger firms are more likely to be constrained than larger firms Older firms are found to have more access to bank credit due to their accumulated social capital and credit history overtime that bank lenders depend on them to make loan decisions (Musamali & Tarus, 2013) Fatoki and Asah (2011) indicated that there is positive relationship between the older firms and access to bank credit More specifically, the older firms are, the greater credit accessibility that firms have, while the authors do not find any evidence supports young firms, between 1 to 4 years old, to access formal financial institution

Together with firm size and firm age, many empirical studies also consider type of firm, industry and region dummy variable as one of the firm’s characteristic and furthermore, receiving assistance from the government is also investigated when one explores which factors affect the ability of getting bank loan

Regarding to type of firm, Demirgüç-Kunt and Levine (2005)’s result pointed out that

while the access to debt finance of state and foreign-owned firms become much more easier, family-owned firms, sole proprietorship (private) and partnership enterprises face greater financial obstacles According to Kumar and Francisco (2005), firms owned by government are more likely to access to bank credit, as compared to private firms or firms owned by

private foreign About the relationship between industry and credit accessibility, Tonin et al

(1998) claimed that the nexus between enterprises targeted for credit and decisions of loan providers reflects economic environment approached by creditors Accordingly, the main sectors, such as retail, trade and service sectors, are more beneficial from formal finance,

relative to the others Another factor affects formal credit is region dummy variable For

example, Tonin et al (1998) claimed that due to lack of financial services, people in sub-urban areas find difficulty in getting bank loan, as compared to in urban areas In addition, some creditors only serve urban areas because of the location of institution, as a result reduces the allocation of resources Regarding to the matter of region/ province and its impact on credit accessibility in Vietnam, Mai (2014) found that firms located in Red River Delta and Southeast areas are more likely to be constrained, compared to those enterprises located in the

other regions And the last is government assistance Follow N T Nguyen (2014)’s finding,

receiving assistance from the government is highly related to bank credit accessibility For example, the state will direct banks to lend money to some firms, so this variable closely supports for obtaining bank loan, but not directly affects firm’s growth

Another factor that plays important role in access to debt finance is networks, and the results are quite mixed between formal and informal finance However, in general the role of networks is confirmed positively related with credit accessibility

In the absence of collateral and face the issues coming from asymmetric information, firms could utilize their banking relationship to overcome the problem and get bank loan (Fraser, Bhaumik, & Wright, 2013) According to Akoten et al (2006), lenders are likely to depend on reputation and the social networks of borrowers when they make loan decisions Furthermore, for those firms have longer relationship with banks are more likely to borrow at lower interest rate as well as less likely to post collateral, compared to the other firms (Berger

& Udell, 1995; Uzzi, 1999) However, the finding is found differently in Essien and Arene (2014)’s study Accordingly, there is no evidence supports the relationship between networks and formal credit access, whereas the role of social capital is significantly positive effect on informal credit access

Furthermore, follow Peng and Luo (2000), using data from China, there are two types of networks including network with government officials and network with other firm’s manager According to these authors, both the former and the latter networks play a significant role in access to bank loan and improving firm performance An explanation given for a positive relationship between government officials and access to bank credit is that procedures with government organizations and banks may be reduced with this closed connection However, according to Le et al (2006)’s findings, using data from small and medium sized enterprises in Vietnam, they indicated that while the connection between manager of other organizations, friends does positively impact bank loan, the nexus between firms and government officials negatively affects bank finance It is because such relations help firms to have more opportunities to obtain government support program and get money, and these sources are cheaper than formal loan

