MINISTRY OF EDUCATION AND TRAINING
FOREIGN TRADE UNIVERISTY

PHD THESIS SUMMARY
ANALYSIS OF THE DETERMINANTS OF
CAPITAL STRUCTURE OF VIETNAMESE INDUSTRIAL FIRMS

Major
Specialized
Code number

:
:
:

Business
Business administration
62.34.01.02

PHAN THANH HIEP

HANOI, 2017


THIS THESIS IS COMPLETED IN
FOREIGN TRADE UNIVERSITY

SUPPERVISOR: ASSOC. PROF, DR NGUYEN DINH THO

Reviewer 1: ASSOC. PROF, DR DINH TRONG THINH


the managers. It would seem that there are some factors impact on capital structure
of firm. So, analyzing how these factors impact on the change of capital structure is
a very significant research and it would help the directors to manage the capital
structure.
It also would seem that capital structure of firm varies from different
industries. For example, capital structure of commerce firms is different from it of
heavy industrial firms. So, it is necessary to focus on a specific industry to analyses.
Base on this view, this thesis has chosen industrial firms which plays important role
in the modernization and industrialization process of Vietnam to research.
Finally, the concept of capital structure and the impact of some determinants
to the capital structure of Vietnamese firms have been studied for years. In theorem
research, these papers refer to almost of capital structure theories such as: trade off
theory, perking order theory, agency-cost theory, signaling model theory, market
timing theory… In empirical research, there have been multiple samples to be
collected and analyzed, but the results seem to be different from each paper.
Especially, most of capital structure papers for Vietnamese firms focus mainly on
the impact of determinants to the capital structure and dropt the impact of the
capital structure to the value of firm. This is a very important issue because it may
lead to the endogenous problem among variables in the model. Most of the papers


2
in this field in Vietnam use Fixed effect model (FEM) and Random effect model
(REM) to run the regression. In the theory, these two models are unable to
overcome the endogenous problem. So, using the model of Generalized Method of
Moments (GMM) which can solve the endogenous issue could be viewed as a new
solution of this thesis in the way to approach the research of determinants of capital
structure.
2. Literature review
This content will be show in the chapter 1 of the thesis.


-

In the content of impacts of some determinants on capital structure, this
thesis approaches the problem by clarifying the determinants into two
categories: (1) business factors and (2) corporate governance factors.

-

Firms in research sample: audited financial report of 95 Vietnamese listed
industrial firms in period of 2006 – 2015.

5. Methodology


3
Following most of economic researches, this thesis uses simultaneously two
study methods: qualitative method and quantitative method to complete the
objectives of the thesis.
6. New contribution of the thesis
In the perspective of science, this thesis contributes some points as: (1) this
thesis runs empirical studies for impacts of determinants on capital structure (2) this
thesis uses two models for studying the impacts of some determinants on capital
structure, the first one uses business factor variables and the other uses corporate
governance factor variables, (3) This thesis use GMM model which could be
viewed as a new economic methodology in Vietnam, (4) Mentioning the optimum
capital structure for Vietnamese industrial companies by running the threshold
regression model, (5) This thesis focuses on Vietnamese industrial companies,
which could be the leading team in the economic growth of Vietnam.
In the perspective of practical application, this thesis makes some new

Bradley

Determinants

et al.

Volatility

Long
and

Titman
Kester

Marsh

Malitz

Wessels
-*

-

and

Bankruptcy threat

-

-*


+

Advertising expenditure

-

-

Research and development fee

-

-

Profitability

+*

The uniqueness of firm
Source: Survey of Harris and Raviv (1991).

+

-*
-


5
In the above table, the plus mark represents the positive relationship, minus

Positive

Deesomsak (2004), Huang and Song (2006), Pandey
(2002), Frank and Goyal (2007), Doan Ngoc Phi Anh
(2010), Okuda and Lai Thi Phuong Nhung (2010), Dang
Thi Quynh Anh and Quach Thi Hai Yen (2014)

Profitability

Positive

Erol (2004), Pandey (2002), Dang Thi Quynh Anh and
Quach Thi Hai Yen (2014).


6
Meyrs and Majluf (1984), Kester (1986), Thies and Klock
(1992), Titman and Wessels (1988), Chen (2003), Hall,
Negative

Hutchison, Michaels (2004), Deesomsak (2004), Tran
Dinh Khoi Nguyen (2006), Pandey (2002), Doan Ngoc
Phi Anh (2010), Okuda and Lai Thi Phuong Nhung
(2010).

Growth

Positive /

Myers (1977), Chen and Zhao (2006), Pandey (2002),


Volatility

Liquidity
Relationship
with bank

Corporate
income tax

(1984), Titman and Witsell (1988), Huang and Song

Positive

Tran Dinh Khoi Nguyen and Ramachandran (2006)

Negative

Deesomsak (2004)
Petersen and Rajan (1994), Donnelly, Berry, and

Positive

Thompson (1985), Tran Dinh Khoi Nguyen and
Ramachandran (2006)

Positive
Negative

M & M (1963)


Positive

Ranti (2013)

Negative

Agyei and et al. (2014)

Without statistic
meaning
Board structure

Some important studies

Positive

Jaradat (2015)
Agyei and et al.(2014)
Agyei and et al.(2014), Tran Dinh Khoi

Organizational ownership
structure

Positive

Nguyen and Ramachandran (2006), Le Thi
Phuong Vy and Phan Bich Nguyet (2015),
Doan Ngoc Phi Anh (2010)


2: THEORIES OF CAPITAL STRUCTURE,

IMPACT OF CAPITAL STRUCTURE ON FIRM VALUE AND
IMPACT OF DETERMINANTS ON CAPITAL STRUCTURE
2.1 The definition and measurements of Capital structure
In the study of Bui Van Van et al. (2015), the definition of capital structure is
defined: capital structure is the concept shows the correlation of parts in total capital
of firm. The total capital of firm is divided into two parts: debt and equity. The term
of debt are also divided into two parts basing on the borrowing time: short term debt
and long term debt. In this thesis, some terms of capital structure are referred as:
total debt ratio, long term debt ratio, short term debt ratio, long term capital ratio
and equity ratio.
The measurement of above terms is:
Total debt ratio = Total debt / total asset
Equity ratio = Total equity / Total asset
Short term debt ratio = Short term debt / Total asset
Long term debt ratio = Long term debt / Total asset
Long term capital ratio = (Equity + long term debt)/ Total asset


9
2.2 The fundamental theories of capital structure
2.2.1 Modigliani and Miller theory
2.2.1.1 Modigliani and Miller (1958)
2.2.1.2 Modigliani and Miller (1963)
2.2.2 The static trade of theory and optimum capital structure
2.2.3 Jensen and Meckling theory about agency cost
2.2.4 The perking order theory of Myers and Majluf
2.2.5 The market timing theory
2.2.6 The theories of signaling, risk avoided philosophy, and others

firms
2.4.2.1 The context of national economy and capital market
2.4.2.2 The context of industry sector
2.4.2.3 Characteristics of industrial firms

CHAPTER 3: THE ACTUAL FACTS OF CAPITAL STRUCTURE
OF VIETNAMESE INDUSTRIAL FIRMS
3.1 Review of Vietnamese industry sector
The growth speed of industry sector of Vietnam has been maintained around
7% per year since 2005. However, this sector has been accounted for around 38% of
GDP for years.
3.2 The actual facts of capital structure of Vietnamese industrial firms in
comparison with non financial listed firms in Vietnam
3.2.1 Most of Vietnamese industrial firms follow the rule of compatibleness


11
3.2.2 The total debt ratio of Vietnamese industrial firms is lower than the
ratio of non financial listed firms
3.2.3 The long term debt ratio of Vietnamese industrial firms is lower than
the ratio of non financial listed firms and getting lower
3.2.4 There are differences in the movement of short term debt ratio of
Vietnamese industrial firms and non financial listed firms in Vietnam
3.2.5 There are differences in the movement of short term debt ratio of and
long term debt ration of Vietnamese industrial firms
3.2.6 Vietnamese industrial firms are borrowing short term debt more than
long term debt
3.2.7 The relationship among capital structure and economy crisis and
interest rate
The economy crisis in the period 2008 – 2011 with two shock of interest rate

4.3.1 The model of business determinants of capital structure
4.3.1.1 Regression model, variables measurement and assumptions
The regression equation is:
CAP.STRUC i t = αi + β1 SIZE + β2 ROA + β3 TOBIN.Q + β4 GROWTH + β5
TANG + β6 VOLATILITY + β7 LIQUIDITY + β8 TAX + β9 INTEREST + β10
YEAR.DUM + εi t
4.3.1.2 Regression methods and tests
The regression method of this model is SGMM. There are three endogenous
variables: the lag of capital structure, profitability, TobinQ and volatility. The
remaining variables are instrument variables. For the tests, SGMM model uses two
basic tests: Arellano-Bond (AR2) and Hansen test.
4.3.2 The model of corporate governance determinants of capital structure
4.3.2.1 Regression model, variables measurement and assumptions
The regression equation is:


13
CAP.STRUC i t = αi + β1 B.SIZE + β2 CEO.DUAL + β3 CEO.GEN + β4
CEO.AGE + β5 SIZE + β6 STATE + β7 GROW + εi t
4.3.2.2 Regression methods and tests
The regression methods of this model are Fixed Effect Model and Random
Effect Model.
For the tests, there are three tests will be checked: autocorrelation test,
heteroskedasticity test, and Hausman test (if necessary).
4.4 The model for testing the impacts of capital structure on the business
effectiveness and firm value.

CHAPTER 5: ANALYZING THE REGRESSION RESULT OF
DETERMINANTS OF CAPTITAL STRUCTURE MODEL
5.1 The model of business determinants of capital structure

855

0,08983

0,139485

0

0,719237

BSDTA

855

0,348872

0,178588

0,006971

0,80502

MTDTA

855

0,445037

0,248727



14
Table 5.2: Descriptive statistics of independents variables

Variables

Number of
observation

Mean

Standard
deviation

Min

Max

tobinq

855

1,184366

0,886036

0,298066

14,92568


0,319884

-0,68624

3,311883

TANG

855

0,250017

0,193343

0,12134

0,97642

volatility

855

2,734085

3,270647

0,12453

44,57315


0,036473

0

0,370879

dumyear

855

0,222222

0,415983

0

1

Source: Calculated by Stata program


15
Table 5.3: Correlation matrix of variables
size

grow

tang

volatility


0,0942

-0,0934

1

tax

0,043

-0,0331

-0,1412

-0,0108

1

liquidity

-0,1664

-0,0556

-0,1237

0,0247

0,0024


0,0765

-0,1161

1

roa

0,0753

0,1613

-0,1426

0,2131

0,049

0,2499

-0,2638

0,3001

Source: Calculated by Stata program
5.1.2 Regression results and tests
The regression results are collected and showed in the table below. The result
of AR2 and Hansen test show that the regression results are confident because there
is not any auto correlation in second order differences and the instrument variables

debt ratio

Market value of
the long term
debt ratio

roa

-0,3999***

-0,2740***

-0,0825***

-0,5389***

-0,4133***

-0,1534***

tobinq

0,0021825

-0,0010

0,0079

-1,3227***


-

tang

-0,015

-0,0609**

-0,028

0,0311

-0,127***

0,031818

volatility

0,0001147

-0,0014

0,0009357

-0,01

-0,1363*

0,0037


-0,1633***

-0,316*

-

- 0,266***
0,0435***

interest

-0,6849

***

YEAR.DUM

0,0093

-0,0037

- 0,015***

0,102***

0,060***

AR2 Test

z = 1,13

66,05
Prob > chi2 =
0,146

61,53
Prob > chi2 =
0,254

80,91
Prob > chi2 =
0,155

82,53
Prob > chi2 =
0,127

71,12
Prob > chi2 =
0,071

Hansen Test

z = 0,81
Pr > z = 0,418


17

5.1.3 Regression results analysis
5.2 The model of corporate governance determinants of capital structure

828

0,1589255

0,320169

-0,68624

3,311883

state

828

0,2608696

0,439374

0

1

b_size

828

1,313281

0,432808


828

50,45169

6,983576

25

72

Source: Calculated by Stata program


18
Table 5.6: VIF Statistic
Variables

VIF

1/VIF

size

1,51

0,66026

b_size

1,44


0,965943

Mean VIF

1,21

.Source: Calculated by Stata program
5.2.2 Regression results and tests
Table 5.7: Woolridge Test result of autocorrelation on each models
Variables
BTDTA

Woolridge Test
F( 1, 91) =

Pr > F

Conclusion

60,290

0,000

autocorrelation

BSDTA

F( 1,


autocorrelation

MSDTA

F( 1,

91) =

58,188

0,000

autocorrelation

MLDTA

F( 1,

91) =

48,751

0,000

autocorrelation

Source: Calculated by Stata program


19

1733,99

0,000 Heterokedasticity

Conclusion

BREUSCH AND

BTDTA
REM

PAGAN
LAGRANGIAN

FEM

WALD
BREUSCH AND

BSDTA
REM

PAGAN
LAGRANGIAN

FEM

WALD

2.700.000 0,000 Heterokedasticity


WALD
BREUSCH AND

MTDTA
REM

PAGAN
LAGRANGIAN

FEM

WALD
BREUSCH AND

MSDTA
REM

PAGAN
LAGRANGIAN

FEM

WALD

3.000.000 0,000 Heterokedasticity

BREUSCH AND

MLDTA

0,0022

grow

0,0144**

0,0003

state

0,0373**

b_size

-0,0004

-0,01244

-0,0031

-0,0189

-0,07056***

-0,00241

ceo_dual

-0,0073



0,00387**
-0,00132

-0,09334*** 0,04366***

0,108425*** 0,091429***

0,00174

ceo_age

0,0001

0,000617

0,0000

0,000425

-0,00014

0,00003

cons

-0,3856

-0,72765


6.2.1 The solutions base on the empirical tests about the impact of
determinants on capital structure
6.2.1.1 Solution base on the impact of size factor on capital structure
6.2.1.2 Solution base on the impact of profitability factor on capital structure
6.2.1.3 Solution base on the impact of growth opportunity factor on capital
structure
6.2.1.4 Solution base on the impact of growth speed factor on capital structure
6.2.1.5 Solution base on the impact of liquidity factor on capital structure
6.2.1.6 Solution base on the impact of interest factor on capital structure
6.2.1.7 Solution base on the impact of tangibility factor on capital structure


22
6.2.1.8 Solution base on the impact of state ownership factor on capital structure
6.2.2 The solutions base on the applications of capital structure theories
6.2.2.1 The perking order theory
6.2.2.2 The trade off theory
6.2.2.3 The market timing theory
6.2.3 Some solutions for specific group of Vietnamese industrial firms
6.2.3.1 For the firms violated the rules of compatibleness in managing capital
structure
These firms need to find out some solutions to overcome the gap between the
value of the long term asset and the value of total long term capital (equity + long
term debt) by increasing the value of equity or long term debt.
6.2.3.2 Review the issue of reducing the long term debt ratio and the solution of
diversifying the capital suppliers
6.2.3.3 For the industrial firms with the total debt ratio exceed the rate of 65%
When the total debt ratio of firm exceeds the rate of 65%, the decision of
borrowing more is very likely lead to the reduction on ROA and ROE indicators. In
opposition, the decision of reducing the debt, could lead to the positive effects in the

6.3.3.1 Supporting the listed industrial firms to raise fund in the stock market and
bond market
6.3.3.2 Developing the bond market, especially the corporate bond

CONCLUSION
The findings from four models in this thesis show that both trade off theory
and perking order theory could be used to explain the movement of capital structure
of Vietnamese industrial firms. Specifically, the trend of how determinants impact
on the capital structure is summarized as: