Financial development and firms’ financing constraints, a study of manufacturing firms in vietnam

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICSFINANCIAL DEVELOPMENT AND FIRMS’ FINANCING CONSTRAINTS: A STUDY OF MANUFACTURING FIRMS IN VIETNAM BY VU THI KHANH MASTER O

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

FINANCIAL DEVELOPMENT AND FIRMS’ FINANCING CONSTRAINTS:

A STUDY OF MANUFACTURING FIRMS IN VIETNAM

BY

VU THI KHANH

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, MAY 2015

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

FINANCIAL DEVELOPMENT AND FIRMS’ FINANCING CONSTRAINTS:

A STUDY OF MANUFACTURING FIRMS IN VIETNAM

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

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

“This is to certify that this thesis entitled “Financial development and firms’ financingconstraints: A study of manufacturing firms in Vietnam”, which is submitted by me infulfillment of the requirements for the degree of Master of Art in DevelopmentEconomics to the Vietnam – The Netherlands Programme (VNP)

The thesis constitutes only my original work and due supervision andacknowledgement have been made in the text to all materials used.”

Vu Thi Khanh

This thesis would not have been possible without the support and theencouragement from many people I would like to make a sincere effort in portraying

my deep sense of gratitude in the form of words

I owe a debt of gratitude to my supervisor, Dr Le Van Chon for his greatgenerosity and dedication in sharing his wisdom, stimulating my critical thinking skillsand guiding me to rethink and to deconstruct my thesis topic He was also not afraid oftime-consuming to explain econometric methods as well as data processing techniques

to me Moreover, he took time to diligently review my final thesis draft and help mecorrect errors and inappropriate words usages Furthermore, I am grateful to Ass.ProfNguyen Trong Hoai and Dr Tran Tien Khai for their valuable comments andsuggestions for my concept note and thesis research design Thank to Dr Pham KhanhNam for his enthusiasm of helping me collecting data I must show my gratitudetoward all lecturers VNP who have broadened my perspectives and encouraged me tothink harder and deeper about the complexity of the world’s realities

Next, I wish to express my thank you to all my friends here at VNP Together

we have walked and struggled through this whole treasured journey of learning andshared memorable and priceless moments Then, I want to say thanks to VNP officers

as well as VNP librarian for their support of comfort lab room and study materials

Finally, I dedicate my thesis to my parents and my brother who are alwaysbesides me and never stop supporting me

: Gross domestic production: General statistics office of Vietnam: Hanoi Stock Exchange

: Ho Chi Minh Stock Exchange: The International Monetary Fund: Liquid liabilities ratio

: Private credit to GDP ratio: Small and medium enterprises: State-owned enterprises: Stock market total valued traded to GDP: The United Kingdom

: The United Nations: Unlisted public company market: The United States of America: Vector Autoregression

: The Vietnam Enterprise Survey: The World Bank

Using panel data from the Vietnamese Enterprise Survey (VES) from

2006-2012, this thesis aims to analyze the relationship between financial development andfinancing constraints of firms in Vietnam The Euler equation approach is applied tomodel firms' investment Investment sensitivity to cash-flow is employed as thevariable to test for the existence of financing constraints To control for endogeneityand firm heterogeneity, I utilize the first difference GMM estimation proposed byArellano and Bond (1991) There is robust evidence that Vietnamese manufacturingfirms face financing constraints and that financial development significantly relaxesfirms' dependence on internal funds for investment In addition, although smaller firmssuffer more severe financing constraints, their constraints are alleviated more thanthose of larger firms in the presence of financial development

Keywords: Financial development, Financing constraints, Corporate Investment

Table of contents

Chapter 1: Introduction 1

1.1 Problem statement 1

1.2 Research questions 3

1.3 The scope of the study 3

1.4 The structure of the study 3

Chapter 2: Literature review 4

2.1 Sources of investment financing 4

2.1.1 External financing 4

2.1.2 Internal financing 5

2.2 Financing constraints on firms 8

2.2.1 Definition 8

2.2.2 Measurement of financing constraints on firm 8

2.3 Financial development and financing constraints on firms 9

2.4 Conclusion 14

Chapter 3: Model specification 16

3.1 Investment modeling 16

3.1.1 Euler investment equation approach 16

3.1.2 Detecting the presence of firm’s financing constraints using Euler equation 24

3.2 Financial development measurement 25

3.3 Empirical model to evaluate the impact of financial development on firm investment 28

Chapter 4: Financial development in Vietnam 30

Chapter 5: Empirical results 37

5.1 Data 37

5.1.1 Sample and variable construction 37

5.1.2 Descriptive (Initial relationship) 40

5.2 Estimation technique: GMM dynamic panel estimation 44

5.3 Regression results 46

Chapter 6: Conclusion 51

6.1 Main findings 51

6.2 Policy implications 52

6.3 Limitations and further research 52

References 53

Appendix 60

List of tables

Table 2.1: Share of financing sources in two industries in the United Kingdom 6

Table 2.2 The proportion of financing sources according to different firm sizes 6

