Location specific determinants of japanese foreign direct investment in selected asian countries

DECLARATION This is to certify that the thesis entitled “Location-specific determinants of Japanese foreign direct investment in selected Asian countries”, is submitted by me in fulfillm

University of Economics International Institution of Social Studies

IN SELECTED ASIAN COUNTRIES

BY HUYNH THAO THUY VI

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

Master of Arts in Development Economics

Under the supervision of Dr Le Cong Tru

HO CHI MINH CITY, 2014

DECLARATION

This is to certify that the thesis entitled “Location-specific determinants of Japanese foreign

direct investment in selected Asian countries”, is submitted by me in fulfillment of the

requirement for the degree of Master of Art in Development Economics to Vietnam –

Netherlands Programme This thesis comprises only my original work and due supervision and

acknowledgement have been made in the text to all other material used

Huynh Thao Thuy Vi

ACKNOWLEDGEMENT

I would not be possible to finish this thesis without the support of people surrounding me

Firstly, I am really grateful to my supervisor, Dr Le Cong Tru, for his guidance, comments and

supervisions

Secondly, I would like to express my gratitude to Dr Truong Dang Thuy for his econometric

guidance

Thirdly, I want to acknowledge all lecturers of Vietnam-Netherlands Programme for the wide

knowledge they provided me during the time I studied here

Next, I would like to express my gratitude to my parents for all of their sacrifice, encouragement

and support for me

Last but not least, I would like to thank my friends and people, who supported for my thesis but

were not above-mentioned

ABBREVIATIONS

FDI Foreign Direct Investment

JBIC Japan Bank for International Cooperation

JFDI Japanese Foreign Direct Investment

JETRO Japan External Trade Organization

MNE Multinational Enterprises

ABSTRACT

This thesis contributes to examine the location-specific determinants influencing on Japanese foreign direct investment This research uses fixed effect method and panel data of ten selected Asian countries in the period from 1995 to 2012 The determinants are classified into three groups: policy factor, business facilitation and economic factor In this study, except for the variables which belong to policy factor and business facilitation, the others of economic factor are categorized in accordance with three main motives for Japanese enterprises investing abroad Those motives are market-seeking, resource-seeking and efficiency-seeking The study finds that market size, natural resource, inflation rate, exchange rate volatility, political risk and infrastructure development are the significant factors

Moreover, among ten selected Asian countries, there are the differences between intercept coefficients of Vietnam and other countries who are Thailand, Indonesia, Vietnam–Philippines and China In other words, except for the determinants in the regression model, there would be other factors making those countries to be more dominant than Vietnam Finally, recognizing the need of Japanese foreign direct investment, the policy makers in host countries should apply the relevant policies to improve the business environment for becoming promising destinations and attract more Japanese foreign direct investment

Key words: Japanese foreign direct investment, panel data, location-specific, Asian countries, fixed effect

TABLE OF CONTENT

DECLARATION 1

ACKNOWLEDGEMENT 2

ABBREVIATIONS 3

ABSTRACT 4

TABLE OF CONTENT 5

LIST OF FIGURES 7

LIST OF TABLES 8

CHAPTER 1 INTRODUCTION 9

1.1 PROBLEM STATEMENT 9

1.2 RESEARCH OBJECTIVES 10

1.3 RESEARCH QUESTION 10

1.4 THESIS STRUCTURE 11

CHAPTER 2 LITERATURE REVIEW 12

2.1 THEORETICAL LITERATURE 12

2.1.1 Early concepts & studies of determinants of FDI 12

2.1.2 Neoclassical Trade Theory (Heckscher-Ohlin model & MacDougall-Kemp model) 13

2.1.3 Hymer – Kindleberger Paradigm 13

2.1.4 Internalization theory 14

2.1.5 The OLI paradigm - Eclectic theory 15

2.2 EMPIRICAL STUDIES 17

2.3 CONCEPTUAL FRAMEWORK 22

CHAPTER 3 RESEARCH METHODOLOGY 25

3.1 VARIABLE DEFINITION AND TESTING HYPOTHESES 25

3.1.1 Market size 25

3.1.2 Natural resource 25

3.1.3 Inflation rate 25

3.1.4 Exchange rate volatility 26

3.1.5 Trade openness 26

3.1.6 Political risk 27

3.1.7 Infrastructure development 27

3.1.8 Labor cost 28

3.2 DATA AND MODEL SPECIFICATION 28

3.3 RESEARCH METHODOLOGY 31

3.3.1 Descriptive Analysis 31

3.3.2 Regression Analysis 32

CHAPTER 4 DATA ANALYSIS 37

4.1 THE OVERVIEW OF JAPANESE FDI IN ASIAN COUNTRIES 37

4.1.1 Before the crisis in 1997 37

4.1.2 After the crisis in 1997 43

4.1.3 Several remarkable characteristics of recent Japanese FDI in Asia 45

4.2 EMPIRICAL RESULTS 47

4.2.1 Descriptive analysis and some general tests 47

4.2.2 Econometric results 52

CHAPTER 5 CONCLUSION AND POLICY IMPLICATION 57

5.1 CONCLUSION 57

5.2 POLICY IMPLICATION 58

5.3 LIMITATION AND SUGGESTION OF FURTHER RESEARCH 59

APPENDIX A: RESULT OF HAUSMAN TEST 60

APPENDIX B: RESULTS OF HETEROSKEDASTICITY & 61

SERIAL CORRELATION TEST 61

Table B-1: Heteroskedasticity test 61

Table B-2: Serial Correlation test 61

APPENDIX C: REGRESSION RESULTS 62

REFERENCES 63

LIST OF FIGURES

Figure 2.1: Conceptual Framework 23

Figure 4.1: Japanese FDI by region (billion Yen) 38

Figure 4.2: Japanese FDI in 10 selected Asian countries (billion Yen) 40

Figure 4.3: Japanese FDI in NIEs (billion Yen) 41

Figure 4.4: Japanese FDI in ASEAN4 (billion Yen) 42

Figure 4.5: The correlation between Political Risk (pol) and Trade Openness (TO1) 49

Figure 4.6: The correlation between trade and GDP 50

Figure 4.7: The response of Japanese enterprises considering Vietnam, Thailand, Indonesia, Philippines and China as the promising countries 56

LIST OF TABLES

Table 2.1: The summaries of some typical theories of FDI 16

Table 3.1: Summary of testing hypotheses 28

Table 3.2: Summary of expected signs of variables 30

Table 4.1: Regional distribution of FDI by Japanese firms 39

Table 4.2: Promising countries for overseas business operation over the medium-term in term of Japanese enterprises 45

Table 4.3: Summary of variables in the study 48

Table 4.4: Correlation coefficients of variables in the study 48

Table 4.5: The VIF and TOL factors before excluding trade openness (TO1) 49

Table 4.6: The VIF and TOL factors after excluding trade openness (TO1) 51

Table 4.7: Summary of Estimation Results 52

CHAPTER 1 INTRODUCTION

For the time being, because of the immobility and long run profitability, foreign direct investment (FDI) is one of the important factors contributing to the economic growth of countries in the world (Nakamura & Oyama, 1998) Through FDI, the financial resources are transferred to the host countries to set up and expand the production conditions in those countries Furthermore, the technological achievements and managerial knowledge are also transferred from the investing countries to the host countries Those factors would contribute to the economic development of the recipients Moreover, the host countries may also take advantage of the networks through the sales and distribution networks of foreign investors

As to the destination of FDI, with the available and potential advantages, Asia has been still the leading region in attracting FDI with 3,740 projects tracked in 2012, which increases its global market share to 31.72% Many countries in Asia have achieved the dominant economic growth through FDI into many specific industries, such as, business and financial services, ICT, chemicals, plastics and rubber, etc As to the source country or home country of FDI, Japan is still the dominant one all over the world, especially in Asia In spite of the decreased number of outward FDI projects from Japan, the ratio of Japanese FDI in Asia still went up from 34.57% in

2011 to 37.37% in 2012 (Fingar, 2013) According to the survey conducted by Japan External Trade Organization (JETRO) on 3,397 JETRO member firms and 6,403 enterprises using JETRO services in 2013, 64.9% of firms intend to expand overseas operation by conducting new investments going with existing operation bases whereas 91.7% set up locations in Asia Pacific

It would be indicated that because of natural disasters and difficult domestic business environment in Japan, for example, labor costs, tax burden, domestic regulations, etc, Japan firms are concentrating on widening their overseas investments (JETRO, 2014) Moreover, as to the conception of Japanese firms, the countries in Asia have been possessing advantages, which would promote them to be promising destinations for JFDI Those advantages can be listed as current size of local market, inexpensive source of labor, social and political stability, etc

Despite many motives for Japanese firms to invest in Asia, there are many issues in this region, which raised concerns for Japanese enterprises As indicated in the report of Japan Bank

for International Cooperation (JBIC) in 2013, the rise in salary, difficult searching for the raw materials, underdeveloped infrastructure, unclear legal system, etc have been the typical issues existing in Asian countries and possibly preventing them from receiving more investment from Japan Combining the pros and cons in Asian region, we can examine the determinants affecting JFDI inflows in Asia In other words, the study of the important factors determining JFDI in Asia is necessary for boosting JFDI into this region and should be based on the empirical studies and the real situations as well Since then, the relevant policies can be suggested for the countries in Asia to become more attractive destinations to Japanese investors

Up to now, there are some studies about factors affecting FDI inflow in Asian countries

It can be stated that the empirical papers presented the different and various factors to demonstrate for the researches and got general findings A few authors have conducted their studies by basing on data of FDI inflow into specific sectors of specific countries to find out what are called sector-specific determinants of FDI

In this study, the research is concentrated on the total amount of FDI into a specific location or host country and come to contribute to on-going researches by examining the location-specific determinants of Japanese FDI in ten Asian countries As to this general objective, this study aims at the following specific objectives:

- To provide the descriptive analysis of the changes in the period of before and after

1997 crisis and some recently remarkable characteristics of JFDI in ten selected Asian countries

- To measure the impacts of location-specific determinants on JFDI in ten selected Asian countries

- To suggest some policy implications for attracting JFDI

To conduct research objectives, this study aims to answer the following questions:

- What are the characteristics of JFDI in the period of before and after 1997 crisis and the changes in recent JFDI in ten selected Asian countries?

- What are the significant factors affecting JFDI inflow in selected Asian countries? Is there any difference between intercept coefficients of ten selected Asian countries?

- What are the relevant policies that can be suggested basing on the analysis to attract

JFDI inflow?

This thesis consists of five chapters Right after Chapter 1 which aims to present the introduction of JFDI inflows into Asian countries, Chapter 2 will synthesize the remarkable theories about determinants of FDI There are five typical theories presented in this chapter from the one with the basic concept to the wider one They are early studies of FDI, Neoclassical trade theory, Hymer – Kindleberger Paradigm, Internalization theory and OLI paradigm - Eclectic theory Moreover, some typically empirical studies will also be reviewed in Chapter 2 By considering the combination of theories and empirical studies, the analytical framework of this thesis will be drawn at the end of Chapter 2

Following Chapter 2, Chapter 3 concentrates on variables definitions and eight testing hypothesis Furthermore, going with the introduction of empirical model used in this thesis, the source for data and the expected signs of eight variables will be summarized in one table The final part in Chapter 3 is the research methodology which describes the methods and typical tests

in panel data regression

In Chapter 4, an overview of JFDI in Asian countries will be presented to describe the changes in JFDI in the period of before and after the-1997-crisis Moreover, the recent trends and characteristics of JFDI in ten selected Asian countries will be also analyzed The next in this chapter is the findings obtained from the estimation results, which will provide the answers for each hypothesis stated in Chapter 2 Since then, the question of significant factors determining JFDI in selected Asian countries will be also discussed

