11U ASSUMPTION UNIVERSITY LIBRAR'f

DETERMINANTS OF VIETNAM'S EXPORTS: AGRA VITY

MODEL APPROACH

By

NGUYEN HAI THO

An Independent Study
Submitted in partial fulfillment of the requirements
· for the Degree of

MASTER OF SCIENCE IN FINANCE AND ECONOMICS

MARTIN DE TOURS SCHOOL OF MANAGEMENT AND ECONOMICS

Assumption University
Bangkok. Thailand

November, 2013
 MARTIN DE TOURS SCHOOL OF MANAGEMENT AND ECONOMICS
MASTER OF SCIENCE IN FINANCE AND ECONOMICS
ASSUMPTION UNIVERSITY "'

This Study by: Ms. Hai Tho Nguyen

Entitled: "Detenninates of Vietnam's Exports: A Gravity Model Approach"

has been approved as meeting the independent study requirement for the:

DEGREE OF MASTER OF SCIENCE IN FINANCE AND ECONOMICS

Dr. Suppan Romprasert, Chairperson

Asst. Prof. Dr. Nopphon Tangjitprom, Member

Dr. Wanida Ngienthi, Advisor

Date of Defense November 18, 2013

 MARTIN DE TOURS SCHOOL OF MANAGEMENT AND ECONOMICS
MASTER OF SCIENCE IN FINANCE AND ECONOMICS
ASSUMPTION UNIVERSITY

Declaration of Authorship Form

I, NGUYEN HAI THO

declare that this independent study and the work presented in it are my own and has
been generated by me as the result of my own original research.

DETERMINANTS OF VIETNAM'S EXPORTS: A GRAVITY MODEL

APPROACH

l confirm that:

1. This work was done wholly or mainly while in candidature for the M.Sc.
degree at this University;

2. Where any part of this independent study has previously been submitted for a
degree or any other qualification at this University or any other institution, this
has been clearly stated;

3. Where l have consulted the published work of others, this is always clearly
attributed;

4. Where I have quoted from the work of others, the source is always given. With ·
the exception of such quotations, this independent study is entirely my own
work;

5. l have acknowledged all main sources of help;

6. Where the independent study is based on work done by myself jointly with
others, I have made clear exactly what was done by others and what l have
contributed myself;

7. None of this work has been published before submission

Signed:_~~
----__._---=-==---- Date: K~ [11(otJJ<l3

II
 MARTIN DE TOURS SCHOOL OF MANAGEMENT AND ECONOMICS
MASTER OF SCIENCE IN FINANCE AND ECONOMICS
ASSUMPTION UNIVERSITY

Student Name: NGUYEN HAI THO

ID: G5519594

ADVISOR'S STATEMENT

I confirm that this independent study has been carried out under my supervision and it
represents the original work of the candidate.

Signed :

Wanida Ngienthi, Advisor

Date:

Ill
 ACKNOWLEDGEMENTS

I would like to express my deep appreciation to Dr. Wanida Ngienthi, my research

advisor, for her guidance and valuable suggestions during the entire duration of my
research work. Her advice has been a great help in the completion of my paper.

I would also like to thank Assistant Professor Dr. Nopphon Tangjitprom for his
assistance and advice especia11y on my research methodology. His willingness to give
his time generously on answering my questions is highly appreciated.

I would also like to extend my special thanks to Dr. Suppanunta Romprasert as the
chairman of the committee for her useful and constructive recommendations, which
helped in improving the quality of my research paper.

Hai Tho, Nguyen

November, 2013

IV
 ABSTRACT

This research paper employed the gravity model to investigate and analyze the
determinants of Vietnam ' s exports to its forty major exporting markets over the
period of seventeen years, from 1995 to 2011. The Hausman test showed that the
fixed effects model was the most appropriate approach to estimate the gravity
regression. The results showed that Vietnam's exports patterns followed the basic
gravity model. In other words, Vietnam's exports increased as its GDP and importing
countries' GDP increased. In contrast, transportation costs, as proxied by geographic
distance, were found to have a negative impact on Vietnam's exports. Vietnam ' s FDI
was surprisingly found to have a significantly negative relationship with Vietnam ' s
exports. The research ' s results asserted the negative relationship between exports and
real bilateral exchange rate, indicating that the depreciation of Vietnam Dong against
the currencies of importing countries stimulated its exports. Importing countries ' GDP
per capita and the Free Trade Agreements dummy variable were found to have no
statistically significant influence on Vietnam ' s exports. The results of this paper can
be beneficial to the Vietnamese government and exporting companies in setting their
exports goals and policies.

v
 THE ASSUMPTION UNIVERSITY LIBRAR1

TABLE OF CONTENTS
Page
COMMITTEE APPROVAL FORM ................................................................. .
DECLARATION OF AUTHORSHIP FORM........... ....... ................................. n
ADVISOR'S STATEMENT...................... ........................................................ m
ACKNOWLEDGEMENTS................................................................................ 1v

ABSTRACT....................................................................................................... v
TABLE OF CONTENTS .. ................................................................................. VI

LIST OF TABLES.............................................................................................. Vlll

LIST OF FIGURES···························································································· Vlll

CHAPTER I: GENERALITIES OF THE STUDY

1.1 Background of the Study ............................................................................. .
1.2 Statement of the Problem.................. ......... ......... ...... .................................... 3
1.3 Research Objectives...................................................................................... 4
1.4 Research Questions.... ................................................................................... 4
l .5 Scope of the Research..... ....... ..... .... .............................................................. 4
1.6 Limitations of the Research. ....... .. .... ... .............. ..... ... .. ............. ..... ...... ......... 4
1.7 Significance of the Study... .. ......................................................................... 5
1.8 Definition of Terms...................................................................................... 5

CHAPTER II: REVIEW OF RELATED LITERATURE AND STUDIES

2.1 Theoretical Review. .... .. ... .......... ....... .............. ... ... ............ ....... .. ..... ............. . 7
2. I . I Comparative Advantage Theory................................................................ 7
2.2 Article Review.......... .............................................. ........ ....................... ....... 7
2.2. 1 Gravity Model... .... ... .... .. ............. ..... ......... ..... ... ..... ... .............. ............ .. ..... 7
2.2.2 Dependent Variables.............. ......................... ........................ ................... 9
2.2.3 Independent Variables............................................................................... 9
2.2.3.1 Exporting Country's Gross Domestic Product .............. ... ....... .. ...... ....... 10
2.2 .3.2 Exporting Country's Foreign Direct lnvestment .................................... 10
2.2.3 .3 Real Bilateral Exchange Rate................................................................. 1J

VI
2.2.3.4 Free Trade Agreements between Exporting Country and each Importing
Country ............................................................................................................ 12
2.2.3.5 Importing Country's Gross Domestic Product....................................... 13
2.2.3.6 Importing Country's GDP per Capita............ .......................... ............... 14
2.2.3.7 Distance. ................................................................................................. 14

CHAPTER Ill: RESEARCH METHODOLOGY

3. J Data ColJection .. ...... ... ... ... ............. ... ... ... ..... ...... .. ... .... .. .......... .. ... .... ..... ........ 18
3.2 Methodology ........................................................................................... ..... 19
3.2.1 SpecificationsoftheModel..................................................... .................. 19
3.2.2 Estimation Process.......................................................... ........................... 21
3.3 Research Hypotheses................................................................ .................... 22

CHAPTER IV: PRESENTATION AND CRITICAL DISCUSSION OF

RESULTS
4.1 Results of the Breusch-Pagan Lagrange Multiplier test, the Huasman Test, and
Checking of Heteroskedasticity .......................................................................... 24
5.2 Gravity Model Estimation Results....... ... .... ......... ...... ....... .. ... ..... ............ .. .... 25

CHAPTER VI: CONCLUSION, IMPLICATIONS AND FURTHER STUDY

BIBLIOGRAPHY........ .... ................ ....... ... ... ........... ................. .. .......... ...... ..... 32

APPENDIX........................................................................................................ 38
I. List ofVietnam·s forty major exporting markets......... ................... ...... .... .... . 38
2. Gravity regression with the fixed effects model........... ...... ......... .. ..... ..... ... .... 38
3. Gravity regression with the random effects model......... ...... ........... ............... 39

VII
 LIST OF TABLES

TABLES Page
Table 2.1: Summary of relationship between the dependent variable and independent
variables.............................................................................................................. 15
Table 2.2: Expectations of the relationships between the dependent variable and
independent variables ............. .. ... .. ......... ....... ....... ....... ... ... .................. ......... ...... 17
Table 3.1 : Measurements of variables.......... .................................... .................. 20
Table 4.1: Breusch-Pagan Lagrange multiplier test results .. .............................. 24
Table 4.2: Hausman test results.......................................................................... 25
Table 4.3: The result of heteroskedascity check...... .. .. ............................... ........ 25
Table 4.4: Gravity model estimation results....................................................... 26

LIST OF FIGURES

FIGURES Page
Vietnam's trade openness index ................................... .................... .
2 Vietnam 's exports percentage to GDP .............................................. . 2

VIII
 MARTIN DE TOURS SCHOOL OF MANAGEMENT AND ECONOMICS
MASTER OF SCIENCE IN FINANCE AND ECONOMICS
ASSUMPTION UNIVERSITY

Form signed by Proofreader of the independent study

I, Mr. Rom Kenneth B. Sales, have proofread this independent study entitled
"DETERMINANTS OF VlETNAM'S EXPORTS: A GRA VlTY MODEL
APPROACH" and hereby certify that the verbiage, spelling and format is
commensurate with the quality of internationally acceptable writing standards for a
master degree in financial economics.

Signed:

h B. Sales, Proofreader
Contact umber I Email Address: "[email protected]"

Date:

ix
 CHAPTER I

GENERALITIES OF THE STUDY

1.1 Background of the Study

In 1986, DoiMoi - an economic and political reform policy was launched in

Vietnam, which marked its transition from central planning and self-isolation to
market mechanism and integration into the global economy (Tuan, 2009). In the early
stages, major focuses of this policy were removing self-imposed barriers in order to
liberalize the domestic market, and attracted new foreign investments. As presented in
Figure I, Vietnam 's trade openness index was increasing over the period 1995- 2011,
which showed that Vietnam's economy was more integrated into the world economy
and the world trade over time.

Figure 1.1: Vietnam's trade openness index (Trade/GDP ratio)

r·----·-··---·-----·---------·-·-·--·- ·········-·-----·---·-----------·-----·-·---·-·--·--·--·-··-····--··-·----··--··--------------···---···--··

I 1800
% ----· Trade openess

140.0%

0 120.0% ; - - - - ·
:;:
...
c:
c.. -~
·-·-·--·----·---·-·----------------------··--
Q
l:)
a> 80.0% ----------··-·--··----·--··
-0
c:
....
E- 60.0% - - - - - -- · - - - - ----------------

40.0% -·---------------------------··-·--------------

20.0% - - - - --·------------------- ---·-·-·-··-------·-----···--··-----·----

O.O'Yo - - - - ----··-··---·--·-·-----·-------- - --------------·-------·-·-···--------··------·----------·

1995 1996 1997 1998 1999 2000 200 1 2002 2003 2004 2005 2006 2007 2008 2009 20 10 20 11

Year
Il...•·-·····--·······---·-- ·····-·----·---··-···------··--··-······ ·········-····-····---··-·--·-··-·-···------·-·--····-····-·-----····----····--····---···-·--·--· ········-·······--····-·········-·····-····--·-·----·---·-···--·····-··-··-········
Source: Vietnam 's General Statistics Office (20 I Ja)
 Since Vietnam opened its economy to the world, exporting activities have
contributed a significant pie to its economic growth (Dang, 2010). In the hard time of
economy, exports have helped to lift up the Gross Domestic Product (GDP) (Anh,
2012). As presented in Figure 2, exports have constantly increased its contribution to
Vietnam's GDP, especially in 2011 with about 78%.

Figure l.2: Percentage of Vietnam's Exports to GDP

Percentage of Exports to GDP

80.00%

70.00%
Q..
Q
\...:) 60.00%
.8
-"',_
0
c..
>(
50.00%

w 4000%
'-
0
<l>
CJ)
30.00%
..:::c:
<l>
u,_
20.00%
<l>
Q..

10.00%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Year

Source: World Bank (2013a)

Vietnam has comparative advantages in exporting crude oil, textiles and

garments, wooden products, agriculture products, leather shoes, fisheries, and
electronic products. However, those products are easily affected by a reduction in
price of competitors, notably China (Trung, 2013). Trung (2013) also pointed out that
Vietnam ' s exports would face many difficulties in the future because those products
are suffering from a demand slowdown in the world.
This leads to a need for conducting a comprehensive research paper on what
factors affect Vietnam's exports, so that the Vietnamese government may generate

2
more suitable policies or solutions to improve its current exports situation. There were
researchers identifying the determinants of Vietnam's exports in the past using
various research models. Most commonly identified factors are exchange rate,
economic growth rate, GDP of importing countries, inflation rate, and inward FDI
(Xuan& Xing, 2006, Hanh&Duc, 2009, Trang, Tam, & Nam, 2010).
The United States (U.S.), Europe, China, Japan, ASEAN, and some other
countries (as listed in the appendix) are major exporting markets of Vietnam, which
usually account for 80%-90% of Vietnam ' s total exports value each year (Vietnam' s
General Statistics Office, 2013b). This paper focused on finding out the determinants
of Vietnam's exports by using the Gravity model on data collected from 1995-2011
on these forty main exporting markets as listed in the appendix.

l.2 Statement of the Problem

Since export is considered as an engme of economic growth (Senhadji &

Montenegro, 1999), it is important to conduct a research on finding out factors that
might have an impact on exports of Vietnam. There has been empirical papers
previously conducted on this topic. However, the results varied. Differences in results
come from the fact that those researches were conducted over different periods of
time, used different sample sizes, or used different research models. This paper
conducted a comprehensive study on the possible determinants of Vietnam ' s exports
to its forty main exporting markets using the data from 1995 to 2011 in order to assert
the results by using Gravity model. The newest available data were used so as to
provide the most relevant and updated results.
Besides commonly considered variables of gravity model- GDP, GDP per
capita, Distance, and Free Trades Agreements, this paper expanded its study to other
two factors including Vietnam ' s FDI and real bilateral exchange rate. The gravity
model was chosen because it has been one of the most successful applications in the
empirical trade (Bae. 2010).

3
1.3 Research Objectives

The study was designed to investigate the factors that have an impact on
Vietnam's exports by applying the gravity model.

1.4 Research Questions

What are the determinants of Vietnam's exports by applying the gravity

model?

1.5 Scope of the Research

Seven independent variables were taken into consideration about their

potential effect on Vietnam's exports to forty major exporting markets. This included
Vietnam ' s GDP, Vietnam's FDI, distance from Vietnam to importing countries,
importing country ' s GDP, importing country's GDP per capita, real bilateral
exchange rate, and a dummy variable - FTAs between Vietnam and importers. The
forty countries were selected because of their significant contribution to the total
value of Vietnam ' s exports each year. The research was conducted based on the
newest available 17-year data (from 1995 to 2011) in order to provide the most
relevant and updated results.

1.6 Limitations of the Research

a) The study has not covered all possible determinants

There were more than seven factors that have been identified in this study.
These factors might have an impact on Vietnam ' s exports. However, this study has
not covered all factors. For example, competition from other countries ' exports of
similar products might be a significant factor that affects Vietnam ' s exports, such as
China' s and Thailand's. However, this factor was not taken into consideration in the

4
study. Under this research, impacts of those unmentioned factors were assumed to be
minor or insignificant. Actually, this is an inevitable issue for all researches.

b) The study has only covered Vietnam's forty major exporting markets

There are more than forty countries to which Vietnam exports. However, this
study has not covered all of them due to lack of data during the examined period. This
might not draw an exact picture about the determinants of Vietnam's exports, but it is
the closest image that can be done because the sample includes all top importers of
Vietnam's exported products.

1.7 Significance of the Study

The results of this study might be beneficial to the Vietnamese government in

identifying which factors have positive and negative impacts on Vietnam's exports.
Therefore, the government can take appropriate actions to achieve their desired
exporting goals through those factors. The study result may also be useful in the
Vietnamese government's plan of action for its future export activities. In other
words, as they see potential changes in those determinants of exports in the future,
they can better adjust their exporting strategies or plans consequently.
Exporting firms are the second group that can benefit from this paper.
Exporting firm can enhance their exporting plans by anticipating or following changes
in those determinants.

1.8 Definition of Terms

Exports: Exports represent the value of all goods and services

that are provided to the rest of the world (World Bank,
2013a).
Gross Domestic Product: It is defined as the market value of all final goods and
services produced within a country in a given period of

5
 time (Parkin, 20 I 0).
GDP per capita It is calculated by dividing the gross domestic product
by midyear population (World Bank, 2013b).
Free Trade Agreements: They are forms of trade pacts between countries, which
are set to eliminate tariffs, quotas and other trading
barriers between those countries (Kepaptsoglou,
Karlaftis, &Tsamboulas, 2010).
Real Bilateral Exchange Bilateral exchange rate is the price at which the money
Rate: of one country is exchanged for that of another country
(Moffett, Stonehill, &Eiteman, 2009). A bilateral real
exchange rate is calculated as the product of the
nominal exchange rate and relative price levels in each
country (Ellis, 200 I).
Foreign Direct Investment: "FDI is an investment of foreign assets into domestic
structures, equipments, and organizations" (Aslanov,
Gasimov, & Isayeva, 2010)

6
 CHAPTER II

REVIEW OF RELATED LITERATURE AND STUDIES . ,.

This section presents theories that are relevant and helpful in determining
factors affecting Vietnam ' s exports. More importantly, results and findings from
various empirical studies on the determinants of exports in both Vietnam and other
countries are discussed in details. Since this study emphasizes on the gravity model,
the literature review focuses more on this model and empirical papers applying it.