Table 5.1: Firm level variable definitions 38

Table 5.2: Panel data structure 39

Table 5.3 The descriptive statistics of the key variables for the whole data sample 41

Table 5.4 Median value of the key variables by different firm types 41

Table 5.5 Correlation matrix 42

Table 5.6 Mean values of the key variables by manufacturing industries 42

Table 5.7 Mean values of the key variables by years 43

Table 5.8 Regression results using first difference GMM estimation 47

Table 5.9 Scenario analysis of firm’s size on investment –cash flow 50

List of figures

Figure 4.1 The mono-tier banking system in Vietnam before banking reforms 31

Figure 4.2 Structure of the two-tier banking system in Vietnam (after May 1990) 31

Figure 4.3 Brief on the Vietnamese banking system 32

Figure 4.4 Liquid liabilities (M3) as % of GDP of some countries 33

Figure 4.5 Credit to the economy 34

Figure 4.6 Private credit to GDP of some countries 34

Figure 4.7 Stock Market Capitalization as % of GDP 35

List of appendix Appendix 1 Banking system development progress from 1990’s 61

Appendix 2 Stock market development process from 1990’s 63

Appendix 3 Boxplot chart for each variable (IK,SK,CFK) by industries 66

Appendix 4 Scatter chart for each variable (IK, SK, CFK) by industries 66

Chapter 1: Introduction

1.1 Problem statement

The study of financial development-economic growth nexus, which is originatedfrom Schumpeter (1912), has attracted many scholars’ concerns The significantpositive correlation has been proved in a number of empirical studies such as Cameron(1967), Goldsmith(1969), McKinnon (1973), Levine (1997), Levine and Zervos

(1998), Beck et al (2003), Huang (2010, 2011) etc In an indispensable research,Levine and King (1993) asserted that economic growth is influenced by the financialdevelopment in the last 10 to 30 years However, the question is in what mechanismfinancial development stimulates economic growth A prevailing research direction is

to investigate the potentially prudent effects of financial development on firms’investment, pointing to the seriousness of firm’s financing constraints, under thecountry financial development background

According to Modigliani and Miller (1958), external finance perfectlysubstitutes internal finance and firm’s investment decisions do not depend on itsfinancing choices under the strict assumption of perfect capital markets Unfortunately,perfect capital market apparently does not exist in reality According to Stiglitz&Weiss(1981) and Myers &Majluf (1984), firms have to tackle difficulties of suffering highercost of external fund due to credit risk and asymmetric information as providers ofexternal finance normally find it very costly to evaluate firms' investmentopportunities Therefore, firms have to resort to internal finance generated throughcash-flow and retained earnings But their investment is subject to fluctuations in cash-flow In other words, firms face financing constraints Rajan and Zingales (1996)argued that financial development induces more efficient reallocation of funds andmitigates the external financing constraints In better-functioned financial systems, notonly the transaction cost of saving and investing is lower but the problem ofasymmetric information is alleviated With the financial market development, some

empirical studies combine cross-country and firm panel data to investigate diversefeatures of the connection between financial development and the extent at whichfirms’ reliance on internal finance Love (2003), processing a panel data of 5,000 firms

in 36 developed and developing countries, proved that financial development lightensthe reliance of firms’ investment on internal fund, especially in small firms Using thesame data but the different method, Love and Zicchino (2006) pointed out that incountries with less developed financial systems, firms face financing constraints moreseriously This also holds at the regional level Sarno (2005) compared southern withnorthern regions of Italy and found that in less financially developed southern regions,small and medium enterprises (SMEs) harshly suffered credit constraints At provinciallevel, O’Toole and Newman (2012) investigated the effect of firms’ financingconstraints on investment opportunities in Vietnam by using the firm panel data takenfrom the Vietnam Enterprise Survey (VES) from 2002-2008 and observe thatprovincial financial development in Vietnam eases investment financing constraints

This study is similar in spirit to O’Toole and Newman (2012) in analyzing therelationship between financial development and firms’ investment in Vietnam Ofparticular interest is the role of financial development via financing constraints on thefirms’ financing investment However, there are some extensions from O’Toole andNewman (2012) Firstly, Euler equation approach is employed to model firms’investment Secondly, investment sensitivity to cash-flow is employed as the variable

to test for the presence of financing constraints, differing from O’Toole and Newman(2012) that used the ratio of investment fund by internal fund to total investmentfinance Thirdly, data in this study cover a more recent period of time from 2006-2012

1.2 Research questions

This study is to find out the impact of financial development on firms’investment By applying quantitative research to a panel data of firms in Vietnam, thisthesis aims to address two main research questions:

1 Does financial development prompt investment or relax financing

constraints of Vietnamese manufacturing firms?