Basing on the findings in Chapter 4, some conclusion remarks and policies recommendations for attracting JFDI into Asian countries will be presented in Chapter 5 which is the last chapter in this thesis Moreover, Chapter 5 will also point out some limitations of thesis and suggest some further research for the study of JFDI

CHAPTER 2 LITERATURE REVIEW

This chapter includes three main parts The first section provides some typical FDI theories whose combinations will be the theoretical basis for this thesis The next part contains the summaries of some empirical studies Finally, basing on the theories and empirical studies, the conceptual framework for this thesis will be presented in the last section

The rise in FDI has led to the extensive research and the development of many theories to explain the determinants of FDI flow To study FDI, it should be the combination of many theories, not just basing on any single one (Faeth, 2009) Therefore, in this study, some theories will be presented from the early to the latest one At the end of this section, the Table 2.1 will provide the summary of these theories

2.1.1 Early concepts & studies of determinants of FDI

The early researches of determinants of FDI were mostly based on the questionnaires The companies in the survey or research were asked to realize and indicate the reasons or factors which encouraged them investing abroad Some major researchers contributing to the initially general building of FDI’s determinants would be mentioned such as Robinson (1961), Behrman (1963), Basi (1966), etc (in (Faeth, 2009) In these early studies, a diversification of factors, indicated as the encouraging ones in multinational enterprises’ investment decisions, included cost factors, trade openness factors, marketing factors and investment climate The degree of significance of those factors were various in every study’s analysis result In some researches, marketing factors whose proxies were market size and market growth were the major determinants of FDI However, in other studies, the availability of low labor and natural resource were considered as the important factors On the other hand, according to the research of Basi (1966), political stability was the most significant determinant It would be said that through the early studies with the researches in specific economies, the initial concepts of FDI’s determinants have been taken into account

2.1.2 Neoclassical Trade Theory (Heckscher-Ohlin model & MacDougall-Kemp

model)

As to the Neoclassical Trade Theory, Heckscher – Ohlin (HO hereafter) model was considered as the first theory attempting to explain the essence and factors of FDI Two Swedish economists, Eli Heckscher and Bertil Ohlin, put the foundation of FDI in HO model through some basic assumptions: no trade barriers, no transportation cost, perfect competition & full employment, no specialization, the same constant return to scale production function, the existence of domestic factor mobility & international factor immobility and the identical technology between countries HO model was based on 2x2x2 model, standing for two countries (home and host country), two factors (capital and labor) and two commodities In HO model, the only difference in relative factor endowment would lead to the comparative advantage and international factor price differentials The key concept in HO model is that the home country would export to the host or foreign country the good, which uses the factor that the home country

is relatively abundant MacDougall-Kemp (MK hereafter) model was one of the earliest theories

of FDI The assumptions in this model were also the full employment, perfect competition and constant return to scale Like HO model, the capital would move from the home country to the host country that has the higher capital rate of returns However, as to MK model, the host country can apply the tax on the capital flow and manage the capital return It could be said that the researches in MK model had took into account the barriers toward the trade or the capital inflow

2.1.3 Hymer – Kindleberger Paradigm

Besides the Neoclassical Trade Theory, after the World War II, the theory of portfolio investment, which is one of the oldest theories of FDI, was used to explain for FDI flows The basis of this theory is interest rate Under this concept, the investors will invest where brings more profits However, this theory was developed by the assumptions of no risks, uncertainties,

or barriers that cannot exist in the reality Therefore, Stephen Hymer devoted his 1960 dissertation to develop another clearer theory of FDI It is called Hymer – Kindleberger paradigm As to Hymer’s theory, besides the unrealistic assumptions, in the interest-rate theory

or portfolio investment theory, there is still a theoretical limitation which is called shortage of explanation of control (Hymer, 1976) As to the portfolio investment theory, the investor would invest his money in a foreign enterprise whose interest rate is higher than domestic one’s

However, he cannot control the enterprise he invested in Therefore, Hymer built two types of direct investment to explain for the reasons why the investors would like to seek the control Since then, the theory of direct investment was formed

In the direct investment of Type 1, Hymer indicated that the investor seeks the control because they would like to make sure that the capital was used prudently Because of the different nationalities, the conflict would happen between the investors about the ratio of reserves that should be kept in particular currency Meanwhile, Hymer also emphasized the difference between international transaction and intra-national one, which would lead to the distrust and consideration among investors Generally, although direct investment of Type 1 is nearly similar to the portfolio investment theory, in which the core concept is interest rate, it would supplement in portfolio investment theory the concept of the necessity of control in the case that there is unbelief between investors and the high fear of expropriation and the exchange rate risk As a result, direct investment of Type 1 can replace the portfolio investment theory

As to the direct investment of Type 2, the other reason for operating the foreign enterprise is the possibility of removal of competition between the firms (Hymer, 1976) It is a very normal phenomenon that in the same market, the competition between foreign enterprises would happen In the case of imperfect market, the combination by gathering the foreign firms and giving the right of control for one firm would be one of the profitable collusion In addition, the ability of enterprise to build the international operation is rather different It would depend on the each firm’s specific advantage in particular industry Furthermore, the firm would have two choices: the first one is to rent or sell its skills or technologies; the other is setting up the international operation and undertaking by itself

2.1.4 Internalization theory

The internalization theory, which was developed by Buckley and Casson in 1976, was considered as a general paradigm explaining for FDI Due to the market imperfection, the internalization would happen Under this theory, the firm internalizes its globally foreign businesses to benefit from the internal network and prevent its operation from the disadvantages

of resource allocation (Buckley & Casson, 2009)

In reality, most enterprises have to buy the inputs from independent suppliers, who would

be international ones In this situation, one question was raised that whether the enterprises should produce inputs by themselves or not This kind of question belongs to what is called

“make or buy decision” in business management or “backward integration” in economics studies This issue also leads to one kind of direct investment, which is called “resource-seeking investment” It means that a global company would like to set up the international operations in a foreign country; and via its subsidiaries in that country, it would use raw materials that could not

be found in any elsewhere to produce intermediate goods In addition, along with “backward integration”, “forward integration” issue was also arisen to mention to the question that whether the enterprise should build its foreign subsidiaries to control the distribution of its goods in foreign markets or not By establishing these subsidiaries, instead of using independent distributors, the enterprise itself can supervise the oversea distribution network

Moreover, Buckley and Casson emphasized in their studies that the multinational enterprises (MNEs) specializing in research and development industries would have higher internalization than the others (Faeth, 2009) On the other hand, MNEs only invested in some countries with specific characteristics going with MNEs’ investment plans As to the study of Buckley and Casson (2009) through literature of development economics, the importance of encouraging-foreign direct investment factors would be analyzed and classified It was stated in the study of Buckley and Casson that as to the MNEs planning to broaden their consuming market into any country, the factors of local market size and local standard of living were the important ones In addition, with the plan of widen markets not only in host country, but also in host country’s neighboring ones, the infrastructure development was the outweighed factor Or with the engagement of constructing focusing-export plants in host country, the MNEs would pay their attention to the factor of labor cost

2.1.5 The OLI paradigm - Eclectic theory

By combining and developing from previous theories, Dunning raised the eclectic theory that is also called OLI paradigm The OLI paradigm is the combination of three factors: O-L-I

which in turn stands for Ownership advantages, Location advantages and Internalization

advantages

When entering the abroad location for production, the multinational firms have to face the additional costs, which are caused by the diversity of legal, cultural, language system, the lack of knowledge of domestic market; etc, would reduce their benefits As to the first factor of OLI paradigm, FDI happens when the multinational enterprises have Ownership advantages or firm specific advantages that can offset the additional costs These advantages can be tangible

(superior technology, products, economies of scale and scope, etc) or intangible (brand name, trademark, etc) (Hosseini, 2005)

As to the second factor in OLI paradigm, a particular country’s characteristics, which are considered as the Location advantages or country specific advantages by foreign firms, would be the drivers of FDI inflows The multinational enterprises tend to combine its ownership advantages and location advantages endowed in host countries to make their investment more beneficial Location advantages can be divided into three groups: economic advantages (including qualitative and quantitative factors related to economics, for example, transport and communication costs, market size, market growth, etc); political advantages (consisting all government policies related to FDI inflows, international trade and production, etc) and social & cultural advantages (including the attitude toward foreign enterprises, the diversity in culture, etc)

The final factor in OLI paradigm is Internalization advantages that would be exploited when the multinational enterprises realize the benefit of using Ownership advantages and investing abroad instead of export or other contractual agreements, such as, licensing, joint ventures, etc (Hosseini, 2005) This is the third leg of OLI paradigm in making clear the scale and geography of foreign activities of multinational enterprises (Dunning, 2001)

Table 2.1: The summaries of some typical theories of FDI

Early studies Cost factors, trade openness factors,

marketing factors and investment climate

Robinson (1961), Behrman (1963), Basi (1966), etc Neoclassical Trade Theory:

- Heckscher – Ohlin Model

- MacDougall – Kemp Model

Higher return on investment, lower labor cost

Heckscher & Ohlin (1933)

(1976) OLI paradigm - Eclectic

theory

Ownership advantages, Location advantages, Internalization advantages

Dunning (1977, 1979)

2.2 EMPIRICAL STUDIES

Up to now, there are many studies researching about determinants of FDI The study of FDI spreads from the FDI inflows to FDI outflows, from locational determinants to sectorial ones of FDI Basing on three conceptions of capital imperfection, special ownership advantages and institutional factors, Buckley et al (2007) investigated the determinants of foreign direct investment by Chinese multinational enterprises The study was based on official data from 1984 – 2001 The authors collected data of forty-nine host countries receiving Chinese FDI, in which, there are twenty-two countries belonging to Organization for Economic Cooperation and Development (OECD) Two models were used in this study: pooled ordinary least squares (POLS) and random effects (RE) Moreover, the authors conducted Lagrangian multiplier (LM) test to conclude that RE is the better model Basing on the general and specific theories of FDI, Buckley et al used fourteen independent variables to be the determinants of Chinese FDI outflows These are market size, market size per capita, market growth, natural resource endowment, host country’s endowment of ownership advantages, political risk, cultural proximity, policy liberalization, exchange rate, inflation rate, geographic distance from China, export & import and openness to FDI Basing on the empirical theories, with the expectation of nonlinear relationship, the authors transformed data into the natural logarithms The results of this study emphasized the important role of host market characteristics, cultural and political factors on FDI inflows In host market factors, market size was the only one having significantly positive influence on Chinese FDI outflow On the other hand, the exchange rate, geographic distance and openness to FDI are all insignificant

In the study of Kinoshita, Campos, and Pankki (2004), the host country’s characteristics, which would be the drivers of FDI, were analyzed by using 1990 – 1998 panel data of 25 transition countries, which include Central and Eastern European and Baltic (CEEB) countries and the Commonwealth of Independent States (CIS) consists of all former Soviet Union countries These two authors used fixed effects and random effects model to find out the importance of host country’s factors in FDI inflows However, the Hausman test rejected random effects model They divided the determinants into four groups Group 1 – Classical Source of Comparative Advantage – referred to the host country’s advantages considered as the motivation for FDI inflows The variables in Group 1 were market size, low labor cost, labor quality and infrastructure The second Group included Macroeconomic Policy and Reform variables in

which inflation rate was the typical one to emphasize the stability of host country’s economy As

to the authors, the lower average inflation rate, the more profit can be brought by investment projects Next, the third Group emphasized the important role of host country institutions To make the investment decisions, the MNEs have to consider not only the economic cost, but also the non-economic one The authors used the “Rule of Law” variable whose data was collected from International Country Risk Guide to measure the corruption of host country In the last Group, the variable of agglomeration was used to demonstrate the feedback effect of past FDI on the current and future FDI Through the estimation results, the authors found that market size, labor cost and natural resource were the important FDI’s determinants In addition, good institution or lower political risk, higher trade openness were also the drivers of FDI inflow Nevertheless, the variable of education and infrastructure were likely to be insignificant

The same results were also demonstrated in the study of Farrell, Gaston, and Sturm (2004) These authors used panel data from 1984 to 1995 of 16 countries and applied pooled and fixed effects model to indicate the important determinants of Japanese FDI They found that market size and labor cost of host country were extremely significant Moreover, the macroeconomic conditions also highly affected JFDI inflow As to the variable of exchange rate, basing on the statistically insignificant results, another finding of the author was that the strength

of Japanese Yen was not likely to be the main determinant of JFDI According to the authors, this finding was also similar to other studies

Using the same type of data from 1979 to 1997 for Japanese FDI and data from 1982 to

1997 for the United States (US) FDI, the research of Nakamura and Oyama (1998) focused on indicating the macroeconomic determinants of FDI from Japan and US to eight East Asian countries They were Taiwan, Korea, Indonesia, Philippines, China, Malaysia, Singapore and Thailand The reason that these two Japanese authors chose to investigate Japan and US as the home countries is that FDI from Japan and US occupied about 50 percent of the total FDI flow into the above-referred East Asian countries According to these authors, although there were many factors influencing on FDI, in their study, they just concentrated on macroeconomic ones, especially the real exchange rate Before conducting the regression analysis, eight host countries were also classified into many different groups basing on their FDI elasticity to macroeconomic variables Remarkably, this classification was similar to the countries’ economic characteristics and development process The study applied fixed effects and random effects method for the

analysis Like the others, Hausman test was also used to decide which method was better As to the regression result, Japanese FDI into eight countries was significantly influenced by the change of exchange rate whereas US FDI was not Moreover, as to both Japanese FDI and US FDI, the coefficient of host country’ GDP was found to be significant in most host countries On the other hand, this study also indicated the link between FDI and trade In other words, FDI strongly affect the export from host countries to Japan and the import of host countries from Japan

Another typical study investigating the factors of Japanese FDI belonged to Urata and Kawai (2000) By using data of 117 countries that were divided into three groups: developed, developing and Asian countries, these two authors found out the importance of host country’s factors in JFDI inflows The independent variables include exchange rate, wage rate, market size, macroeconomic stability, labor quality, infrastructure, agglomeration and governance Accompanied with the analysis of relationship between those factors and JFDI, the authors also used SME dummy to indicate the difference between small, medium-sized enterprises (SMEs) and the large ones in the way of making investment decisions The results of this analysis showed that low wage labor, good infrastructure, factors related to local market have the significant meanings on JFDI inflows When investing in developing countries, Japanese multinational enterprises aim the export production that is the reason why they consider much more about the production conditions On the other hand, investing in developed countries, Japanese firms would be interested in the local sales As a result, market-related factors would be always the decisive ones Moreover, industrial agglomeration was found to be very significant in attracting JFDI Japanese investors always consider the inter-firm relationship between Japanese enterprises As to their conception, this relationship would bring them benefits in procurement and sales The other finding referred in the study is the SMEs are more sensitive to the change of locational conditions than the large ones Therefore, with the investment tendency from Japanese SMEs, the countries, who want to attract JFDI inflows, need to build, maintain and enhance local business environment

Similarly, with the dataset from 1975 to 2007, Vijayakumar, Sridharan, and Rao (2010) contributed one more study to the research of FDI’s determinants In their study, a new term was generated – BRICS standing for Brazil, Russia, India, China and South Africa This term represented for the world’s four continents whose economies were considered significant The

FDI inflow into those regions was extremely complex In this study, the authors used three methods for analysis: OLS pooled regression, fixed effects and random effects The fixed effects method was also rejected by Hausman test The variables were classified into seven groups They were market size, labor cost, trade openness, currency value, infrastructure facilities, gross capital formation and economic stability & growth prospect The analysis results showed that most variables had the important influence on FDI Nevertheless, the economic stability & growth prospect measured by inflation rate and trade openness measured by the ratio of total export plus import to GDP were found to be insignificant determinants of FDI in BRICS countries

Basing on the OLI framework of Dunning, Wadhwa and Reddy (2011) conducted the study about the impact of market-seeking, resource-seeking and efficiency-seeking factors of host countries on FDI inflows into those countries The study used panel data from 1991 to 2008

of ten developing countries, which are Bangladesh, China, India, Indonesia, Iran, Malaysia, Pakistan, Thailand, Turkey and Vietnam The author classified the determinants into three kinds

of factors As to market-seeking factor, GDP and population growth were used as the proxies of market size In efficiency-seeking factor, the author used inflation rate and exchange rate to be the variables in this category Infrastructure indexes including internet users, mobile subscribers and roads paved were used to demonstrate for the resource-seeking factor In this study, the author used fixed effect model with some necessary tests, for instance, Augmented Dickey-Fuller test checking the stationary of data and multicollinearity test The regression results showed that all the factors have the significant impacts on FDI inflows into ten Asian countries referred above

Basing on the data sets from 1992 to 2009 of provinces of China, Xin-Zhong (2005) also conducted panel model to investigate the location determinants affecting to FDI inflow into provinces in China The author classified these determinants into three categories They were investment environment improving, macro-economic and investment cost factors The group of investment environment improving factors included many determinants in which the typical ones are openness level of economics, policy index and infrastructure level The factors of macro-economics included market size, growth rate of economy, economic developing level and human capital The final group related to investment cost included labor cost and the exchange rate The empirical results indicated that all variables were statistically significant However, as to the

wrong signs of the variables of economic developing level and human capital, this author considered the existence of collinearity among variables As a result, this author chose the method of eliminating these two variables and recombining the independent variables to create more models and remove the collinearity The derived results showed the good performance of these two variables and the highly statistic significance of other variables Basing on the results, some policy implications were also suggested to attract more FDI inflows According to this author, the policy should concentrate on promoting GDP, liberal trade and re-educational projects Moreover, besides applying the basic education, it is really important for the government to pay attention to research and development (R&D) policy to achieve the productive labor force

Another research about FDI was conducted by Delaunay and Torrisi (2012) These authors studied about determinants of FDI in Vietnam by using time-series data which was from

1991 to 2008 and collected from the reliable sources, for instance, IMF, UNCTAD, Vietnamese GSO, etc According to these authors, the models of FDI determinants concentrated on two groups They are economic and non-economic factors In this study, the authors classified political stability, institutional efficiency and corruption indicators into the group of non-economic factors whereas market size, market growth, trade openness, etc belonged to economic factors On the other hand, because the research object was Vietnam whose figures or measures

of these factors were not available, this study just examined the impact of economic factors Moreover, the authors used ASIAN dummy variable to measure the impact of Vietnam’s membership in Asian trading zone on FDI into Vietnam The results derived were similar with other studies GDP, labor cost and exchange rate are predictably significant to FDI in Vietnam whereas GDP growth rate was not significant Furthermore, according to the results, the impact

of Vietnam’s membership in Asian trading zone on FDI into Vietnam was not clear The authors explained that although the intra-trade increased when Vietnam join Asian, this intra-trade has been rather low That is the reason why the re-export from Vietnam to ASIAN market is limited, which would reduce the attractiveness for FDI inflows It could be concluded fairly that ASIAN has not succeeded in being a trade integration mechanism yet

Different from previous studies, the joint research of Vuong and Yokoyama (2011) brought the variety and interest in researching FDI’s determinants, especially JFDI’s In the study, they used Importance Performance Analysis (IPA) method to investigate the factors that

stimulating JFDI inflow Since then, the authors evaluated and compared the attractiveness of Vietnam with China and Thailand in the process of being the dominant destination of Japanese enterprises The study was conducted by IPA method, which has been using not only in economic planning for strategic management problem, but also in forming framework for demonstrating changes Conducting qualitative research methods as referred by Dunning and Lundan (2008), these two authors carried out the survey and research based on the participation

of 1500 Japanese companies: 900 companies located in 15 districts in Japan and 600 companies set the operation in Vietnam Through the research, the attributes whose means were higher than four (>4), were considered as the determining factors of JFDI in Asia They were political stability, availability of skilled labor, infrastructure conditions, labor cost, access to raw material, etc Furthermore, the authors also indicated the important difference in between two group countries in the way that Japanese companies evaluated Vietnamese factors motivating JFDI As

to the enterprises investing in Vietnam, they paid attention and appreciated the political stability and the strength of Japanese Yen toward Vietnam Dong As to the companies not having projects

in Vietnam, they were found to be optimistic about Vietnamese investment environment and the quality of labor On the other hand, in making investment decisions, they also considered the easy access to raw material, infrastructure development and the corruption condition According

to the findings in this study, Vietnam was considered to be more advantageous than China and Thailand in production cost and labor characteristics However, to attract more JFDI, Vietnam should pay attention to the factors related to macroeconomic conditions and investment environment

Based on the literature review presented in the previous sections, the location-specific determinants affecting to FDI inflow are classified and described in the following figure:

Figure 2.1: Conceptual Framework

Following the World Investment Report (UNCTAD, 1998), the location-specific or host country determinants of FDI are classified into three groups They are policy factor, business facilitation and economic factor The policy factor refers to the index measuring the stability of politics and economics, the rules related to the entry and operation of multinational enterprises and the trade policy related to the trade openness, which is the necessary factor for the host countries to receive FDI inflows from foreign countries While the policy factor aims at creating the framework for the operation of foreign investors, the factors belonging to business facilitation are considered to facilitate their businesses in host countries The measures in business facilitation group are mostly new, in which the reduction of “hassle cost” is really

Location-specific (Host country)

- Economic, political and

1 Market – seeking - Market size

2 Resource/Asset – seeking - Raw materials, natural resource

- Physical infrastructure (power,

energy, telecommunication)

3 Efficiency - seeking - Lower labor cost

- Exchange rate

- Inflation rate

important Demonstrating the reduction of “hassle cost” includes measuring the improvement of government effectiveness or the reduction of corruption In this study, in term of policy factor and business facilitation, we use the variables of trade openness, the average figure of Rule of Law, Regulatory Quality and Government Effectiveness to be the proxies

The third group of host country determinant is the economic factor The economic factor includes many determinants belonging to the host country’s economic indexes that can have negative or positive effects on FDI inflow As indicated in World Investment Report (UNCTAD, 1998), the potential destinations for FDI belong to the host countries which have the advantages sought by the foreign countries However, to consider and decide which country is suitable for investment, the foreign firms from home countries also base on their strategies and motives According to the study of Dunning and Lundan (2008), there are three primary motives for the multinational enterprises to decide investing abroad They are market-seeking, resource/asset-seeking and efficiency-seeking FDI According to the above conceptual framework, the economic determinants are divided and classified relatively into each type of FDI motive As indicated in the study of Dunning and Lundan (2008), the prerequisite reason for market-seeking FDI is that a multinational enterprise finds it necessary to set the business operation in the important markets in which its competitors are serving When engaging market-seeking FDI, the foreign firms consider their affiliates to be independent production units rather than a part of network of cross-border activities The output will be mostly consumed in host country, so the market size of host country is top leading determinant that the Japanese firms always appreciate (JBIC, 2013) Next, the resource-seeking FDI, as it name, would happen when the multinational enterprises aim to acquire the source of physical infrastructure and raw materials (Buckley et al., 2007) Furthermore, because Japan is the country with limited natural resource, natural resource endowment is always considered to be significant determinant (Urata, 1993) The purpose of the last motive, efficiency-seeking FDI, is the lower cost reduction and the more efficient business operation As a result, the lower labor cost is classified in this type of FDI motive Moreover, in this group, inflation rate and exchange rate are the determinants highly evaluated While the inflation rate implies the macroeconomic stability, the exchange rate is the factor that the foreign investors would take advantage to reduce the cost of production (Wadhwa & Reddy, 2011)

CHAPTER 3 RESEARCH METHODOLOGY

This chapter includes three parts Basing on the conceptual framework discussed in the previous chapter, the first part introduces the independent variables used in this thesis Going with the definition of variables, each specific hypothesis is also derived The second section is data and model specification In this section, the way of calculating and the source of dependent variables are illustrated The final part is research methodology which consists of descriptive analysis and regression analysis

3.1.1 Market size

According to UNCTAD (1998), market size is one of the important traditional determinants influencing on FDI flows The developed or home countries look at the size of host country’s economy to decide whether they should conduct FDI or not The larger scale of economy, the more markets and more benefit the foreign enterprises can get Therefore, market size is the cardinal factor reckoned by multinational enterprises when they aim to seek markets through FDI Since then, the hypothesis can be derived as follow:

Hypothesis 1: Host country’s market size will have a positive impact on JFDI

3.1.2 Natural resource

Japan is a country with the limited natural resource This is the reason why Japanese FDI has been concentrating on host country’s natural resource to ensure the provision of raw materials for the production, for example, petroleum drilling in Indonesia; iron ore mining in Malaysia; copper mining in Philippines, etc (Urata, 1993) Moreover, in the research of Kinoshita et al (2004), the abundance of natural resources is regarded as the driver of FDI inflows Thus, I derive the following hypothesis:

Hypothesis 2: Host country’s natural resource will have a positive impact on JFDI

3.1.3 Inflation rate

Unpredictable and instable inflation rate can be the disadvantage for host country to attract FDI It is clearly stated that profit expectation of foreign firms would not be ensured because of volatile inflation rate In addition, high inflation rate can be one of main reasons leading to the lowered real earnings in domestic currency of foreign firms As a result, the study of Wadhwa

and Reddy (2011) indicated the negative relationship between inflation rate and FDI In this study, inflation rate was classified into the group of efficiency-seeking factors, which concentrate

on the motives of lower cost reduction and the more efficient business operation of foreign firms Hereby, it can be concluded the third hypothesis as follows:

Hypothesis 3: Host country’s inflation rate will have a negative impact on JFDI

3.1.4 Exchange rate volatility

As to economic theory, because of the depreciation of host country’s currency, the foreign firms can reduce the costs of buying the local inputs that leading to the lower production costs In the study of Xing (2006), the role of exchange rate is referred as the critical variable affecting Japanese FDI in China Under this author’s research, the appreciation in the Yen associated with the increase in FDI into China and vice versa Specifically, the author found that the fluctuation of Japanese FDI in China can be explained by the fluctuation of real bilateral exchange rate between China and Japan Moreover, in the study of Urata and Kawai (2000), the exchange rate does not have the important influence on FDI inflows whereas the exchange rate volatility has negative impact on JFDI inflows Therefore, the fourth hypothesis can be derived

of total export plus import to GDP to demonstrate the degree of trade openness The fifth hypothesis is as follows:

Hypothesis 5: Host country’s trade openness will have a positive impact on JFDI

3.1.6 Political risk

According to Buckley et al (2007), in the host countries with the higher rate of political risk, the multinational firms who want to find markets would choose to substitute exporting or licensing for owing production in local area Moreover, in the study of finding out the linkages between political risk, institution and foreign direct investment, Busse and Hefeker (2007) analyzed the relative importance of twelve indicators standing for political risk of a particular country on FDI inflows The result of this study shows that FDI inflows are significantly sensitive to the change in host country’s indicators of political risk such as government stability, internal and external conflict, law and order, quality of bureaucracy, corruption and ethnic tensions and democratic accountability of government The higher value of political risk index demonstrates the higher political stability In this study, we use the average of Rule of Law, Regulatory Quality and Government Effectiveness The sixth hypothesis can be derived as follows:

Hypothesis 6: The increase in host country’s political risk will have a negative impact on JFDI

3.1.7 Infrastructure development

By statistical analysis, Urata and Kawai (2000) examined and emphasized the importance

of the availability of infrastructure in attracting FDI by Japanese small and medium-sized enterprises For the time being, in some developing countries, it is rather hard for them to solve the problems related to infrastructure development because of the lack of finance The insufficient infrastructure is one of the main problems that Japanese firms would encounter and consider when investing in a particular country As to the study of Urata and Kawai (2000), most Japanese firms, who want to enter and set the investment in any country, always pay attention to the availability of electricity because it is considered as the significant factors to produce high quality products However, as to Vijayakumar et al (2010), there are some indicators used to measure infrastructure development, for example, the availability of transportation, telecommunications, electricity and water The combination of those indicators is necessary to reflect the host country’s infrastructure development Therefore, after considering the availability

of data, in this study, the infrastructure development index is built by indexing the level of electricity generation per person, the telephone lines (per 100 people) and the energy use (kg of oil equivalent per capita) The seventh hypothesis is set as following:

Hypothesis 7: Host country’s infrastructure development will have a positive impact on JFDI

3.1.8 Labor cost

With the purpose of seeking efficiency and market in host countries, the firms from developed countries outweighed the labor-related issues Among them, Mirza and Giroud (2004) referred in their study the significant role of labor cost in attracting FDI into Vietnam Because

of the pressure from other countries that have been in process of exploiting advantages for FDI inflows, it is really necessary for Asian developing countries to base on its endowment, such as, low labor cost, largely potential markets, a degree of innovatory capacity, etc Thus:

Hypothesis 8: Host country’s labor cost will have a negative impact on JFDI

In summary, the below table indicates eight hypotheses that will be tested in this thesis:

Table 3.1: Summary of testing hypotheses

H1 Host country’s market size will have a positive impact on JFDI

H2 Host country’s natural resource will have a positive impact on JFDI

H3 Host country’s inflation rate will have a negative impact on JFDI

H4 Host country’s exchange rate volatility will have a negative impact on JFDI H5 Host country’s trade openness will have a positive impact on JFDI

H6 The increase in host country’s political risk will have a negative impact on JFDI H7 Host country’s infrastructure development will have a positive impact on JFDI H8 Host country’s labor cost will have a negative impact on JFDI

The dataset includes yearly observations from 1995 to 2012 for ten Asian countries: Vietnam, Thailand, Indonesia, Malaysia, Philippines, Hong Kong, Korea, Singapore, India and China The dependent variable in this study is the Log of Japanese FDI (LJFDI) inflow in billion Yen The independent variables, chosen thoroughly basing on the previous studies and the consideration of the availability of dataset, include Purchasing Power Parity adjusted Gross Domestic Product (GDP), natural resource (NRE), inflation rate (IFL), exchange rate volatility (EXCV), political risk (POL), trade openness (TO1), infrastructure development (INFRA2) and labor cost (ARWAG)

To archive the second objective, which indicates the determinants of Japanese FDI in ten selected Asian countries, a panel data econometric model for this study was constructed based on the basic theories, the empirical studies and the availability of data as follows:

LnJFDI it = α0 + β1 GDP it + β2 NRE it + β3 IFL it + β4 LnEXCV it + β5 POL it + β6 TO1 it + β7

INFRA2 it + β8 ARWAG it + ε it

Where:

GDP it is Purchasing Power Parity adjusted Gross Domestic Product in constant 2005 international $ for country i at time t and is the proxy of market size (Billion US dollars)

NRE itis Natural Resource and is measured by the ratio of ore and metal exports to merchandize exports for country i at time t (%)

IFL itis the inflation ratefor country i at time t (%)

at time t The simple exchange rate index is constructed for selected countries as:

Y t = X t / X 1995

X 1995 is the value of exchange rate in 1995 for each country

X t is the value of exchange rate at time t for each country

Y t is the index value of exchange rate at time t for each country

TO1 itis Trade Openness for country iat time t and is measured by the ratio of total import and export to GDP (%)

POL itis Political Riskfor country i at time t and is measured by the average of Rule of Law, Regulatory Quality and Government Effectiveness Those data are collected from Worldwide Governance Indicators (WGI) This is the source of dataset produced by Revenue Watch and Brookings Institution, World Bank Development Research Group and World Bank Institute There are six aggregate WGI measures, whose data are in a standard normal distribution, mean

of zero, standard deviation of one and run from -2.5 to 2.5 The higher values stand for better governance

POL it = ∑3 Yjt / 3

j

Y jt is the value of the jth indicator at time t for each country

INFRA2 itis Infrastructure Development Index for country i at time t This is the average of the index of the level of electricity generation per person, the telephone lines (per 100 people) and the energy use (kg of oil equivalent per capita) Those data are collected from data source of World Bank The simple Infrastructure Development Index is constructed for selected countries as:

Y jt = X jt / X j1995

X j1995 is the value of jth indicator in 1995 for each country

X jt is the value of jth indicator at time t for each country

Y jt is the index value of the jth indicator at time t for each country Then we take the average

of the above Y jt to get the Infrastructure Development Index for each country i as follows:

INFRA2 it = ∑3 Yjt / 3

j

This data is collected from the International Labor Organization (ILO)

e it is the error time over the time t

Table 3.2: Summary of expected signs of variables

US dollars

Political risk POL: the average of Rule of Law

Index, Regulatory Quality Index and Government Effectiveness Index

Index

+

Worldwide Governance Indicators Infrastructure

development

INFRA2: The average of the index of the level of electricity generation per person, the telephone lines (per 100 people) and the energy use (kg of oil equivalent per capita)

3.3.1 Descriptive Analysis

In this section, by applying the method of descriptive statistics, the overview of JFDI before and after the crisis in 1997 will be presented The year of 1997 is chosen because it was the first threshold on which JFDI underwent the large change The descriptive analysis is conducted by going from the whole context to the detailed selected Asian countries By describing the changes of JFDI towards the crisis in 1997, this part aims to synthesize and analyze the specific factors of each selected Asian countries, which made them to be more or less attractive destinations of Japanese firms in this period

In the second section of this part, the ranking table of promising countries for JFDI from

2009 to 2013 will be also given with the purpose of discussing about the recent characteristics of JFDI in ten selected Asian countries By combining the figures, empirical studies and recent surveys, the analysis in both two sections partially contributes for the outline of determinants of JFDI

3.3.2 Regression Analysis

Data used for estimating the econometric model in this study is a panel dataset The panel data will be in the period of 1995 to 2012 and consist of 10 selected Asian countries, including Vietnam, Thailand, Indonesia, Malaysia, Philippines, Hong Kong, Korea, Singapore, India and China

There are three common methods used to analyze panel data They are Common Constant Method, Fixed Effects Method (FEM) and Random Effects Method (REM) Theoretically, increasing the accuracy in calculation is the main advantage of panel data, which can be easy to realize As to each individual, by the combination of many periods of data, the number of observations is accelerated The second benefit of panel data is in FEM, which allows the researchers to consider unobserved heterogeneity that may be correlated with the regressors On the other hand, one more advantageous method referred in panel data is REM, which manages the unobserved heterogeneity to be independent with the regressors Nevertheless, if the model includes the fixed effects, REM and Common Constant Method seem to be inappropriate.(Cameron & Trivedi, 2005)

The general model for panel data can be presented as follows:

Yit = α it + Xit β it + uit, i = 1, ,N; t = 1, ,T Where i stands for cross section or individual, t stands for the time, Yit is a dependentvariable,

Xit is aKx1 vector of independent variables, uit is disturbance term

The common constant method

The Common Constant Method is also called as Pooled OLS method This method is conducted under the basic assumption that there is no difference among data matrices of cross section dimensions In this model, the coefficients and intercepts are the same Therefore, the model of Pooled OLS method is as follow:

Yit = α + Xit β + uit

Despite being a simple method in panel data, Pooled OLS method still has the drawbacks It is very common that each individual i has time-invariant but unique effects whereas Pooled regression model ignores the heterogeneity between individuals and assumes the same coefficients for all of them, and then, the above-referred effects will be included in the error term