2.1 Theoretical Review

2.1.l Comparative Advantage Theory

Comparative advantage theory states that each country would be specialized in

producing particular products for which it possess absolute advantage, then it can
exchange those products for goods which are produced cheaper in other countries
(Moffett et al. , 2009).This theory provides a basis for explaining and justifying why
international trade- import and export occurs, and why particular countries are most
suitable for exports of particular goods and services. Clearly, a nation's export is
induced by its owing distinct advantages in production, such as cheap labor cost, high
technology, etc. in comparison with other countries.

2.2Article Review

2.2.l Gravity Model

The gravity model was first introduced by Tim bergen ( 1962) and Linneman
( 1966) to explain the international flows of trade. However, only empirical evidence
was provided at that time. Later, it has been widely adopted by various researchers to
analyze patterns and performance of international trade. Despite of its lack of

7
theoretical foundation, the gravity model has brought considerable empirical
robustness and explanatory power for depicting trade flows, said Porojan (2001).
The model applies the universal gravitation law that was discovered in 1687
by Newton who proved that any two objects exert a gravitational force of attractive on
each other. The magnitude of the force is proportional to the product of the
gravitational masses of the objects, and inversely proportional to the square of the
distance between them. This law was further applied to bilateral trade between two
countries which states that the bigger and closer countries are to each other, the more
trade will be conducted between them (Eita, 2008).
In McCallum's study (1995), which is the basic foundation for this research
paper' s model, he studied exports from each Canadian province to other provinces or
to states in the U.S. using the 1988 data. The model is generalized below in linear
form:
Ln Xi.i = a + ~ rln Yi + ~i ln Y.i + j33bij + ~4lndij + Eij,
where Xij is the exports from country i to country j ; Yi(Yj) is the GDP of the country i
U); Oijis a dummy variable that is equal to unity if trade is between two Canadian
provinces and zero otherwise; dij is the distance between any two provinces or states.
He found out that the domestic trade among Canadian provinces is 22 times larger
than the cross-border trade between the Canadian provinces and the U.S. states. This
is due to the border effects (U.S.-Canada). In other words, trading activities between
two studied destinations were affected by the distance between them.
In case of Vietnam, Bac's research paper (2010) is found to be closest to this
study since he also applied gravity model on a panel data from 1986 to 2008 covering
15 largest exporting markets of Vietnam. He developed two gravity model- static and
dynamic one. His models were modified from models of McCallum (1995), and
Harris and Matyas ( 1998). The static gravity model is represented as follows:
LnEXPOi1 = <D, + <D2lnINC, + <D3lnPINC1 +<D4lnREMOTi + <l>sASEANi 1+
<D6EXCHi1 + Ljt,
where EXP0i 1 is Vietnamese exports to country i 111 the year t; INC. (PINCii) is
Vietnamese income (importing country's income) proxied by GDP in the year t;
REMOTi is the distance from Ho Chi Minh city to the biggest economic centre of
country i; ASEANi 1 is a dummy variable that is equal to unity if country i is the

8
member of ASEAN and 0 otherwise, EXCHit is the average real exchange rate
between Vietnam's currency and country i' s currency. The dynamic gravity model
was modified by adding the independent variable- EXPOii-i, which is Vietnam's
exports to country i in year t-1. The results showed that the dynamic model could fit
the data better. There is a positive correlation between Vietnam's exports and its
importers' income. Transportation costs (proxied by distance) had a significantly
negative effect on Vietnam's exports. Other factors, such as exchange rate and
ASEAN membership also played an important role in determining Vietnam's exports.
Additional variables were also included. For instance, in their researches,
Bergstrand (1985) included population size, Oguledo and Macphee (1994) added
price variable, and Harris and Matyas (1998) added exchange rates, into the model. In
this study, besides basic variables- GDP and distance, other variables, which are
discussed in the following part, are added in order to improve the basic formulation
and better explain the dependent variable.

2.2.2 Dependent Variable

The dependent variable in this paper is Vietnam's exports value to each forty
major exporting markets. There are some empirical researches done to find out the
factors determining exports of Vietnam. Factors which have positive impacts on
Vietnam ' s exports are commonly recorded as Vietnam' s GDP (Bae, 2010), Vietnam ' s
FDJ (Xuan & Xing, 2006), importing country ' s GDP (Bae, 2010) and income per
capita, etc. On the other hand, factors having an inverse relationship with Vietnam
exports are commonly found to be Bilateral exchange rate (Hanh & Due, 2009), and
Distance from Vietnam to its exporting markets (Bae, 20 I 0), etc.

2.2.3 Independent Variable

Independent variables are classified into export side factors and impo11 side
factors.

9
Export side factors

Those factors reflect the export supply capacity of the exporting country,
which include exporting country's GDP, exporting country's FDI,real bilateral
exchange rate, and a dummy variable - FTAs between exporting country and each
importing country. According to Fugazza (2004), countries having superior supply
conditions are anticipated to export more.

2.2.3.1 Exporting country's GDP

GDP is considered as a proxy of economic mass, which is a basic variable of

gravity model. Bhagwati (I 988) noted that an increment in GDP is generally
corresponded by an expansion of trade. This can be understood in a way that when a
nation produces more products, their exporting capacity will also increase.
Most empirical papers around the world support the idea that GDP has a
positive impact on exports. In his paper, Eita (2008) found out that an increase in
Namibia' s GDP caused exports to increase during the examination period of 1998-
2006. In a more recent study. Aslanov et al. (2010) also found out that an increase in
GDP Jed to higher exports, and vice versa, of three countries in South Caucacus,
including Azerbaijan, Georgia and Armenia. Aurangzeb (2012) got the same result in
Pakistan but basing on a longer time series data from I 981 to 2009. Majeed and
Ahmad (2006) also conducted their studies in 75 countries and the results were
asserted again that there is a positive relationship between GDP of a country and its
exports.

2.2.3.2 Exporting country's FDI

There has never been a single conclusion about the effect of FDI on export
from previous studies because they showed different results. Some papers concluded
that FOi inflow has a positive impact on a national exports while other found negative
or insignificant impact.

10
 THE ASSUMPTION UNIVERSITY LIBRAd

It was found out that trade between South East Asian countries and the United
States and Japan is positively impacted by inward FDI (Linda & Michael, 1998).
Gunawardana and Sharma (2009), and Pemasiri and Sharma (2010) also pointed out
that inward FDI has a significant positive influence on Australian manufacturing
exports in the long-term. Specifically for Vietnam, Xuan and Xing (2006) concluded
that FDI is one of the major factors that led to the rapid growth of Vietnam's exports
from 1990-2004. Hanh and Due (2009) got a similar result. It states that higher FDI
lead to higher export from 1990-2007. Some studies (Lipsey& Weis, 1984; Anwar &
Nguyen, 20 I 1) further showed that FDl and export are supportive of each other.
ln contrast, Jeon (1992) and Sharma (2000) proved a negative relationship
between exports of the host country and its inward FDl in Korea and India
respectively. It is explained that those foreign-funded enterprises want to obtain the
domestic market share by moving products for exports back to that host country (Liu
& Shu, 2003). Furthermore, the relationship was found to be insignificant in Central
and Eastern European countries by Hoekman and Djankov (1997), and in Ethiopia by
Taye (2009).
4222 Q,_ -1
2.2.3.3 Real bilateral exchange rate

Exchange rate is a fundamental determinant of a country's trade. Therefore

there are numerous empirical studies being conducted on how this variable would
affect a country's exports. In fact, some studies show positive impact of a national
currency depreciation on its export while others show no impact. There are various
results from previous research papers.
ln their papers, Chen, Rau and Chiu (20 I I). and Aljebrin (2012), supported
the idea that a fall in currency value makes costs of domestically produced goods
lowered , which consequently enhances competitiveness in export in China and Saudi
Arabia, respectively. In their research about Pakistan' s exports, Najia and Irfan (2012)
also proved that the nation' s currency value influence its export volume inversely. In
their papers, Tilak and Tan ( 1998) further explained that an appreciation in national
currency would damage exports. However, it would be reduced due to the existence
of high import content in the exporting products. For Vietnam specifically, Hanh and

I1
Due (2009) concluded that a real depreciation of Vietnamese currency versus the
foreign currency led to increments in exports basing a panel data from 1990 to 2007.
The results do not always show that there is an inverse relationship between
exports of a country and its currency value (proxied by bilateral exchange rate). In
fact, in many papers, it was proven that there is no causality between these two
variables. For instance, Miyao (2003) found no causality between exchange rate and
exports in the mid 1980s in Japan, while Alam (2010) also found no impact of a real
depreciation of Taka on exports of Bangladesh.