2 How does financial development affect investment of different types of firms?

1.3 The scope of the study

The study will examine relationship between financial development and firms’financing investment in the case of Vietnamese manufacturing firms The firm paneldata for this research is taken from VES from 2006 to 2012 Meanwhile, the macrodata of financial development is gathered from World Development Indicators ofWorld Bank’s database

1.4 The structure of the study

The paper contains 6 main chapters Following this first chapter of introduction,chapter 2 presents the critical literature review on sources of investment financing, firmfinancing constraints, and the impact of financial development on financingconstraints Next, model specification in chapter 3 specifies the methodology to test thepresence of firm financing constraints using Euler equation approach as well as theempirical model to examine the effect of financial development on financingconstraints on firms Chapter 4 gives a brief overview of financial sector development

in Vietnam Then, the result of the study is presented in chapter 5 which contains threemain parts The first part concentrates on estimation technique, the second variableconstruction as well as data collection and the third regression results Finally, chapter

6 concludes the main findings, limitations and further researches

Chapter 2: Literature review

2.1 Sources of investment financing.

For the purpose of this study, investment refers to capital expenditure (Capex),which is measured by spending on plant, property and equipment Basically, firm’sinvestment is financed by using two sources of finance: internal finance and externalfinance Internal finance, i.e self-financing, may consist of four main components: pastprofits which are not distributed to shareholders; additional capital from the owners;depreciation and amortization Besides taking the advantages from itself whencollecting internal sources for investment, firm might also use external source fromloans of banks and other financial institutions or the issue of new equity This section isdivided into two subsections: the former covers external financing and the later internalfinancing

2.1.1 External financing

Early in mainstream economic point of view, external financing is viewed as asource of firm financing and financial institutions, especially banks has been proved toplay a virtual role in financing investment Schumpter (1912) believed that anentrepreneur must borrow to afford his new business if his purchasing power is notstrong enough In this case, “he cannot become an entrepreneur by previouslybecoming a debtor” In addition, he considered the banker as the key agent in thisprocess Later, it is widely accepted by many authors when investigating finance-growth relationship that investment financing comes from the financial system(King&Levine, 1993; Levine & Zervos, 1998; Levine et al., 2000; Warchtel, 2003etc.) In econometric model of finance-growth, they use the share of broad moneysupply and private credits to gross domestic products (GDP) as the proxies of financialdevelopment The findings state that there is a significantly positive relationshipbetween financial development and economic growth Therefore, they jump to a

conclusion that a developed financial system could promote long-term economicgrowth by the channel that financial intermediaries provide credits to finance firms’development plans.

However, there are some economists arguing these results Arestis (2004; 2005)and Guirat and Pastoret (2009) raised a serious question that what type of loans thatfinancial intermediaries offer firms, long-term or short-term It is clear that the role ofshort-term loans is to facilitate firms’ cash-flow rather than to finance investment.Hence, when choosing proxies for financial development, authors need to take intoconsideration the type of credit or loans Nevertheless, this seems to be missing in thetwo proxies, the share of broad money supply and private credits to GDP, in theprevious studies Another argument is coming from commercial banks’ difficulty inmandating long-term credits Apparently, most recent commercial banks’ sources comefrom short-term deposits

2.1.2 Internal financing

Table 2.1 Share of financing sources in two industries in the United Kingdom (UK)

There is a vast literature discussing the importance of internal finance sources.According to Sweezy (1968), the firm’s profits growth mainly contributes to the capitalaccumulation gain for the purpose of production expansion Moreover, Mayer (1990)used firms’ data in the United States (US) from 1970-1985 to provide an empiricalevidence that in this case, firms mostly rely upon internal sources The result is similar

to the study of British firms of chemical and electrical engineering industries from1949-1984 that bank loans and bond securities do not play a significant role to fixedinvestment financing (see Table 1) Instead, bank loans which are considered as themain source of external finance mostly contribute to firms’ working capital

Using the same data of two industries in UK but with the different approach,Mayer (1990) examined the variation of capital structure according to the firm size.Table 2 points out that large firms seem to require more on internal financing incomparison with external financing

Table 2.2 The proportion of financing sources according to different firm sizes

Source: Business Monitors (M3) in Mayer (1990)

Corbett and Jenkinson (1997) provided another evidence to support the Mayer’sfindings by analyzing the source of finance of three developed countries includingGermany, the US and the UK from 1970-1994 The study shows that in the period1970-1994, the firms in these countries mostly rely on the internal source of financewith the average proportion of the internal finance and total fixed investment of 78.9

percent in Germany, 93.3 percent in the UK and 96.1 percent in the US These arestories in developed countries.