2.2 FINANCING CHOICES AND SALES GROWTH

2.2.1 Theoretical studies

According to Modigliani and Miller (1958), in MM world, where the information is symmetric, tax is not distorted, no transaction cost and bankruptcy cost, the capital structure does not affect the value of firm and the cost of capital In particular, there is no difference between debt and equity finance, or the choice between internal and external financing could

be substituted perfectly And if the MM world’s assumptions are relaxed in Modigliani and Miller (1963), the change in capital structure is associated with firm’s value and cost of capital More specifically, an increase in debt over equity will lead to an increase in the value

of firm and a decrease in weight average cost of capital (WACC) However, although the benefit retrieved from tax-shield when using higher leverage ratio, it is not completely perfect method due to higher risk such as financial distress, bankruptcy cost, asymmetric information, and so on as possible trade-off options against the benefit of tax-shield According to which, the trade-off theory claimed that higher profitability is associated with higher debt over equity thanks to tax-shield benefit, although this perception is not at the level of 100%

Pecking order theory indicated the order of priority in the use of fund According to which, enterprises initially used internal financing source, follow debt and equity It means that if firms get more profit, they tend to use their retain earnings to finance their business operation, rather than borrow from external finance So the more profitable firms are, the less usage of their debt is

Another theory that reflects the nexus between capital structure and firm performance is agency theory According to this theory, there is agency problem between stockholders and debt holders In particular, the former want to take riskier projects and demand for higher return However, debt holders are more likely to choose less risky projects and accept lower return In the case of success of the project, an extra return will be transferred to stockholders and if the project is failure, all losses will be divided between stockholders and debt holders (Jensen & Meckling, 1976) As a result, enterprises that are more indebted take lower risk projects In addition, follow Myers (1977)’s discussion, the differences in goals between shareholders and debt holders result in firm’s underinvestment More specifically, according

to the author, firms with higher growth have more options for future investment than those with lower growth However, the more indebted firms are, the less investment opportunities firms have This is simply because firm’s wealth will effectively be transferred from firm’s ownership to debt holders As a consequence, higher growth firms are related with lower debt-to-equity ratio

However, the negative relationship mentioned above could be mitigated if firms use short term debt instead of long term debt (Myers, 1977) It is likely that there exist the positive relationship between short term debt and firm performance if firms replace long term credit by short term credit In my opinion, this preposition is more relevant in the case of small and medium sized enterprises in Vietnam where short term loan is dominant, as compared to long term loan

Table 2.1 Summary of theoretical studies

Modigliani and Miller (1963) Positive Performance impacts debt

Agency problem (short term debt) Positive Debt impacts performance

2.2.2 Empirical studies

While the majority of papers focus on internal and external financing or the effect of bank finance and credit constraint on sales growth, a few papers investigates how formal and informal credit affects firm performance Dependent variable is firm performance, and there are two methods of calculation for this variable, such as the change in logarithm of sales revenue (Ayyagari et al., 2010; Degryse et al., 2013), or the percentage change in sales revenue (Allen et al., 2012; Saeed, 2009) Both methods of calculation will be considered in this thesis later

To explore how financing choices affect firm performance, dependent variable is usually constructed as a function of financing sources (formal and/or informal and/or internal dummy), firm age, firm size, type of ownership dummy, industry dummy and credit constraint dummy According to (Allen et al., 2012; Ayyagari et al., 2010; Saeed, 2009)’s discussion,

bank finance dummy is endogenous, so they used collateral (Ayyagari et al., 2010) or firm

owned by state in any percentage (Saeed, 2009) or the total bank credit per firm disbursed by the bank in the state (Allen et al., 2012) as instrumental variables to solve the problem

Further, Yiu, Su, and Xu (2013) claimed that informal finance dummy is suffered from

endogeneity issue, and underground finance afterward dummy is used as instrumental

variable (IV) All of these things show that financing choices, for example, formal and informal source are endogenous, and using instrumental variable is the popular method to overcome the problem Based on Ayyagari et al (2010)’s discussion, due to proprietary information observed by bank, the problem of sample selection bias occurs, and Heckman two-stage procedure is employed to overcome the issue This method is also used in empirical study of Degryse et al (2013) According to the authors, dataset is used to run regression less than full observations, so there is sample selection bias However, the authors do not mention the endogeneity problem in their study Throughout some papers mentioned above, the effect

of formal and informal credit on sales growth is ambiguous For example, regarding to the impact of formal source on firm performance, Ayyagari et al (2010) and Saeed (2009) claimed that official credit plays significantly positive role in improving firm performance Whereas Allen et al (2012) stated that they do not find any evidence that support to the positive relationship between standard financing sector and firm performance

That finding is also similar to (Yiu et al., 2013) when they investigate the effect of alternative financing channel on sales growth With regard to the effect of non-standard financing source on firm performance, on the one hand, the positive role of this type of fund is found in (Allen et al., 2012) and (Yiu et al., 2013) On the other hand, this role is negatively impact on firm performance (Saeed, 2009) or there is no evidence support the effect of non-official financing source on outcomes (Ayyagari et al., 2010)

For more details, the nexus between financing and growth has been investigated by many researchers, and empirical studies worldwide claimed that the underdeveloped financial system or the limitation of credit accessibility is one of the main reasons that constraints firm performance, particularly in small-medium sized enterprises For example, Beck et al (2005) used the cross-countries dataset to investigate the effect of financial, legal constraints and corruption on firm performance They indicated that financial constraints adversely affect growth rate of firm, particularly small firms

Rahaman (2011) utilized firm-level panel data from 1991 to 2001 to investigate whether the heterogeneity in firm’s growth could be explained by the differences in financial structure and in the degree of access to internal and external funding Dependent variable is measured

by the change in logarithm of employment, which is considered as firm performance It is a function of lag of growth, firm characteristics and financing sources Internal fund and bank credit accessibility are introduced as different sources of finance, and all of them are measured

by amount According to Rahaman (2011), there exists some econometric problems such as endogeneity issue due to a set covariate financing source correlated with the error term, autocorrelation due to the lag of growth, and the time period is smaller than the number of observations In order to solve these problems, different GMM estimation is the most appropriate method The results suggest that internal and external financing source is not perfectly substituted and the cost wedge between them gives a rise to external financing constraint Internal financing source plays significant role on firm performance, however, this effect reduced with an increase in access to bank credit Although financial constraint significantly affect firm performance, an alleviation of credit limit help firms rely less on internal finance and switch to external The latter source of finance is considered as primary source to foster firm performance and this pattern of transition is pronounced in small firms Using cross section data for four countries include Vietnam, Lao, Philippines and Indonesia in 2009, Shinozaki (2012) investigate the impact of external fund on SMEs growth Dependent variable is denoted by total annual sales value which is a function of a firm size dummy variable, the value of approved loan and lines of credit which is considered as external financing, and an interaction term between them Simply apply OLS method to run regression, and the result suggests that formal financing channel accelerate SMEs growth More specifically, one percentage increase in bank credit accompanied with 0.4 percentage point increase in sales growth for large company, and this figure expand to 0.68 percentage points in SMEs at 1% significant level

Becchetti and Trovato (2002) determined factors affect SMEs growth and evaluated the role of external finance on firm performance by using panel data (including 4000 Italian SMEs

from 1989 to 1997) In this study, the left hand side is the employment growth rate, which is a function of firm size, firm age, total amount of ownership, leverage, a set of dummy variable such as sector, industry, region, government subsidy, export and rent variable External financing source is denoted by ration binary variable, equal to 1 if firms requested but do not receive any bank loan Employing OLS method, the authors pointed out that SMEs growth is not only affected by firm size and age, but the limitation of bank credit as well Moreover, small firms have higher potential average growth rate than the others However, this potential may be reduced if bank credit constraints are available

There are two contradicting views on how different financing sources affect firm performance On the one hand, the predominant views claimed that significantly positive role

of formal financing sector contributes to higher enterprises growth, while there is no evidence

or the findings are mixed for informal one On the other hand, some papers argued that in transition economies where the lack of professional financing market and the limited of credit accessibility of small firms are occurred, informal finance is seen as an alternative financing channel for small firms to seek recourse (Jain, 1999)

Regarding to the views that argue for the positively important role of formal finance, for example, Ayyagari et al (2010) employed 2400 Chinese firms across 18 different cities in

2003 to investigate whether informal source is a perfect substitute channel and supports firm performance faster than formal finance In this study, dependent variable is denoted by the log change in firm sales, which is a function of bank dummy, access to bank dummy, self-financing dummy variable, and a set of other dummy variables such as firm age, firm size, type of ownership, cities, industry and competition In order to evaluate the impact of formal fund on firm performance, according to Ayyagari et al (2010), bank dummy is subject to endogeneity problem and due to proprietary information observed by bank, sample selection bias occurs The authors used collateral as instrument variable and employed Heckman two-stage procedure to overcome these problems The results suggest that formal credit plays positive effect on sales growth To compare the role of formal and informal in improving firm performance, the authors divided into two panels for each source of finance and employed

OLS method regardless of endogeneity problem between formal source and sales growth Again, the author claimed that while formal finance plays a significantly positive role in promoting higher firm performance, there is no evidence for informal financing source In addition, non-standard financing mechanism is not effective channel to substitute formal

sector for making higher firm performance, and the contribution of Reputation and

relationship to fastest growing firms in less-developed countries may be overestimated by

Allen et al (2012)

Saeed (2009) utilized firm level data from World Bank Enterprise Survey (WBES), comprising 1539 Brazilian small and medium scaled enterprises (SMEs) from 2000 to 2005 to find down the nexus between different financing sources and growth The left hand side of the estimated equation is percentage change in fulltime employment or sales in the last three years, which is a function of the amount of formal, informal, and internal finance Besides, the index of financial liberalization, financing constraint dummy variable are also inserted as explanatory variables into the model According to Saeed (2009), formal financing is endogenous variable, thus firm ownership defined by firms owned by state in any percentage

is used as instrumental variable and employed 2SLS The authors state that formal financing source plays an important positive role on enterprise performance, while informal fund goes in the opposite direction Besides, internal finance is considered as one of the most critical indicators, which contributes to the improvement in firm performance While financial constraint has a negative impact on firm performance, an increase in financial liberalization index helps to enhance this situation

Together with the important role of standard finance mechanism on supporting firm performance, non-legal mechanism is considered as alternative financing channel in imperfect information market Allen et al (2012) employed aggregate country-level data and firm-level data from Prowess database in India from 1996 to 2005 to investigate whether access to bank credit results in higher firm performance Dependent variable is denoted by percentage change

in sales revenue, which is a function of firm age, firm size, industry dummy, firms publicly listed dummy, and lagged bank finance dummy Follow Allen et al (2012), bank finance

dummy is endogenous, thus 2SLS procedure with two instrumental variables including the number of bank branches per firm in the state (in a particular year) and Total bank credit per firm disbursed by the bank in a state are used to overcome the problem However, the authors

do not find any evidence that support the positive relationship between formal financing sector, backed by legal system- and firm performance Instead, non-standard financing sector, backed by non-legal system, is preferred form of fund in India

Using the dataset comprise of 284 Chinese private firms across 19 cities in 2006, Yiu et

al (2013) investigated the effect of alternative financing channel such as underground finance

on sales growth Dependent variable is return on asset represented as firm performance, which

is a function of underground finance dummy, formal finance dummy, firm age, firm size, leverage, industry dummy, operating cash flow, and the interaction term between provincial marketization index and underground finance According to Yiu et al (2013), underground

finance dummy is suffered from endogeneity problem, so underground finance afterward

dummy is used as instrumental variable and employed 2SLS to solve the problem In addition, follow Yiu et al (2013), the dataset just comprise of private firm data, so the issue of sample selection occurs The authors applied Heckman two-stage procedure to overcome this problem After all, the results suggest that in transition economies likes China, while non-market source of finance, namely underground finance and trade credit, is significantly positive effect on higher growth rate of private firms, formal financing channel is not Moreover, the crucial role of underground finance is clearly portrayed in provinces with less government assistant, less credit marketization and lack of economic development of non-state

The same finding is also found in Degryse et al (2013) There are three research questions arisen from these authors, such as i) funding from informal source is higher connected with small firm’s growth, ii) co-funding, specifically, formal and informal finance,

is an optimal choice for small firms to enhance growth rate, and iii) in the case of existence, a minority proportion of informal source is closer related to sales growth of small firms The dataset comprises of 3837 Chinese firms across 108 cities in 2005 used to find

co-down the answers Dependent variable is denoted by the log change in sales, which is a function of formal dummy, informal dummy, firm age, firm size, type of ownership dummy, province and industry dummy Employing Heckman two-stage procedure, the authors argued that informal finance has positive effect on sales growth of small firms, but negative impact on large firms In addition, co-funding is an optimal choice for small firms to foster sales growth, and this type of finance does surmount the difficulty of bank’s lending in information asymmetric market

CHAPTER THREE: DATA AND METHODOLOGY

In this chapter, data source and the methods of analyzing the research problems are presented We first present the data source Then we present the bivariate model for analyzing the choice of formal and informal finance After that the model for analyzing the impacts of financing choice on firm performance is presented The method addressing endogeneity in this model is also discussed

3.1 DATA

The main data source of this research comes from the survey of Small and Medium sized enterprises (SMEs) in 2013 conducted by Central Institute Economic Management (CIEM) The survey deeply interviewed more than 2500 small and medium firms, which are located in

10 regions, including Phu Tho, Ha Noi, Ha Tay, Hai Phong, Nghe An, Quang Nam, Khanh Hoa, Lam Dong, Ho Chi Minh, and Long An In which, Ho Chi Minh City (636 observations), Nghe An (358 observations), Ha Tay (345 observations), followed by Ha Noi and Phu Tho with 285 and 262 observations, respectively In addition, majority of observations in this dataset is household establishment with the number of employees is less than and equal to 10 people Dataset refers to many firm’s characteristic aspects, however, I just focus on the information about the access to formal and informal finance According to the data, over 2500 enterprises sample, there are 533 firms access to formal loan while these figures are 150 and

58 enterprises that obtain loan from informal and both source of fund, respectively

3.2 METHODOLOGY

There are two research questions deeply explored in this thesis:

1 Which determinants do affect the probability of access to formal and informal financing source?

2 How different sources of fund impact on sales growth?

3.2.1 Determinants affect firm’s access to finance

Regarding to research question 1, we have two equations as follow:

Formal i = α0 + α1 Firm age + α2 Firm size + α3 Government assistance + α4 Networks + α5

Age + α6 Gender + α7 Education + α8 Ownership + α9 Industry + α10 Region + ε1 (1)

Informal i = α0 + α1 Firm age + α2 Firm size + α3 Government assistance + α4 Networks + α5

Age + α6 Gender + α7 Education + α8 Ownership + α9 Industry + α10 Region + ε2 (2)

Definition of each variable in equation (1) and (2) is reported in Table 3.1, in which networks comprise of network with firm, bank, official and network with others

Table 3.1 Definition of variables

 Dependent variable

Formal = 1 if firms obtain formal loan,

0 otherwise Informal = 1 if firms obtain informal

loan,= 0 otherwise

 Explanatory variables

Expected sign of formal

Expected sign of informal

Firm age The number of firm’s age from

(Akoten et al., 2006; Levenson & Willard, 2000;

effect

(Bigsten et al., 2003; Essien &

Arene, 2014;

Saeed, 2009) Government

assistance

= 1 if firms receive either financial support or technical assistance, 0 otherwise

2000)

Network

with banks

The number of banks that firms

(Berger & Udell, 1995; Fraser et al., 2013)

(Le et al., 2006; Peng & Luo, 2000)

Network

with others

The number of others that

Age Age of business owner from

(Akoten et al., 2006; Vos et al., 2007)

Gender = 1 if owner is male, 0

(Akoten et al., 2006; Yaldiz et al., 2011)

Education = 1 if firm reported in specific

(Nikaido et al., 2012; Omboi & Wangai, 2011;

Rand, 2007)

Ownership = 1 if firm reported in specific

ownership and 0 otherwise

(Demirgüç-Kunt

& Levine, 2005; Kumar &

Francisco, 2005) Industry = 1 if firms located at specific

region, 0 otherwise

(Tonin et al., 1998)

Region = 1 if firms located at specific

According to Akoten et al (2006)’s discussion, making a decision to borrow money from official or non-official financing source may be occurred in simultaneous manner So I am planning to conduct bivariate probit model if the error term in two equations is correlated with each other Otherwise, probit model can be applied separately

Because there are two financing choice and I define them as binary, a bivariate probit model is appropriate In this model, there are two binary outcomes y1 and y2 (for more than two, we must apply multivariable probit), so I also have two latent outcomes y1* and y2*

y1, y2 are equal to 1 if and only if the latent variable y1* and y2* are greater than zero, 0 otherwise

In detail, model for two equations looks like

𝑦1∗ = 𝑥1′𝛽1+ 𝜀1 𝑦1 = 1 𝑖𝑓 𝑦1∗ > 0, 0 otherwise (3)

𝑦2∗ = 𝑥2′𝛽2+ 𝜀2 𝑦2 = 1 𝑖𝑓 𝑦2∗ > 0, 0 otherwise (4)

So, if the two disturbances are related with each other by 𝜌 value, we can employ

bivariate probit model (and maximum likelihood method is employed to estimate the value of

𝜌, 𝛽1, 𝛽2) (1), if not, probit model can be applied separately for the two equations

3.2.2 Financing choices and firm’s growth

The initial growth’s equation is reported as follows:

GROWTH_1 = α0 + α 1 Formal + α 2 Informal + α3 Firm age + α4 Firm size + α5 Government assistance + α6 Age + α7 Gender + α8 Education + α9 Ownership + α10 Industry + ε (5)

GROWTH = α0 + α 1 Formal + α 2 Informal + α3 Firm age + α4 Firm size + α5 Government assistance + α6 Age + α7 Gender + α8 Education + α9 Ownership + α10 Industry + ε (6)

In which, formal and informal is considered as financing choices And the definition of each variable is described in Table 3.2

Table 3.2 Definition of variables

Growth The percentage change in sales revenue (Allen et al., 2012)

 Independent variables

Formal = 1 if firms obtain formal loan, and 0 + (Ayyagari et al.,

otherwise 2010; Essien &

Arene, 2014; Saeed, 2009)

Informal = 1 if firms obtain informal loan, and 0

Firm size is proxied by total assets in billion VND

(Kumar & Francisco, 2005) (Demirgüç-Kunt & Levine, 2005)

Gov

assistance

= 1 if firms receive either financial

support or technical assistance, 0

otherwise

+

Age Age of business owner from year of

Education = 1 if firm reported in specific

Ownership = 1 if firm reported in specific

ownership and 0 otherwise

Based on Saeed (2009)’s discussion, there exist the endogeneity problem coming from the relationship between formal finance and sales growth This issue is also mentioned in Allen and Qian (2009) due to self-selection bias of owner in financing choice Furthermore, according to Yiu et al (2013)’s discussions, the relationship between firm performance and informal finance experienced endogeneity issue In short, either formal or informal fund, hereafter we call financing sources, is suffered from endogeneity If we ignore this problem and employ OLS method, the estimated results are biased and inconsistent The most popular and painful way is to find instrument variable (IV) in order to solve this issue Once we find

the best IV (strong connection with endogeneity variable, but not directly affect sales growth), then we predict results, and finally OLS method can simply be applied

More specifically, we have:

+ Endogeneity variable is financing choices

+ Networks are planned to be used as an instrumental variable

We have two assumptions for an instrumental variable (IV):

(i) IV is not correlated with the error term (ε)

(ii) IV is highly correlated with the endogenous variable

In this situation, Networks could be seen as an IV In addition, we have four types of

Network in the dataset, including Network with firms, Network with government officials, Network with banks and Network with others An increase in the number of networks can support firms have more favorable conditions to access to official or non-official finance Looking back to the case of Vietnam where the weakness of financial system exist, the information is not transparent enough in almost transactions, and the high level of corruption,

so it is an advantageous for those firms have wide social capitals For example, having good connection with banks allows firm’s loan procedures is processed quickly than other applicants Or even in the access to non-official finance, one person may be become a guarantor who pledges the debt will be paid and the loan is easily given And once firms have more money to finance their business operation, it is likely that their performance is more

optimistic As a consequence, Networks could be represented as IV This is because Networks

are highly associated with the ability of loan accessibility but not directly affect firm performance

Stage 1: Run IV regression

Formal i = α0 + α1 Firm age + α2 Firm size + α3 Government assistance + α4 Networks + α5

Age + α6 Gender + α7 Education + α8 Ownership + α9 Industry + α10 Region + ε1

Informal i = α0 + α1 Firm age + α2 Firm size + α3 Government assistance + α4 Networks + α5

Age + α6 Gender + α7 Education + α8 Ownership + α9 Industry + α10 Region + ε2

Stage 2: Run growth’s equation with predicted value of financing choices

After running these equations above, Formal and Informal hat value are predicted in Stage

1 These values are inserted into growth’s equation to solve endogeneity problem So now we have:

GROWTH_1 = α0 + α 1 Formal hat + α 2 Informal hat + α3 Firm age + α4 Firm size + α5

Government assistance + α6 Age + α7 Gender + α8 Education + α9 Ownership + α10 Industry +

GROWTH = α0 + α 1 Formal hat + α 2 Informal hat + α3 Firm age + α4 Firm size + α5

Government assistance + α6 Age + α7 Gender + α8 Education + α9 Ownership + α10 Industry +

ε (8)

The estimates from (7) and (8) are then fixed for bias due to endogeneity The next chapter presents the results from these regression analyses

CHAPTER FOUR: RESULTS

Chapter 4 first presents the overall descriptive statistics of the dataset The empirical results on the access to formal and informal finance, as well as the different effects between official and non-official financing source on firm performance will be presented and discussed

4.1 DESCRIPTIVE RESULTS

Table 4.1 Summary statistics of the sample

Dependent variable

Independent variables

 Firm characteristics

the number of firms that enterprises regularly contact with, including those firms in the same and in different business, is nearly 28 firms, ranging from 0 to 300 The overall number of banks that firms have closed relationship is 1.39 financial institutions, with min value equal to

0 and max value is 13 Other two variables reflect firm’s social networks are network with officials and networks with others Regarding to the former network, the results in Table 4.1 indicated that on average, the number of government officials that firms regularly contact with

is 1.62, ranging from 0 to 21 The latter network showed the other relationship with an average number is 5.45, ranging from 0 to 100 Following the results in Table 4.1, the average owner’s age is approximately 45 years old, with min value is 18 and max value is 69

Table 4.2 Sales growth by gender and education

Statistics of sales growth

Not completed elementary

Graduated from elementary

Graduated from secondary

Graduated from high school

-Regarding to educational attainment, the number of owners graduated from high school is dominant, relative to the other groups However, the positive mean of growth for those firms have owners with highest educational attainment just stands in the second position, as opposed

to those firms with managers who graduated from elementary (0.17% versus 2.44%) Besides, the negative growth rate is observed in firms with owners who graduated from secondary and not completed elementary school More specifically, a greater reduction in average sales growth is found with owners not completed elementary (2.74%) and maximum growth rate is

recorded at 35% while that figures are 0.56% and 181%, respectively, for firms with owners graduated from high school

Table 4.3 Sales growth by the type of ownership

Statistics of sales growth

6 Joint stock company with

is also significantly small, both coming from joint stock company with state capital, and that

figures are -1.71%, 32.06%, respectively

Table 4.4 Sales growth by region

Statistics of sales growth

REGION