2.2.3.4 FTAs between exporting country and each importing country

FTAs have been widely used among countries in order to facilitate trading or
reducing trading barriers among them, such as North American Free Trade Agreement
(NAFT A), and ASEAN FTA. Predicting effects of the FT As on trade among involved
parties has become one of major application of the gravity model (Baier &
Bergsrtand, 2007). Most research papers showed that the existence of FTAs helped
increase exports the (trading) value of a country to its FTA partners.
Assarson (2005) found out that South Africa's exports to the European Union
increased by 75% between 1999 and 2004 due to the creation of European Union -
South Africa free trade agreement. In another paper, Korinek and Melatos (2009)
showed that the creation of FTAs increased the trade of agricultural products (proxied
by exports) among those countries involved, by applying the gravity model. Hur,
Alba, and Park (20 I 0) also proved in their researches conducted in 96 countries
covering the period 1960-2000 that FTAs have a significant positive influence on
ex ports between FTA partners.
However, it was interestingly found that Vietnam ' s exports were affected
negatively by the ASEAN free trading agreements (Bae, 2010). This study also
appli ed the gravity model on a data period of 1995-2006. This could be explained that
exports of Vietnam to A SEAN are basically underexploited. In his research, Hatab et
al. (20 I 0) even found that the regional trade agreements could not help in determining
exports volume of Egyptian agricultural products (i.e. there is no statistically
significant relationship between exports and trade agreements).

12
Import side factors

These factors describe the accessibility of foreign markets and their importing
capacity. Redding and Venables (2003) proved that supply capacity conditions and
foreign market access are equally vital for export development of a country. Import
side factors are comprised of importing country's GDP, importing country ' s GDP per
capita, and distance.

2.2.3.5 Importing country's GDP

While exporting country' s GDP represents productive capacity of exporters,

importing country's GDP measures absorptive capacity of importers (Hatab,
Romstad, & Huo, 20 l 0). Even if there has been conflicts in results of empirical papers
on the relationship between the exports of one country with the GDP of its trading
partners, a positive relationship between those two variables are expected to have.
In most researches applying the gravity model , the results revealed that an
increase in importing country's GDP would lead to an increase in the export volume
of the exporting country. For example, in the study, which tested 20 countries
including European Union members, Chile and four members of Mercosur plus,
Martinez-Zarzoso and Nowak-Lehman (2001), by gravity model, found out that GDP
of importing countries has a positive effect on export value of exporters. In the case of
Vietnam, Bae (2010) again confirmed that the importing country ' s GDP has a positive
relationship with exports (of Vietnam), by applying the gravity model. In other
studies, using different models, the results still supported the positive relationship
between these two variables. For instance, Ong, Yoong, Lim, and Tong (2009) found
out that the real GDP of Japan and the US have a positive impact on Thailand ' s real
exports. while Anh and Thang (2008). and Tien (2009) found similar results in
Vi etnam .
However, it was surprisingly found in another research done byTrang et al.
(2010) that GDP of importing countries do not affect Vietnam's exports of different

13
product groups much. Their paper also used the gravity model with a panel data from
2004 to 2008 on Vietnam's 61 importing countries.

2.2.3.6 Importing country's GDP per capita

Importing country's GDP per capita is a proxy of country developing and

market potential (Fugazza, 2004). It is normally expected that an increase in GDP per
capita of importing country will lead to a greater demand for products from the
exporter, said Ha ta bet al. (2010).
There are various results regarding the relationship between exports of a
country and GDP per capita of its importers. The variety in the empirical results
comes from the fact that different countries exporting different products, which are
consequently influenced differently by GDP per capita of their importing partners.
Hermawan (2011) applied both standard and augmented gravity model to find
out the determinants of Indonesia's textile products exports. The paper found out that
there was a positive and significant relationship between trading partners ' per capita
income (proxied by total GDP/population)and Indonesia's exports of textile products.
In contrast, Eia (2008) found out that an increase in importer's GDP per capita
led to a decrease in Namibia's exports to its 38 trading partners basing on annual data
from 1998 to 2006. In other words, there is an inverse relationship between the two
variables in Namibia.
In their studies, Hatab et al. (2010) found out that Egyptian agricultural
exports to their major trading partners from 1994 to 2008 were not statistically
influenced by the importer's GDP per capita. Even if this study only covered Egyptian
agricultural exports rather than total exports, it could still be a good reference for
Vietnam because agricultural products are also one of the country's major export
products.

2.2.3. 7 Distance

Distance between an exporter and its importers is used as a proxy for

transportation costs. This is the basic variable of the gravity model. It is normally

14
stated that as counlries slay far rom ach oth r the tnmsp rtation costs etween them
ar higher. nsequently they tend to trade less.
Eita (2008 pro ed in his study thatNambian exports decrea as the distance
g t larg r. Hatab et al. 2010 found in their research that transportation c sts, proxied
by distance, iniluen ed Egyptian agricultural exp rts negati ely ase on time series
data fr m I 94-20 8. Weldemariam 2009 and rindi 2011 also Ii und out that
exp rls reduced a ' distance b tl een th m and their imp rters g t larger in thiopia
and Kenya respectively. In the case of Vietnam, ac (2010) con luded 1n his research
which was conducted fr m 1< 1 to 2006, that an increase in ge graphical distance
b tween Vi tnam nd its trading partners tends to reduce Vietnam s export t those
countries. imilarly, Rahman (2 JO found out that the distance had a negative, ut
Country's inward Weis, 1984 • Sharma, 2000 Djankov, 1997
FDI •Linda and • Taye, 2009
Michael, 1998
• Xuan and Xing,
2006
• Gunawardana
and Sharma,
2009
• Hanh and Due,
2009
• Pemasiri and
Sharma, 2010
•Anwar and
Nguyen, 2011
• Tilak and Tan,
1998
•Tri , 2006
• Khedhiri and
Bouazizi , 2007
• Miyao, 2003
3. Real Bilateral • Chen et al., 2011
•Alam, 2010
Exchange Rate • Aljebrin, 2012
• Najia and Irfan,
2012
• Hanh and Due,
2009

• Korinek and
4. Free Trade Melatos, 2009 • Hatab et al. ,
• Hur, Alba, and •Bae, 2010 2010
Agreements
Park, 2010
• Assarson, 2005
• Zazoso and
Lehmann, 2001
• Khedhiri and
Bouazizi, 2007
•Anh and Thang,
5. Importing • Trang et al.
2008
Country' s GDP 2010
• Ong et al. , 2009
•Tien. 2009
•Bae, 2010
•Ibrahim, 2011
• Alj ebrin, 201 2
6. Importing • Hatab et al. ,
• 1-Iermawan,
country' s GDP per • Eita, 2008 2010
2011
capita
7. Distance • Eita, 2008 •Rahman,

16
 • Weldemaria, 2010
2009
• Hatab et al.,
2010
-
•Bae, 2010
• Orindi, 2011

Referring to findings and results of empirical papers about the relationship

between exports and each independent variable, Table 2.2 describes an expectation
about the relationship between Vietnam's exports and each considered independent
variable in this study. It is predicted that Vietnam's exports will have a significantly
positive relationship with Vietnam's GDP, Vietnam's FDI, importing country' s GDP,
importing country's GDP per capita, real bilateral exchange rate, and FTAs variables.
In contrast, Vietnam's exports are anticipated to have a significantly negative
relationship with distance variable.

Table 2.2: Expectations of the Relationships between the Dependent Variable

and Independent Variables
Independent Expected Relationship with Vietnam ' s exports
Variables Positive Negative Insignificant
1. Vietnam ' s GDP
. x
.·~ i_
- .,~

2. Vietnam's FDI
:..;u
··x. ""

,·
·.
3. Real Bilateral < <

Exchange Rate ~ c ;;:

x •.
'·
4. Free Trade
~

..
Agreement between
.'

Vietnam and its x

forty exporting "' '
markets
5. lmporting
"
·' ' .'';?''"1i ·'
....."
country ' s GDP
x
6. Importing '
country ' s GDP per x
capita
7. Distance
x
Source: Author (2013)

17
 CHAPTER III

RESEARCH METHODOLOGY

This chapter describes in details the process of data collection, methodology,

and hypotheses of the research.

3.1 Data Collection

The annual data covered forty importers for the period from 1995 to 2011 with
one dependent variable and six independent variables, and a dummy variable (a total
of n = 680, N = 40, and T = 17). All variables were expressed in natural logarithm
except the real bilateral exchange rate and the dummy variable. The forty major
importers of Vietnam are listed in the appendix, which altogether contributes
approximately 80%-90% of Vietnam's exports each year during the examination
period.
The dependent variable, Vietnam ' s exports to each country, were collected
from Vietnam ' s General Statistics Office (http://www.gso.gov.vn). While the data on
GDP, GDP per capita, nominal bilateral exchange rate, Consumer Price Index (CPI),
and FDI were collected from World Bank database (http://data.worldbank.org). Since
the nominal bilateral exchange rate between Vietnam Dong and other currencies were
not available. they were calculated indirectly by dividing the VND/ 1USD rate for X/
1USD, where X is the number of an importing country's currency units per one unit
of USD. The real bilateral exchange rate was calculated using the following formula
developed by Ellis (200 I):
reri ,t = e·l ,t x P*u
Pc ,

where P, is the consumer price index of Vietnam at time t; P*i.1 is the consumer price
index of foreign country i at time t: and ei.i is the nominal bilateral exchange rate
between Vietnam dong and foreign currency i at time t, which is expressed as the
number of Vietnam Dong units per one unit of foreign currency i. The base year for
CPI is 2005.

18
 The dummy variable- FTAs was especially collected from Vietnam Trade
Promotion Agency (http://www.vietrade.gov.vn/en/). The distance between Hanoi and
other studied capital cities were taken from the Indo's website
(http://www.indo.corn/distance/).

3.2 Methodology

3.2.1 Specifications of the model

Bac's model (2010) is the most relevant foundation for developing this paper' s
model. Two more variables were added, which are Vietnam's inward FDI value and
GDP per capita of importing countries. The FTAs and distance variables were slightly
modified from Back' s study.
FDI has been commonly studied in empirical papers and usually found to have
positive impacts on exports. Inwards FDI is expected to improve production capacity
of Vietnam, which consequently enhances its exporting capacity. Including FDI into
the model is anticipated to better explain the dependent variable - Vietnam ' s exports.
Importing country's GDP per capita is another variable that was added into the model.
Instead of the economic mass (proxied by GDP), the income level of residents in
exporting countries (proxied by GDP per capita) was also considered to see how this
variable influences Vietnam ' s exports.
Bae (2010) considers whether the studied trading partner is a member of
ASEAN or not, whereas in the present paper, the free trade agreements between
Vietnam and the studied trading partners were considered. It is perceived to provide
better and wider information about the effects of trade agreements on Vietnam ' s
exports.
Distance variable used in the present paper is the geographical distance
between Hanoi (Vietnam ' s capital) to the biggest economic center of each trading
partner. Jn Bac 's paper. Ho Chi Minh city was used inslead.
After making adjustments and modifications, the model of the present study is
developed as shown below:

19
 lnEXPit = Jlo+ P1lnVNGDP, + P2lnICGDPi1 + fJ3lnVNFDI, + fJ4REXCHit+
1lslnICGDPPCi1+ IJ61nDISi + ll1FT Asi + µit,
where lnEXPit is logarithm of Vietnam's total exports to country i at the year t;
lnVNGDP1 is logarithm of Vietnam's GDP at the year t; lnICGDPi 1 is logarithm of
importing country i' s GDP at the year t; InVNFDI 1 is logarithm of Vietnam ' s FDI at
the year t; REXCHit is the real bilateral exchange rate between Vietnam Dong and
country i' s currency at the year t; lnICGDPPCit is logarithm of importing country i' s
GDP per capita at the year t; lnDISi is logarithm of distance between Hanoi capital to
each importing country' s biggest economic center; FTAsi is a dummy variable, which
is equal to unity if Vietnam and a importing country has any form of FTA, and 0
otherwise; µi 1is the error term, and ~ 0 ___ ~ 7 are parameters to be estimated. Table 3.1
provides measurements on those variables in details.

Table 3.1: Measurement of Variables

Symbols Descriptions Measurements
Total value of Vietnam ' s exported goods
Vietnam' s total exports
EXP it and services to country I at the year t,
to country i
which is measured in US dollar.
Total value of final goods and services
Vietnam ' s Gross being produced within Vietnam for a
VNGDP,
Domestic Product specific period of time (a year), which is
measured in US dollar.
Total market value of investment from
foreign companies and countries that
Foreign Direct
VNFDI, Vietnam receives for a certain period of
Investment into Vietnam
time (a year), which is measured in US
dollar term .
It is measured by multiplying the nominal
Real Bilateral Exchange bilateral exchange rate between Vietnam
REXCH; 1
Rate Dong and the foreign currency i with a
ratio of foreign country's CPI divided by

20
 Vietnam's CPI at time t.
The geographical distance between tJanoi
Distance between Hanoi
and each importing country ' s biggest
DI Si to capital city of the
economic center, which is measured in
importing country i
kilometer term.
Total market value of final goods and
Importing Country i's
ICGDPit services being produced in country i at the
Gross Domestic Product
year t, which is measured in US dollar.
It is measured by dividing country i's
Importing Country i's
ICGDPPC, GDP for its midyear population, which is
GDP per Capita
in dollar term.
It is equal to 1 if there is any forms of free
Free Trade Agreements
trade agreements between Vietnam and
FTAi1 between Vietnam and
each studied forty exporting markets, and
Importing Country i
0 otherwise.

3.2.2 Estimation process

In this study, it was not necessary to run serial correlation tests because it was
only a problem in macro panels with over 20-30 years data (Oscar, 2013). According
to Oscar (2013), cross-sectional dependence problem is also not much of a problem in
micro panels. Furthermore, unit root tests for micro panels (i.e. small number of time
periods and a large number of cross-sectional units), do not receive much concerns
from researchers like that for macro panels (Baltagi & Kao, 2000). Therefore, those
tests were exempted.
The Breusch-Pagan Lagrange multiplier test (LM test) was run in order to
decide whether a random effects regression or a simple OLS regression should be
used. The null hypothesis is that variances across entities is zero (i.e. a simple OLS
regression is appropriate). If the null hypothesis of LM test is rejected (i.e. p-value is
less than 0.05), the random effects regression should be used.
The Hausman test was used to examine whether the random effects is
consistent and etlicient. If the null hypothesis of Hausman test is rejected (i.e. p-value

21
 THE ASSUMPTION UNIVERSITY LIB~

is less than 0.05), the random effects is not appropriate and the fixed effects model
should be used instead. One of the problems associated with the fixed effects mo?el is
that time-invariant variables like distance wi11 be automatica11y omitted, which is not
a case for random effects model. Then the coefficient of distance variable wi11 be
especia11y reported basing on the random effects model.
Heteroskedasticity causing estimators inefficient, 1s only available to be
checked in the fixed effects model. The null hypothesis is that variance is constant. If
the nun is rejected (i.e. p-value is Jess than 0.05), there exists the heteroskedasticity
issue. The "robust" option was used to control for heteroskedasticiy in both random
and fixed effects model.

3.3 Research Hypotheses

Each independent variable might or might not have an impact on the

dependent variable. In testing this, hypotheses were set for each individual
relationship between the independent variable and the dependent variable.

HI 0 : There is no significant linear relationship between Vietnam ' s exports and

Vietnam's GDP.
H 1a: There is significant linear relationship between Vietnam's exports and
Vietnam ' s GDP.

H2 0 : There is no significant linear relationship between Vietnam ' s exports and

importing country ' s GDP.
H2a: There is significant linear relationship between Vietnam's exports and
importing country's GDP.

1-Bo: There is no significant linear relationship between Vietnam ' s exports and
Vietnam's FDI
H3a: There is significant linear relationship between Vietnam's exports and
Vietnam ' s FDJ.

22
 H4 0 : There is no significant linear relationship between Vietnam's exports and
Vietnam's real bilateral exchange rate.
H4 3 : There is significant linear relationship between Vietnam's exports and
Vietnam's real bilateral exchange rate.

H5 0 : There is no significant linear relationship between Vietnam's exports and

importing country's GDP per capita.
H5 3 : There is significant linear relationship between Vietnam's exports and
importing country's GDP per capita.

H60 : There is no significant linear relationship between Vietnam's exports and

distance between Vietnam and importing country.
H6 3 : There is significant linear relationship between Vietnam's exports and
distance between Vietnam and importing country.

H7 0: There is no significant linear relationship between Vietnam's exports and

FTAs dummy variable.
H7 3 : There is significant linear relationship between Vietnam's exports and
FTAs dummy variable.

23
 CHAPTER IV

PRESENTATION AND CRITICAL DISCUSSION OF RESULTS

This chapter discusses in details the results after estimating the equation based
on the methodology. There are two main parts. Part l reports the results of the
Breusch-Pagan Lagrange multiplier test, Hausman test, and testing for
Heteroskedasticity. Part 2 discusses the estimated gravity regression using the
preferred model from Hausman test. Relationships between the dependent variable
and each independent variable are also analyzed in details in this part.

4.1 Results of the Breusch-Pagan Lagrange Multiplier Test, the Hausman Test,
and Checking for Heteroskedasticity

Table 4.1 represents the results of the Breusch-Pagan Lagrange multiplier test.
It indicates that a random effects regression is more appropriate than a simple OLS
regression (i.e. the null hypothesis of zero variances across entities is rejected).

Table 4.1: Breusch-Pagan Lagrange multiplier test results

Breuscn ano Pagan Lagrangian mu1t1p11er test for random errects
logexp[countryc,t] = Xb + u[countryc] + e[countryc,t]

Estimated results:
var sd = sqrt(Var)
logexp .656521 .8102598
e .0498073 . 2231756
u .095502 .309034
Test: var(u) = O
sh "i!(i) lbM . 8!1
~b > chi2 = 0.0000 ~

Table 4.2 represents the Hausman test results, which indicates that the fixed
effects model is preferred to the random effects model because the probability is less
than 0.05 (i.e. the null hypothesis of a consistent and efficient random effects model is
rejected). In other words, the random effects model is suffering from the violation of
the Gauss-Markov theorem and end up with biased and inconsistent estimates, in

24
contrast, the fixed effects still remains unbiased and consistent (Park, 20 I 0). Since the
fixed effects model is chosen based on Hausman test results, the time-invariant
variable like distance will not be estimated (Oscar, 2013).

Table 4.2: Hausman test results

- - Coeffi ci ent:s - -
(b) (8) (b-8) sqrt:(diag(v_b-V_8))
fixed random Difference S . E.

logvngdp 1.725462 1.715533 .0099296 .0691846

logvnfdi -.2608017 -.3178508 .0570491 .0126177
ft: as -.0343612 .0338501 -.0682113 .0166566
rexch 9.23e-06 .0000107 -1.45e-06 1.07e-06
logicgdp 2.564262 .6337591 1.930503 .5195346
logicgdppc -2.405171 -.1518352 -2.253336 .5081368
b = consist:ent: under Ho and Ha; obt:ained from xt:reg
8 = inconsist:ent: under Ha, efficient: under Ho; obt:ained from xt:reg
Test:: Ho: difference in coefficient:s not: syst:emat:ic
chi2(4) = (b-8)'[(V_b-V_8)A(-l)](b-8)
r~ ~-

crob>chi2 = 0.0000
::::>

The test result for heteroskedascityis displayed in Table 4.3, which shows that
there is a presence of heteroskedascity in this study~ s panel data (i.e. p is less than
0.05). ln other words, the variance is not constant. As mentioned in the estimation
process, the "robust" option can be used to solve for this problem.

Table 4.3:The result of heteroskedascity check

Modified Wald t:est for groupwise het:eroskedast:icity

in fixed effect: regression model
HO: sigma(i)A2 = sigmaA2 for all i

4.2 Gravity Model Estimation Results

Table 4.4 represents the estimated equation by both the random effects model
and the fixed effects model. According to Hausman test results in Table 4.2, the fixed
effects model is better for estimating the relationships between Vietnam ' s exports and

25
other seven independent variables. Accordingly, interpretation of the results focuses
more on the fixed effects model.

Table 4.4: Gravity Model Estimation Results

Variables Fixed Effects Model Random Effects Model
logvngdp 1.725462 1.715533
(8.98) *** (9.84) ***
logvnfdi -0.2608017 -0.3178508
(-2.81) *** (-3.38) ***
logicgdp 2.564262 0.6337591
(2.11) ** (5.99) ***
logicgdppc -2.405171 -0.1518352
(-2.01) (-0.83)
logdis ---- -I .070503
(-4.85) ***
ft as -0.0343612 0.0338501
(-0.34) (0.34)
rexch 9.23e-0.6 0.0000107
(2.34) ** (2.77) ***
constant -27.57177 -9.78504
(-3.23) *** (-6.32) ***
Adjusted RL 0.7942 0.7841
Observations 680 680
Notes:***/**/* significant at 1%/5%/10% level. z and t-statistics are in parentheses.

Table 4.4 shows that under the fixed effects model , the dependent variable-
Vietnam ' s exports has statistically significant relationships with five independent
variables- Vietnam·s GDP, Vietnam ' s FDI, importing country·s GDP. real bilateral
exchange rate, and di stance because their p-value is less than 0.05. Jn contrast, it has
no statistically significant relationships with importing country·s GDP per capita and
FTAs because their p-value is more than 0.05. The detail of the gravity regression
results with the fixed effects model are shown in Appendix 2.

26
 The coefficient sign for Vietnam's GDP variable is reported to be positive as
expected. In other words, holding all other independent variables constant, a 1%
increase in Vietnam's GDP causes a 1.72% increase in Vietnam's exports
approximately. This finding is consistent with the findings of Eita (2008), Bae (2010),
and Aurangzeb (2012) that there is a positive relationship between exports of a
country and its GDP. Looking at the coefficient of importing country's GDP, it is also
reported to be positive as expected. This means that holding all other variables
constant, a 1% increase in importing country's GDP leads to an approximate increase
of 2.56% in Vietnam's exports. This result is consistent with the findings of Anh and
Thang (2008), Tien (2009), and Bae (2010) that there is a positive relationship
between the exports of a country and GDP of its trade partners. Under the fixed
effects model, the time-invariant variables like distance were automatically dropped.
In other words, their effects cannot be estimated. Since distance is one of the main
variables of gravity model, its effect should be considered in this paper. Therefore, the
coefficient of distance, was taken from the random effects model estimation, and was
recorded to be negative as predicted. This finding is consistent with the literature
about the gravity model, that as a country stays far from its trade partners (i.e.
transportation costs are higher), it tends to trade less and vice versa (Eita, 2008 ; Hatab
et al., 201 O; Bae , 2010). These findings of positive relationships between Vietnam' s
exports and Vietnam' s GDP, and between Vietnam's exports and Importing country's
GDP, and negative relationship between Vietnam's exports and distance, are
consistent with the main idea of the gravity model that as the size of economies
increases and geographical distance reduces, bilateral trades are increased.
The coefficient of Vietnam ' s FDI is found to be negative, which is not the
expected result of this study. This result means that holding all other variables
constant, a 1% increase in inward FDI into Vietnam causes an approximate 0.26%
decrease in its exports. This finding is supported by Jeon ( 1992) and Sharma (2000).
As explained by Liu and Shu (2003) that foreign-funded enterprises might want to
obtain and dominate domestic markets rather than exports. Therefore they move
products for exports back. This might also be the case of Vietnam where inward FDI
is for the purpose of obtaining domestic market rather than exports. Therefore there is
a negative relationship between Vietnam ' s exports and its inward FDI.

27
 The coefficient for the importing country's GDP per capita variable is
negative. This means that an increase in the importing country's GDP per capita
-
causes Vietnam's exports to decrease. However, as mentioned earlier that the p-value
of this variable is more than 0.05 indicating that importer's GDP per capita is not a
statistically explanatory variable for exports (Hatab et al., 2010). This result is not
consistent with the study's expectation.
The FTAs variable is also found to have no statistically significant impact on
Vietnam ' s exports because the p-value is more than 0.05 . This result is contradictory
to this study's expectation. Most of the FTAs of Vietnam are between Vietnam and
ASEAN countries, which is considered a small market for Vietnam ' s exported
products in comparison with non-ASEAN countries like the U.S ., European, China,
and Japan. As Vietnam is more integrated into trades with non-ASEAN countries than
that with ASEAN members, this might explain why the FTAs variable is not
statistically significant enough to explain the development of Vietnam ' s exports. In
other words, trade gains from FTAs has been minimal (Hatab et al. , 2010).
The coefficient for real bilateral exchange rate is found to be positive as
expected. It means that an increase in the real bilateral exchange rate (i.e. Vietnam
dong depreciates against foreign currencies), causes an increase in Vietnam 's exports.
This finding is also consistent with most empirical papers, such as Tri (2006), and
Hanh and Due (2009) that there is a negative relationship between a country' s exports
and its currency value.

28
 CHAPTERV

CONCLUSION, IMPLICATIONS AND FURTHER STUDY

After the Doi Moi policy was launched in 1986, Vietnam has become more
integrated into the global trades. Parallel to this, Vietnam's exports have gained more
successes and become more important to its economy. Recognizing the importance of
exports in Vietnam's economy, this study was designed to find out the factors
influencing Vietnam' s exports to its forty major exporting markets.
The gravity model approach was chosen to find out the factors that have
impacts on Vietnam' s exports, which is considered as one of the most successful
application in explaining bilateral trades. The period of 1995 to 2011 was chosen
because the study geared towards providing the most updated results. Besides basic
variables of the gravity model- Vietnam ' s GDP, importing country's GDP and
distance, additional variables including Vietnam ' s FDI, importing country's GDP per
capita, real bilateral exchange rate, FTAs, were included in order to improve the basic
formulation and better explain the dependent variable.
The Hausman test has been conducted and it was found out that the fixed
effects model is preferred than the random effects model to estimate the gravity
regression. Therefore, the results have been interpreted based on the fixed effects
model estimation. The effect of distance variable was especially explained using the
coefficient from the random effects model because it could not be estimated in the
fixed effects model.
The results have shown that Vietnam ' s exports patterns follow the basic
gravity model. In other words. Vietnam's exports increases in proportion to its GDP
and importing countries' GDP, and it decreases in proportion to its distance to those
trading partners. The results imply that in order to expand its exports. Vietnamese
government and exporting companies should focus on promoting exports to rich
economi es, which are located in a close distance. Furthermore. since a long di stance,
which is a proxy of high transportation costs. is a barrier that discourages trade
between countries, finding ways to reduce the transportation costs is important in

29
overcommg such barrier. Key points to reduce transportation costs included the
improvement of transportation infrastructure as well as logistics system.
Vietnam's FDI is reported to have a significant negative relationship with its
exports. This finding indicates that foreign investments into Vietnam might be for
obtaining domestic markets rather than for producing exports. The similar results have
been found in other countries like Korea (Jeon, 1992) and India (Sharma, 2000). In
fact, this relationship is subject to changes if another research period was chosen
because the objectives of foreign-funded enterprises might have changed then. In
contrast, the importing countries' GDP per capita was found to be a statistically
insignificant determinant of Vietnam's exports. This implies that exports follow the
GDP pattern centering on the overall economy size, rather than the GDP per capita
pattern centering on the richness of residents in the importing countries (Hatab et al.,
201 O).The FTAs dummy variable was also found to be an insignificant factor in
determining Vietnam ' s exports. This result indicates that the trades gained from FTAs
were too small to explain Vietnam's exports. The result further implies that Vietnam
is under-exploiting its advantages of holding free trade agreements with countries in
the region because most FTAs were found to be between Vietnam and ASEAN
members. However, the reason might be that most countries in the reg10n are
exporting quite similar products. Thus, bilateral trades among them cannot be
developed further even if they hold free trade agreements. In short, it is expected that
in the future when Vietnam succeeds in signing FTAs with big trading partners, such
as the European members and the United States, the FTAs variable can have a more
explanatory power to Vietnam ' s exports. It is not surprising that the real bilateral
exchange rate has a significant and positive relationship with Vietnam ' s export. In
other words, a depreciation of Vietnam dong versus importing countries' currencies
causes an increase of Vietnam ' s exports. This result indicates that in order to achieve
its export goals, Vietnamese government can take appropriate actions in response to
depreciation or appreciation of its currency against foreign currencies.
In a report done by World Bank (2013c ), it shows that Vietnam 's total ex ports
rose by 16% in the first half of 2013 compared to the same period last year. This
increase in exports could be projected basing on either its positive relationship with
Vietnam 's GDP as found in this study, which is recorded to grow at 5.25% in the

30
second quarter of 2013, or its negative relationship with Vietnam Dong value as
proven in this study, which is reported to depreciate by 1.6% over the past 12"'months
(World Bank, 2013c). This shows that up to a certain extent the findings of this study
can explain the growth in Vietnam's exports in the first half of2013.
Thus, the results of this research paper regarding the determinants of
Vietnam's exports might be useful and supportive for both Vietnamese government
and exporting companies in setting their export goals and policies. However, referring
the study alone is not enough. In other words, besides the findings of this study, the
Vietnamese government and exporting companies should conduct researches on many
other aspects related to exports in order to have a better inference. This includes the
development of Vietnam's transportation infrastructures, the effect of domestic
demand, or the competition from other exporters, etc. Furthermore, more
comprehensive policy recommendations can be configured to improve the
performance of Vietnam' s exports in the international market. Also, it should be noted
that the costs and benefits of each action should be taken into consideration carefully.
For instance, a depreciation of Vietnam Dong might help to increase exports, but it
raises the burden of foreign debts on both the government and exporting firms.
Further researches should be conducted on the basis of solving the limitations
of this study. It can expand to study more variables, such as competition from China's
exports and the importance of logistics, to provide better inference, or expand the
sample size by increasing either number of importing countries or number of time
periods to draw more accurate results. The study can also be modified to find out the
determinants of Vietnam ' s exports to specifically interested exporting market like the
U.S. or Japan.

31
 BIBLIOGRAPHY

Anh M. 2012). Tangluang, xufttkhiu 'cliu' G P 2012. Retrieved April 5

20 l", fr m: htlp://ndhmoney. vn/web/guesl/s02/-/j um.al content/lang-lu ng-xuat-

khau-cuu-gdp-2012 .
Anh, T.T. & Thang, D.N. (2008 . ac nMin ti> anh hucmg toi mire di) 1~p trung
thucmg mQi cua Vi~t Nam vm Asean 3. Vietnam Ee n mic and P !icy R se. rch
center (VEPR .
Alam, R. 2 I 0 . The link betwe n reaJ exchange rate and export earning: a
coinlegration and granger causality analysi n anglad h. Internal Re iew of
Bu iness Research Papers, 6(1) p. 2 5-214.
Aljebrin I. . (2012). he determinants f Arab countries cmand for audi
exp rt : pan I data e idence. Resear h in World E onomy, 3 2). _Q-28.
Anwar .. & Nguy n .P. (2011). F reign direct investment and xp rt
spjllo er: viden e from Vietnam. International Bu iness Review, 20 p. f 77-193.
 1111!! ASSUMPTION UNIVERSITY LIBRAR't

Bergstrand, J.H. (1985). The gravity equation m international trade: some

microeconomic foundations and empirical evidence. The Review of Economics and
Statistics, 6 7(3), p. 474-481.
-.
Bhagwati, J. ( 1988). Protectionism. The MIT Press, Cambridge,
Massachusetts.
Chen, K. M., Rau, H. H. , & Chiu, R. L. (2011). Determinants of China' s
exports to the United States and Japan. The Chinese Economy, 44( 4 ), 19-41.
Dang, N.H . (2010). D6ng g6p cua xuAt kh~u vao tang truemg kinh t~, cua hoc;tt
d(mg xufit kh~u vao nang cao chAt luqng san phb. Retrieved April 14, 2013, from:
http://voer.edu.vn/bai-viet/kinh-te/dong-gop-cua-xuat-khau-vao-tang-truong-kinh-te-
cua-hoat-dong-xuat-khau-vao-nang-cao-chat-luong-san-pham.html .
Eita, J. H. (2008). Determinants of Namibian exports: a gravity model
approach. University of Namibian.
Ellis, L. (2001 ). Measuring the real exchange rate. RBA Research Discussion
Paper No. 2001-04. P. 70-73.
Fugazza, M. (2004). Export performance and its determinants: supply and
demand constraints. Policy Issues in International Trade and Commodities Study
Series No. 26.
Gunawardana, P. J. , & Sharma, K . (2009) . The impact of inward FDI , labour
productivity and industry assistance on manufacturing exports of Australia. Working
paper series, School of Economics and Finance, Victoria University, Melbourne.
Hanh, T.H.P., &Due, T.N. (2009). Foreign direct investment, exports and real
exchange rate linkages: Vietnam evidence from a cointegrtion approach. 1-24.
Retrieved April 25, 2013, from: http://www. univ-
orleans.fr/ gdre09/articles/PHAM .pdf.
Harris, M., &Matyas, L. (1998). The econometrics of gravity models.
Melbourne Institute WPS/98.
1-Jatab, A. A., Romstad , E., &Huo, X. (2010). Determinants of Egyptian
agricultural exports: a gravity model approach. Modern Economy, 1, I 34-143.
Hermawan, M. (2011 ). The detern1inant and trade potential of export of the
Indonesia' s textile products: a gravity model. Global Economy and Finance Journal,
4(2), 13-32.

33
 Hoekman, B., &Djankov, S. (1997). Determinants of export structure of
countries in Central and Eastern Europe. The World Bank Economic Review, 11(3), p.
471-487.
Huberman, G., & Wang, Z. (2005). Arbitrage pricing theory. Retrieved April
22, 2013, from: http://wwwO.gsb.columbia.edu/faculty/ghuberman/APT-Huberman-
Wang.pdf.
Hur, J., Alba, J. D., & Park, D. (2010). Effects of hub-and-spoke free trade
agreements on trade: a panel data analysis. World Developemnt, 38(8), 1105-1113.
Ibrahim, M. A. (2012). Merchandise export demand function for Egypt: a
panel data analysis 1990-2008. Applied Econometrics and International Development
(AEID), 12(1), 107-116.
Jeon, Y. ( 1992). The determinants of Korean foreign direct investment in
manufacturing industries. Review of World Economics, 128(3), 527-542.
Khedhiri, S., &Bouazizi, T. (2007). An empirical analysis of the demand
elasticity for Tunisian exports. Applied Econometrics and International Development,
7(1 ), 134-148.
Kepaptsoglou, K. , Karlaftis, M. G. , &Tsamboulas, D. (2010). The gravity
model specification for modeling international trade flows and free trade agreement
effects: a 10-year review of empirical studies. The Open Economics Journal, 3, 1-13.
Korinek, K. , &Melatos, M. (2009). Trade impacts of selected regional trade
agreements in agriculture. OECD Trade Policy Working Papers, No. 87.
Kumar, N. (1998). Multinational enterprises, regional economic integration,
and export-platform production in the host countries: an empirical analysis for the US
and Japanese corporations. WeltwirtschafthchesArchiv, 134, 45-83.
Linda, G. , & Michael , K. (1998). Foreign direct investment, trade and real
exchange rate linkages in developing countries. managing capital flows and
exchanges rates: perspectives from the Pacific Basin. Cambridge University Press
1998, 73-100.
Linneman, H . ( 1966). An econometric study of world trade.flows. Amsterdam:
North-Holland.
Lipsey, R.E ., & Weiss, M.Y. (1984). Foreign production and exports of
individual firms. The Review of Economics and Statistics, 66(2). 304-308.

34
 Liu, X.,Shu. C.(2003).Determinants of export performance: Evidence form
Chinese industries. Economics of Planning 36, p.45-67.
Majeed, M. T., & Ahmad, E. (2006). Determinants of exports in developing countries.
The Pakistan Development Review, 45:4 Part II, p. 1265-1276.
Martinez-Zarzoso, I., & Nowak-Lehmann, F. (2001). Augmented gravity
model: an empirical application to Mercosur-European union trade flows. Journal of
Applied Economics, 6(2), 291-316.
McCallum, J. (1995). National borders matter: Canada-U.S. regional trade
patterns. American Economic Review 85, p. 615-623.
Miyao, R. (2003). The effects of Yen depreciation policy in Japan . Economic
Reviews, 54, p. 114-25.
Moffett, M.H. , Stonehill, A.I. , &Eiteman (2009). Fundamentals of
multinational finance. Boston: Pearson Prentice Hall. 3rd edition.
Najia, S., &Irfan, S. (2012). Exchange rate volatility and its effect on
Pakistan ' s export volume. Advances in Management and Applied Economics, 2(4),
109-124.
Oguledo, V. I., &Macphee, C.R. (1994). Gravity models: a reformulation and
an application to discriminatory trade arrangements. Apphed Economics, 26. 107-1 20.
Ong, H. B., Yoong, Y. J. , Lim, S. L. , & Tong, G. K. (2009). The effect ofreal
exchange rate on Thailand' s export. IUP Journal of Applied Economics, 8(5/6), 48-
54.
Orindi, M. N . (2011). Determinants of Kenya' s exports: a gravity model
approach. International Journal of Economic and Political Integration, 1( 1), 3-14.
Oscar, T-R. (2013). Panel data analysis fixed and random effects (using Stata
10.x). Retrieved June 10, 2013 , from:http://www.princeton.edu/~otorre s/Panel 101.pdf

Park, H. M . (2010). Practical guides to panel data analysis. Retrieved

December 10, 2013 , from :
http://www.iuj .ac.jp/faculty/kucc625/writing/panel_guidelines.pdf .
Parkin, M. (2010). Economics. Pearson Education Publication. 8111 edition.

35
 Pemasiri, J. G., & Sharma, K. (2010). The determinants of Australian
manufacturing export performance. Journal of International Finance and Economics,
10(2).
Porojan, A. (2001 ). Trade flows and spatial effects: the gravity model
revisited. Open Economic Review, 12, 265-280.
Rahman, M. M. (2010). The factors affecting Bangladesh's exports: evidence
from the gravity model analysis. Journal ofDeveloping Areas, 44( 10), 229-246.
Redding, S., &Venables, A. J. (2003). Geography and export performance:
external market access and internal supply capacity. NBER Working Paper 9637.
Senhadji, A. S., Montenegro, C. E. (1999). Time series analysis of export
demand equations: a cross-country analysis. IMF Staff Papers, 46(3), 259-273.
Sharma, K. (2000).Export growth in India: Has FDI played a role. Yale
University Economic Growth Center Discussion Paper, No. 816.
Sharma, K . (2003). Factors detern1ining India' s export performance. Journal
ofAsian Economics, 14, 435-446.
Taye, Y. T. (2009). Determinants of Ethiopia' s export performance: a gravity
model analysis. Trade and development discussion paper No. 01/2009.
Tilak, A. , & Tan, L. Y. (1998). Exchange rate appreciation and export
competitiveness: the case of Singapore. Department of Economics and Statistics,
National University of Singapore, l 0 Kent Ridge Crescent, Singapore.
Tien, D.N. (2009). Determinants to Vietnam ' s export flows and government
implications under the global crisis. Research on international Trade policy
Conference. Foreign Trade University, Hanoi , Vietnam.
Timbergen, J. (1962). Shaping the world economy. New York: Twentieth
Century Fund.
Trang, N. T. H. , Tam, N. T. T., & Nam, V. H. (2010). An inquiry into the
determinants of Vietnamese product export. Faculty of Economics and International
Business Foreign Trade University, Vietnam .
Tri , D. T. (2006). A gravity model for trade between Vietnam and twenty-
three European countries. Department of Economics and Society.
Trung, N. T. (2013). Thvc tn;mg phat tri~n xu~t nh~p kh~u cua Vi~t Nam va dv
bao d~n nam 2015 . Retrieved April 19, 2013, from: http://aidv.vn/san-pham-dich-

36
vu/dich-vu/thuong-mai/thuc-trang-phat-trien-xuat-nhap-khau-cua-viet-nam-va-du-
bao-den-nam-2015.html .
Tuan, H. A. (2009). DoiMoi and the remaking of Vietnam. Global Asia- A
Journal of the East Asia Foundation. Retrieved April 12, 2013, from:
http://www.globalasia.org/l.php?c=e2 l 5 .
Vietnam's General Statistics Office (2013a). Trading value and price index.
Retrieved May 15, 2013, from:
http://www.gso.gov.vn/default.aspx?tabid=393&idmid=3&lternID=l4552 .
Vietnam's General Statistics Office (2013b). Exports value by country and
geographic area. Retrieved May 15, 2013, from:
http://www.gso.gov.vn/default.aspx?tabid=393&idmid=3&ItemID=l4548.
Weldemariam, H. G. (2009). Modeling determinants of Ethiopia's export: a
gravity approach. Retrieved June 15, 2013, from:
http://papers.ssm.com/sol3/papers.cfm?abstract id=l 525443 .
World Bank. (2013a). Exports of goods and services (% of GDP). Retrieved
April 5, 2013, from: http://data.worldbank.org/indicator/NE.EXP.GNFS.ZS.
World Bank. (2013b). GDP per capita (current US$). Retrieved April 5, 2013,
from: http://data.worldbank.org/indicator/NY.GDP.PCAP.CD.
World Bank. (2013c). An update on Vietnam's recent economic development
July 2013: Key findings. Retrieved September 10, 2013, from:
http://www.worldbank.org/en/news/feature/2013/07/l2/taking-stock-july-2013-an-
update-on-vietnams-recent-econornic-devel oprnent-key-findings .
Xuan, N. T., & Xing, Y. (2006). Foreign direct investment and export: the
experiences of Vietnam. GSJR Working Paper, 6-11.

37
 APPENDIX

1. List of Vietnam's forty major exporting markets

ASEAN United States
Cambodia China
Indonesia Japan
Laos Others
Malaysia New Zealand
Philippines Australia
Singapore Russian Federation
Thailand India
Euro~ean Union Saudi Arabia
Bulgaria Korea, Rep.
Czech Rep. Hong Kong
Poland Ukraine
Austria Norway
Germany Switzerland
Belgium Canada
Hungary United States
Italy Mexico
Netherlands Algeria
Slovenia South Africa
Denmark
Spain
France
United Kingdom
Sweden

2. Gravity regression with the fixed effects model

F1xea eTTects lW1th1nJ regression Number of obs - 680
Group variable: countryc Number of groups- = 40
R-sq: within = 0.7942 Obs per group: min = 17
between = 0.1900 avg = 17.0
overall = 0.2239 max = 17
F(6,39) = 91.71
corr(u_i, Xb) = - 0.8861 Prob > F = 0.0000
(Std. Err. adjusted for 40 clusters in countryc)
Robust
logexp coef. Std. Err. t P> !ti [95% Conf. Interval]
logvngdp 1. 725462 .1921005 8.98 0.000 1.336902 2.114022
logvnfdi - .2608017 .0926989 - 2.81 0.008 -.4483029 - .0733005
ft as -. 0343612 .0998426 - 0.34 0.733 -.2363119 .1675895
rexch 9.23e- 06 3.94e-06 2 . 34 0.024 1. 25e-06 . 0000172
logicgdp 2.564262 1. 214334 2.11 0.041 .108039 5.020485
logicgdppc - 2 .405171 1.198898 - 2.01 0.052 - 4.830171 .0198286
logdis (omitted)
_ cons -27.57177 8. 540813 - 3.23 0.003 - 44 . 8472 -10.29635
sigm<L..u 1.4855954
sigm<L..e .22317557
rho .97793012 (fraction of variance due to u_i)

38
•:
3. Gravity regression with the random effects model
Ranoom- e~~ects GLS regression Number o~ obs = 680
Group variable: countryc Number of groups = 40
R- sq : within = 0.7841 obs per group: min = 17
between = 0.6887 avg = 17.0
overall = 0.7152 max = 17
Random effects u_i - Gaussian Wald chi2(7) = 616.69
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
(Std. Err . adjusted for 40 clusters in countryc)
Robust
logexp coef. Std. Err. z P>IZI [95% Conf. Interval]
logvngdp 1.715533 .1742754 9.84 0.000 1. 373959 2.057106
logvnfdi - .3178508 .0939428 - 3.38 0.001 - .5019752 - .1337263
ft as .0338501 .1009871 0 . 34 0.737 - .1640811 .2317812
rexch .0000107 3.85e-06 2.77 0.006 3.13e- 06 . 0000182
logicgdp .6337591 .1058124 5.99 0.000 .4263705 .8411477
logicgdppc - .1518352 .1832778 - 0.83 0.407 -. 5110532 .2073827
logdis -1.070503 .2205377 - 4 . 85 0 . 000 - 1.502749 - .6382571
_cons - 9.78504 1. 547605 - 6.32 0 . 000 - 12 . 81829 - 6. 751789
sigma...u .30903399
sigma...e .22317557
rho .65723239 (fraction of variance due to u_i)

THE ASSUMPTION UNIVERSITY LJBRA8'

39