What happens in the countries with the state control over financial systems? Inthese countries, state development banks and other state owned banks follow lowinterest rate policies to stimulate investment (Singh, 1998), especially in the early stage

of development The poof indicates that the most fundamental source of finance forfixed investment comes from internal finance; however the share of long-term bankloans are still high compare to the rate in developed countries mentioned above.Choosing Japan before becoming a developed country in 1990’s as the case of statecontrol over financial system, Tsuru (1993)figured out that the share of internalfinancing over gross manufacturing investment accounted for 60 percent in the late1950s, increasing to 75 percent in the late 1970’s and around 100 percent in the late1980’s The long-term bank loans covered the rest of investment financing, and hadcontinued declining from 1950’s China is a typical example for the country with thefinancial system regulated by the government Specially, the Chinese government plays

a central role in the financial system It is demonstrated that the proportion of the totalassets of state-owned commercial banks and development banks to total banks’ assetaccounts for roughly 60 percent Mlachila and Takebe (2011) argued that due to thegovernment regulation, these banks can lend lower interest rate to firms, especiallywith state-own enterprises investing in construction, logistic and heavy manufacturing

2.2 Financing constraints on firms

2.2.1 Definition

According to Modigliani-Miller (1958), firm’s capital structure is irrelevant toits value in perfect market In other words, external finance perfectly replaces internalfinance However, in financial friction or imperfect financial market, this is not true.External finance is more expensive than internal finance due to asymmetric

information, transaction costs and agency problems (Fazzari et al.,1988, Mayer, 1990,Schiantarelli, 1995, Blundell et al., 1996) Kaplan and Zingales (1997) suggested that afirm faces financing constraints when its costs or accessibility to external sourcesbecome barriers preventing the firm from profitable investment projects Laeven(2003) summarized a large body literature to introduce a definition of financingconstraints on firms that a firm’s investment is considered as financially constrained if

an unexpected increase in internal funds leads to a higher level of investment spending

2.2.2 Measurement of financing constraints on firm

In order to test the existence of financing constraints on firms, Euler investment

equation approach and q investment model are commonly applied in most empirical

studies Both two approaches contain their own advantages and disadvantages

The q model originates from Tobin (1969) and is developed to q-investment

model by Hayashi (1982) Fazzari et al (1998) first introduced the procedure to test the

presence of financing constraints on firm using q-investment model If all markets are

perfectly competitive, a firm’s production and installation functions exhibit constantreturns to scale, and stock markets are strongly efficient, then a firm’s investment only

depends on marginal q which is equal to average q However, when these strict assumptions are not satisfied, the q model is not a good proxy for firm investment (Hayashi, 1982).In the improvement process of q investment model with a novel

approach, Abel and Blanchard (1986) tried to predict the expected present value of thecurrent and future profits produced by an additional unit of fixed capital Theadvantage of this approach is that it does not use stock price data However, thedisadvantage of this approach is that a certain stochastic process on the variables needs

to be assumed

Another approach is Euler investment equation, which is a necessary conditionfor an optimal policy employed by a firm which seeks to maximize its value The first-

order conditions for investment remove the shadow cost of capital from the Eulerequation The Euler investment equation is first introduced by Albel (1980), andapplied by many authors with different adjustments (White, 1992; Hubbard, Krashyap

& White, 1995; Bond & Meghir, 1994; Harris, Schiantarelli & Siregar, 1994, Gilchrist

& Himmelberg, 1999; Love, 2003; Laeven, 2003; Forbes, 2007, Chan et al., 2012,etc.) The Euler investment equation does not use stock price data and not have toassume linear homogeneity of the net revenue function Moreover, it does not require

to assume a certain stochastic process on the variables as in the q investment model as

in the q investment model and the Abel and Blanchard (1986) model Nevertheless, theEuler equation needs to assume and adjustment cost function and a firm’s smoothinginvestment expenditures over time

2.3 Financial development and financing constraints on firms

Basically, financial sector development occurs if financial instruments, marketsand intermediaries are less prone to market asymmetry information, limitedenforcement and transaction costs However, Čihák et al (2013) argued that if onlycapture the ease of market imperfection, the financial development measurementshould cover the actual function of financial development to the whole economy.Therefore, many authors (Merton, 1992; Demirguc-Kunt & Levine, 1996;Demirguc-Kunt&Maksimovic, 1998; Rajan & Zingales, 1998; Levine, 2005; Čihák et al., 2013etc.) adopted a broader definition of financial development as improvements in five keyfinancial functions, including: “(1) producing and processing information aboutpossible investments and allocating capital based on these assessments; (2) monitoringindividuals and exerting corporate governance after allocating capital; (3) facilitatingthe trading, diversification, and risk management; (4) mobilizing and pooling savings;and (5) easing the exchange of goods, services, and financial instruments.” As theresult, barriers to access to banking sector have been reduced and capital market hasbeen triggered The gap between external and internal source of finance has gradually

lowered, and the firm’s entry to banking sector has been stimulated In other words, it

is commonly thought that under the existence of financial development, costs of capitalfor firms are decreased, and barriers for firms to external sources are graduallyremoved After all, from theoretical point of view, with the presence of financialdevelopment firms’ financing constraints are ameliorated and their investmentincreases

The empirical examination of the relationship between financial developmentand firms’ investment is of contemporary concerns The current empirical findings offinancial development and financing constraints nexus in developing countries are stillambiguous Demirguc-Kunt and Maksimovic (1998, DM) used cross country firm-level data to investigate whether financial development affects the degree to whichfirms are constrained from investing in profitable growth opportunities Using the sameindicators to measure financing needs among firms in different countries, DM casteddoubt on Rajan and Zingales (1998) who used industry level data to investigate the tiebetween financial development and economic growth DM argued that it would beprecise if those differences are addressed at firm level data Firstly, DM computed therate for individual firm at which firm can grow in two cases (1) it is only funded byinternal sources and (2) both internal funds and short term loans are employed Then,the proportion of firms whose growth rates are in excess of the maximum rate that firmuse only internal fund is calculated It is denoted as Pr_faster Based on this, theproportion of firms in each country relying on external fund would be estimated Inorder to examine the impact of financial development on firm’s growth, DM run thecross country regression:

Pr_fasterit = β1FDi,t+ β2CVi,t + εi,t (1)

in which FD is financial development, CV a set of control variables and ε thedisturbance term To measure financial development variables, DM use a set offinancial indicators including the ratio of market capitalization to GDP, the total value

of shares traded divided by market capitalization and the rate of bank assets to GDP.For control variables, DM employed the different mix of macroeconomic variable such

as economic growth, inflation, the average market to book value of firms in the wholeeconomy, total amount of subsidies from government to firms, the average ratio of netfixed assets over total assets, the real GDP per head, the law and order tradition of theeconomy Finally, they found that the proportion of firms whose growth rates are inexcess of the maximum rate that firm use only internal fund is positively related tofinancial development and legal system variables Applying the same approach, butextending the sample data of the largest listed manufacturing firms in 26 countries,Beck et al (2001) confirm the findings

However, Love (2003) stated that the firms’ investment opportunities indifferent time and countries are not mentioned in DM’s approach She asserted that totest whether financial development enhances the allocation of capital within a country,growing firms need to be identified, given their investment opportunities She usedpooled cross country firm-level data in an investment model based on the Eulerequation approach It embodies financing constraints by parameterizing the shadowcost of external funds as a function of the firm’s cash-flow which is built on Gilchristand Himmelberg (1999) This approach is widely applied by Chan et al (2012), Forbes(2007), Harrison, Love and McMillan (2004) when they analyze the effect of financereform, financial liberalization or banking reform on firms’ investment decision Forfinancial development proxies, five standardized indices, including GDP, total valuetraded over GDP, total value traded over market capitalization, M3/GDP, financialdepth (credit to private sector to GDP) from Dermirguc-Kunt and Levine (1996) arecombined into a single measure This index is standardized to have the mean of zeroand the standard deviation of one Love used a panel data of 5,000 large, publiclytraded firms in 36 countries Her evidence indicates that financial development makes

investment of firms, especially small ones, less sensitive to internal finance This is inline with the theory and many previous studies.

Leaven (2003) conducted the study of firm-level data from 13 developingcountries to examine the difference of investment sensitivity to cash-flow The levels

of liberalization scaled from 1 to 6 are different across countries He found thatalthough investment is largely constrained by cash flow with both firms, there aredifferent effects of financial liberalization on small and large firms Small firms arefound to be more financially constrained before liberalization and less financiallyconstrained after liberalization in comparison with large firms Large firms also arefound to face more financially constrained after liberalization It might come from thefact that large firms often receive incentives from the state such as directed creditprogram of state development banks or other preferential loans which must be scaleddown after liberalization

A different approach is represented by O’Toole and Newman (2012) Despite ofthe same purpose of study the influence of financial development on firms’ investmentfinancing via financing constraints, they differed from previous studies in some points

Firstly, O’Toole and Newman (2012) proposed using fundamental q model based on

Gilchrist and Himmelberg (1995) and applied empirically by Love and Zicchino (2006)with following procedure:

In which xit includes a proxy for the firms' marginal value product of capital In thisresearch, a sale to capital ratio has been used as a proxy for the marginal product ofcapital This is a valid proxy under constant return to scale of Cobb- Douglasproduction structure as detailed in Galindo, Schiantarelli, & Weiss (2007) andrepresent time and firm specific effects, respectively The vector autoregression is used

to estimate the coefficient matrix A which is then included in equation (2) Themarginal value product of capital ratio is identified by vector c, represents thecombined discount and depreciation rate O’Toole and Newman (2012) argue that the

q model might be flexible because it relates investment to the firm’s individual

investment opportunities Secondly, firm’s financing composition mix is the first timedefined by using the differential cost of capital relating to the financing optionsavailable to firms Thirdly, financial development indicators are calculated atprovincial level based on Vietnam Enterprise Survey data before these added in thefirm’s investment model Provincial financial development indicators here areclassified into three families, including financial depth, state intervention and thedegree of market financing in the economy The idea is taken from Guariglia andPoncet (2008) in China due to the similarity between Chinese and Vietnameseeconomies To investigate the impact of financial development on firm’s investment,financial development indicators are interacted with firm’s financing composition mix.O’Toole and Newman’s results support the view that financial development reducesfirms’ financing constraints though credit expansion or through more efficient creditallocation However, the impact differs across firms Financial development hassignificant influences on domestic private firms, but negligible effects on foreigninvested firms

In consistent with the findings in previous studies, Harris, Schiantarelli &Siregar (1994) for the case of Indonesia, Gallego and Loayza (2001) for the case ofChile, Gelos and Werner (2002) for Mexico, they both jump to the conclusion thatfinancing constraints are relaxed for small firms but not for large ones after financialdevelopment However, the empirical findings of the relationship between financialdevelopment and firms financing constraints have not been conclusive Using the samemethod in Love (2003) but only examining the financial development in singlecountry, China and Chile, respectively, Chan et al (2012) and Forbes (2007) found

interesting findings Financial reforms in China do not benefit SMEs and SOEs interms of alleviating the financing constraints This also occurs similar to the case ofChile in the capital control period from 1991 to 1998 Chile small firms sufferedsignificant financing constraints which did not appear before and after the capitalcontrol (Forbes, 2007) Jaramillo et al (1996) found an insignificant relationshipbetween financial development and firm’s investment when examining the case ofEcuador.

2.4 Conclusion

To conclude, in order to investigate the effect of financial development onfirms’ investment, the recent approach widely used is via analyzing firms’ financingconstraints The problem is that how to formulate the firms’ investment and throughwhich the tie between financial development and firms’ financing constraints is

clarified Q model and adjusted Euler specification model are two main approaches of

modeling firms’ investment Formulating financial development indicators is also ofconcern Many different indicators are employed to represent financial development,and they have varied in different studies Most of them are financial depth, the ratio ofmarket capitalization to GDP, the total value of shares traded divided by marketcapitalization and the rate of bank assets to GDP, M3/GDP The impact of financialdevelopment on financing constraints on firms is still ambiguous Most empiricalstudies show that financial development has positive impact on firms’ investment,relaxing firms’ financing constraints Nonetheless, several studies in some developingcountries point out a contradictory result that firms seem to not benefit from thefinancial development process

In this paper, Euler investment equation approach is employed to detect thepresence of financing constraints on firms due to its outstanding advantages inaccordance with Vietnamese firm data and economy context The model specificationand hypotheses development will be presented in the next chapter

Chapter 3: Model specification

This chapter contains three main parts The first part displays the theoreticalframework to model investment As mentioned in the previous part, Euler investmentequation approach as well as how to use it to test for the presence of financingconstraints would be presented The second part discusses the measurement offinancial development indicators The final part specifies the model which is expected

to test the impact of financial development on firm investment

3.1 Investment modeling

3.1.1 Euler investment equation approach

Since the first introduction in Abel (1980), the investment Euler equationapproach has been widely used by many authors (White, 1992; Hubbard, Krashyap&White, 1995; Bond & Meghir, 1994; Harris, Schiantarelli & Siregar, 1994, Gilchrist &Himmelberg, 1999; Love, 2003; Laeven, 2003; Forbes, 2007, Chan et al., 2012) Theequation is acquired after reorganizing first order conditions when solving themaximizing firm’s value problem

In this paper, the investment model based on Euler equation approach closelyfollows models in previous studies, especially Laeven (2003) and Bond & Meghir(1994) The firm is assumed to maximize its present value, which is the sum of theexpected discounted value of dividends subject to the capital accumulation and externalfinancing constraints

Let be the firm value at time t, the dividend paid to shareholder at time t,and + = ∏ =1 (1 + + −1 )−1 the j–period discount factor for j≥ 1 where is the risk-free expected rate of return and = 1 Then, the function of firm

value at time t is ∑∞ =0 + + , the expected value of present values of future dividend payments conditional on time t information Let П = ∏

( , , ) denote the profit function

15

where is the firm’s capital stock at time t, the vector of variable input, and the firm investment at time

t The profit function is assumed to be concave and bounded according to Gilchrist and Himmelberg (1999) With the depreciation rate and the investment expenditure at time t, , the capital stock

fictions are taken into account via the assumption that the marginal source of external finance is debt that risk-neutral debt holders require an external premium, = ( ) This premium is expected to be increasing in the amount of debt due to agency costs Following Gilchrist (1999) and Laeven (2003), the gross required rate of return on debt is assumed to be equal (1 + )(1 + ( )) .

To ensure that the only debt is the firm’s marginal source of finance, a negativity constraint on dividends is introduced It also means that shareholders favordividends paid out instead of retention For all of the above, the manager’s problem ofmaximizing firm’s value is as follows:

non-∞

{ + , + , , + } =0 ∞

16

Substituting in (2) into (1) and using (3) to eliminate in the profit function, theLagrangian function is constructed as:

Secondly, the first order condition for debt Bt is:

= {(1 + ) − [(1 + )(1 + )(1 + ( )) ]} = 0

+1 +1

According to the study of Bond & Meghir (1994) on the hierarchy of finance approach, when firm pays positive dividend, > 0 and not issue new equity, firms generate sufficient profit to finance its dividend payments and investment from retained earnings It is the case that +1 = = 0, the shadow value of internal fund is zero, or the firm is not financially constrained Then, the Euler equation (8) transforms to the standard investment model without financial friction:

(1− )

[П +1 ] =

П + П

Where ( , ) is a constant return to scale production function; ( , ) is convex adjustment

cost function of installing , which is assumed to be homogeneous in and ; p t is the output price;

is the vector of input prices; is the price of investment

18

Let Y = ( , ) − ( , ) be net output which is assumed to be homogeneous in both K t and L t To allow for imperfect completion, is assumed to depend on Y t with

a constant price elasticity of demand | |>1.

Adjustment costs function:

The estimation requires an adjustment cost function A common specification widely employed is:

( , ) = 1 ( − ) 2 ,

which is strictly convex in I t and homogeneous of degree one in I t and K t

19

From this adjustment cost function, differentiating with respect to :

Regarding to the above assumption of production function ( , ) that it is

homogeneous with , , ( , ) could be expressed as:

2

=− the cash-flow to capital ratio

3.1.2 Detecting the presence of firm’s financing constraints using Euler equation

In order to test the presence of the firm’s financing constraints on investment, the empirical specification is constructed from equation (20) with the exclusion of the cost of capital J due to its measurement difficulty.

And the subscript i refers to the firm, f i and d t denote firm- and time-specific effects

Under the null of no financial friction, Bond and Meghir (1994) demonstrated that 1 > 1; 2 < −1; 3 > 0 and 4 <0 If any of these restrictions is not satisfied, investment diverges from its optimal path Especially, if 4 >0 or investment responds positively to an increase in cash-flow, the firm faces financing constraints The intuition behind is that an expected increase in cash-flow lead to a higher level of investment which is funded before internal finance is up Moreover, a firm could be considered more financially constrained if the coefficient

4 is more positive.

23

3.2 Financial development measurement

Based on the definition quoted in 2.3, Čihák et al (2013) constructed measures

of four main characteristics consisting depth, access, efficiency and stability offinancial institutions and financial markets A 4x2 framework for benchmarkingfinancial systems (see below) in their study has been used by most mainstreameconomists and adopted into policy guidelines of World Bank, IMF and UN

Table 3: A 4x2 framework for benchmarking financial systems

domestic private debt securities to GDP

 % of firms with line of credit (all firms)

years)

Ratio of domestic to total debt securities

 Ratio of new corporate bond issues to GDP

market

Price synchronicity (co-movement)

 Profitability (return on assets, return on

Boone indicator (or Herfindahl or

exchange

 Settlement efficiency

 Z-score (or distance to default)  Volatility (standard deviation / average) of stock

price index, sovereign bond index

Skewness of the index (stock price, sovereign

Source: Čihák et al (2013)

In empirical studies, authors used wide range of proxies of financialdevelopment Among them, Dermirguc-Kunt and Levine (1996) employed fivestandardized indices, including GDP, total value traded over GDP, total value tradedover market capitalization, M3/GDP, financial depth (credit to private sector to GDP).Love (2003) inherited these proxies for his research Demirguc-Kunt and Maksimovic(1998) used a set of financial indicators including the ratio of market capitalization toGDP, the total value of shares traded divided by market capitalization and the rate ofbank assets to GDP Notably, they are country level proxies of financial development.Another approach of financial development proxies could be found in Guariglia andPoncet (2008) when studying the case of China In this study, financial development isdecomposed into three features, including financial depth and intermediarydevelopment; state involvement and the degree of market financing All the featuresare calculated for provincial level These proxies seem to be more appropriate for thecontext of transition economies from the central planned to the market-oriented.O’Toole and Newman (2012) employed these proxies for their study of Vietnam theclose similarity in the economy context between Vietnam and China However,problems might come from calculating process in using VES data The VES sampleselection mainly focuses on provinces with high number of enterprises Using VESdata for computing provincial level data might cause bias results

Due to no existence of a single aggregate index in the literature and theavailability of data, in this paper I employ three principal indicators that are widelyused as proxies for financial development Those are liquid liabilities ratio, privatecredit to GDP ratio, and stock market total value traded to GDP (Dermirguc-Kunt &Levine, 1996; Levine et al., 2000; Love, 2003; Huang, 2011b; Čihák et al.,2013).

Liquid liabilities ratio (LLG)

Liquid liabilities ratio is one of the primary indicator to measure the size offinancial intermediaries compare to the whole economy (Huang, 2011b; Love, 2003;Beck et al., 2001; Levine et al., 2000); Demirguc-Kunt& Levine, 1996; and King &Levine, 1993) It is measured by the ratio of the liquid liabilities of banks and non-bankfinancial intermediaries (i.e M3, the currency plus demand and interest-bearingliabilities) to GDP According to Levine et al (2000), LLG might not an efficientindicator to measure the effectiveness of financial system in relaxing asymmetryinformation problem as well as the transaction costs However, due to the positiverelationship between the size of financial intermediaries and the feature and quality offinancial services, LLG is accepted by authors as a measure of financial depth

Private credit to GDP ratio (PCG)

Private credit to GDP ratio is calculated by the credit from banks and otherfinancial intermediaries to private sector over GDP This not only excludes the credit togovernment, government agencies and state-owned companies but also the credit frompolicy and development banks De Gregorio and Guidotti (1995) argued that thisindicator measures the allocation of fund to private sector, which triggers investmentand investment efficiency Levine et al (2000) argue that although PCG indicatorpoorly explains the improvement of financial development on dealing with informationasymmetry and transaction costs problem, higher level of PCG indicates betterfinancial services and thus the greater financial development

Stock market capitalization to GDP (SCG)

Together with financial institutions, stock market also plays a crucial role in thedevelopment of financial sector as well as economic growth (Levine & Zervos, 1998)

By taking part in the stock market, investors can diversify their investment portfolioand reduce transaction costs and their risk Čihák et al (2013) argued that stock marketcapitalization to GDP, which is measured by the ratio of total value of all listed in thestock market to GDP, is a common choice for financial market depth due to the itsapproximating of the stock market

3.3 Empirical model to evaluate the impact of financial development on firm investment

This thesis aims to examine whether financial development relaxes firm’sfinancing constraints or stimulates the firm’s investment For this purpose, Love(2003), Laeven (2003), Forbes (2007), Chan et al (2012) proposed that the interactionterm of the cash stock variable and commonly used financial development indicatorsshould be added in the model (21) above Moreover, to investigate whether the impact

firm’s size (SIZE =Size of the firm measured by Log of total assets_lnTA), and financial development indicators (FD) is included in the model (20) The final empirical function is as follows: