Effects of Foreign Aid: Evidence from Thailand (New Frontiers in Regional Science: Asian Perspectives, 50) 981162481X, 9789811624810

Citation preview

New Frontiers in Regional Science: Asian Perspectives 50

Hiroaki Sakurai

Effects of Foreign Aid Evidence from Thailand

New Frontiers in Regional Science: Asian Perspectives Volume 50

Editor-in-Chief Yoshiro Higano, University of Tsukuba, Tsukuba, Ibaraki, Japan

This series is a constellation of works by scholars in the field of regional science and in related disciplines specifically focusing on dynamism in Asia. Asia is the most dynamic part of the world. Japan, Korea, Taiwan, and Singapore experienced rapid and miracle economic growth in the 1970s. Malaysia, Indonesia, and Thailand followed in the 1980s. China, India, and Vietnam are now rising countries in Asia and are even leading the world economy. Due to their rapid economic development and growth, Asian countries continue to face a variety of urgent issues including regional and institutional unbalanced growth, environmental problems, poverty amidst prosperity, an ageing society, the collapse of the bubble economy, and deflation, among others. Asian countries are diversified as they have their own cultural, historical, and geographical as well as political conditions. Due to this fact, scholars specializing in regional science as an inter- and multi-discipline have taken leading roles in providing mitigating policy proposals based on robust interdisciplinary analysis of multifaceted regional issues and subjects in Asia. This series not only will present unique research results from Asia that are unfamiliar in other parts of the world because of language barriers, but also will publish advanced research results from those regions that have focused on regional and urban issues in Asia from different perspectives. The series aims to expand the frontiers of regional science through diffusion of intrinsically developed and advanced modern regional science methodologies in Asia and other areas of the world. Readers will be inspired to realize that regional and urban issues in the world are so vast that their established methodologies still have space for development and refinement, and to understand the importance of the interdisciplinary and multidisciplinary approach that is inherent in regional science for analyzing and resolving urgent regional and urban issues in Asia. Topics under consideration in this series include the theory of social cost and benefit analysis and criteria of public investments, socio-economic vulnerability against disasters, food security and policy, agro-food systems in China, industrial clustering in Asia, comprehensive management of water environment and resources in a river basin, the international trade bloc and food security, migration and labor market in Asia, land policy and local property tax, Information and Communication Technology planning, consumer “shop-around” movements, and regeneration of downtowns, among others. Researchers who are interested in publishing their books in this Series should obtain a proposal form from Yoshiro Higano (Editor in Chief, [email protected]) and return the completed form to him.

More information about this series at http://www.springer.com/series/13039

Hiroaki Sakurai

Effects of Foreign Aid Evidence from Thailand

Hiroaki Sakurai Faculty of Business Administration Bunkyo University Adachi, Tokyo, Japan

ISSN 2199-5974 ISSN 2199-5982 (electronic) New Frontiers in Regional Science: Asian Perspectives ISBN 978-981-16-2481-0 ISBN 978-981-16-2482-7 (eBook) https://doi.org/10.1007/978-981-16-2482-7 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Preface

This book summarizes the impact of foreign aid or official development assistance (ODA) from the perspective of economic growth, fiscal budget, and aid agencies, as in the case of Thailand. One of the most important desires of human beings is to be able to fulfill the minimum standards with regard to food, sanitary system, clothing, and housing. Since there has been economic growth all over the world post the Cold War, the abovementioned desire is bound to be attained, at least in calculation in the coming years. The assumption is that the developed countries donate 0.7% of GNI and that the developing countries use the donations efficiently. While foreign aid is an important instrument to fulfill this desire, there is no established theory in the field of economics and related disciplines that helps us understand if foreign aid really helps developing countries improve the livelihood of their poor. Thailand attained economic growth and poverty reduction during the same time it received foreign aid from Japan (one of its main donors), alongside the influx of foreign direct investment. There is a belief among aid practitioners that foreign aid, primarily aimed toward widening the access to social infrastructure, is one of the important elements necessary for increasing the living standards of people. This book focuses on the impact of foreign aid or official development assistance (ODA) from several aspects, using the case of Thailand. While there are many studies on the development of each country, there is a need for research that brings out the important elements of foreign aid, which are as follows. First, it is time that we investigate the role of foreign aid again, for poverty reduction is still an important global problem and small conflicts due to poverty and economic conflict still remain. This book presents a case of poverty reduction. Second, this approach is applicable to other developing countries as well, especially the neighboring countries of Thailand. Vietnam has come to rapidly grow in a similar way. Myanmar and Cambodia have also started to grow rapidly. This book presents a good example in this regard. Third, the bad reputation the word “ODA” carried was overthrown in Japan. In the 1980s, ODA was broadcasted as being associated with corruption or being useless. As a result, Public Opinion Survey by the Government of Japan shows v

vi

Preface

that the word “ODA” has come to carry a bad image, while the word “International Cooperation” has come to carry a good image in Japan. In fact, both these terms indicate the same meaning. As per the analysis advocated by this book, foreign aid to Thailand contributes toward economic growth. Moreover, the Thai government has a well-governed foreign aid and has managed to keep a sound management of finance by mainly reducing domestic borrowing as a substitute for foreign aid. Finally, a form of interdependent strategic relationship has come to be well established and well managed by the aid agencies. These results, obtained using the long-term period data, are consistent with the widely believed things, while the effect of foreign aid itself is still under discussion. This book is intended toward aid practitioners, including policy makers, researchers, and Ph.D. students. In addition, the findings of this book imply that other developing countries with similar policies could reinforce themselves by observing the experience of Thailand. There are many people whom I would like to thank, without whom this book would not have been possible. First, I am grateful to Dr. Yoshiro Higano, Regional Science Association International (RSAI), and Mr. Yutaka Hirachi, Springer Japan, who gave me the opportunity to publish this book. Next, this book is a revised version of my doctoral dissertation presented at Hitotsubashi University. I also thank Dr. Professor Motohiro Sato, who was my dissertation’s primary supervisor, and Dr. Professor Yukichi Mano, who was dissertation’s co-supervisor. They guided me well on my project. In addition, I am grateful to my dissertation committee members, namely Dr. Professor Izumi Yokoyama, Professor Masaki Nakahigashi, and Dr. Professor Hiroyuki Taguchi, for their critical and valuable comments. I would like to thank Dr. Professor Jota Ishikawa, whose seminar I attended during my second grade and who continues to advise me. I would also like to thank Dr. Hiromitsu Ishi, who was my supervisor during my undergraduate seminar. In addition, I would like to thank many people who stay in Thailand, in particular, the National Economic and Social Development Board (NESDB) staff and the researchers of Thailand Development Research Institute (TDRI), who taught me a lot about statistics, practical methods, and the policy of Thailand. I received great hospitality during my stay at the Economic Planning Agency, the Cabinet Office, the Ministry of Foreign Affairs in Japan, and Embassy of Japan in Thailand. Finally, the financial support for this research came from JSPS KAKENHI Grant-in-Aid for Research Activity Start-up 18H05685 and 19K20886, and Bunkyo University. I am immensely grateful for their funding. Tokyo, Japan April, 2021

Hiroaki Sakurai

Contents

1

Foreign Aid, Poverty Reduction, and Economic Growth in Thailand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Economic Growth and Poverty Reduction in Thailand . . . . . . . . . 1.2 Official Development Assistance (ODA) to Thailand . . . . . . . . . . 1.2.1 Definition of ODA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.2 History of ODA to Thailand . . . . . . . . . . . . . . . . . . . . . . 1.2.3 Thai Government as a Recipient Country . . . . . . . . . . . . . 1.2.4 Plaza Agreement in 1985 and FDI . . . . . . . . . . . . . . . . . . 1.3 Purpose of This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

1 1 3 4 5 9 12 12 13

2

Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Foreign Aid and Economic Growth . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Impacts of Foreign Aid . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Productivity of Social Infrastructure . . . . . . . . . . . . . . . . . 2.3 Foreign Aid and Fiscal Condition . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Relationship Among Aid Facilities . . . . . . . . . . . . . . . . . . . . . . . 2.5 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

15 15 16 16 17 17 18 19 19

3

Statistics in Thailand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Region and Population Statistics in Thailand . . . . . . . . . . . . . . . . 3.3 National Accounts of Thailand . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.1 National Income in Thailand . . . . . . . . . . . . . . . . . . . . . . 3.3.2 GPP and Socio-economic Survey in Thailand . . . . . . . . . . 3.3.3 Capital Stock in Thailand . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Foreign Aid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Fiscal Data in Thailand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 Labor Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

23 23 24 24 27 28 29 29 31 32 vii

viii

4

5

6

Contents

3.7 Price . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.8 Ways of Acquiring Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

33 34 34 35

Foreign Aid Loans and Economic Growth in Thailand . . . . . . . . . . 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Estimation of the Effect of the Foreign Aid in the Entire Country . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.3 Estimation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.4 Marginal Production Effects . . . . . . . . . . . . . . . . . . . . . . 4.2.5 Comparison of the Public Capital . . . . . . . . . . . . . . . . . . . 4.2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Estimation of the Effects of the Foreign Aid (Yen Loan) in Regional Panel Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1 Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.3 Estimation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. .

37 37

. . . . . . .

38 39 40 40 42 42 43

. . . . . . .

44 45 46 47 49 49 49

Fiscal Effects of Foreign Aid in Thailand . . . . . . . . . . . . . . . . . . . . . 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Methodology and Estimation Results . . . . . . . . . . . . . . . . . . . . . 5.3.1 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.2 Total Governmental Expenditure . . . . . . . . . . . . . . . . . . . 5.3.3 Categorizing Consumption Expenditure and Capital Expenditure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.4 Impact of Foreign Aid in the 1960s and the 1970s . . . . . . 5.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

51 51 52 54 54 55

. . . .

57 58 62 62

Relationships Between Aid Agencies . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Methodology and Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.2 Simple Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.3 Data and Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Estimation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

65 65 67 67 67 68 69 74 75

Contents

7

Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1 Summary of Research Results . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Research Results and Further Research . . . . . . . . . . . . . . . . . . . . 7.2.1 Research Results and the Purpose of This Book . . . . . . . . 7.2.2 Further Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 ODA Has a Bad Impression in Japan . . . . . . . . . . . . . . . . . . . . . 7.4 One Suggestion to the Future Foreign Aid . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix

. . . . . . . .

77 77 78 78 79 79 81 82

Abbreviations

Foreign Aid ODA OOF

Official Development Assistance Other Official Flows

Royal Thai Government BOT MOF MOL NESDB NESDC TICA

Bank of Thailand (Central Bank in Thailand) Ministry of Finance Ministry of Labour National Economic and Social Development Board National Economic and Social Development Council (Organizational change from NESDB in 2018) Thailand International Cooperation Agency

Aid Facilities in Japan JBIC JICA OECF

Japan Bank for International Cooperation (foreign aid loan including yen loan from 1999 to 2008) Japan International Cooperation Agency (dealing with technical assistance, grant aid, aid loan after 2008) Overseas Economic Cooperation Fund (foreign aid loan including yen loan up to 1999)

International Facilities ADB DAC OECD WB

Asian Development Bank Development Assistance Committee Organisation for Economic Co-operation and Development World Bank

xi

xii

Abbreviations

Currencies THB USD

Thai Baht US Dollar

Chapter 1

Foreign Aid, Poverty Reduction, and Economic Growth in Thailand

Abstract This chapter describes the background knowledge of foreign aid in Thailand over 50 years, including economic growth, poverty reduction, and foreign direct investment from the practitioners’ point of view. The main points are summarized for the following three points. First, it is shown that Thailand experienced nearly 5% growth for this period continuously, and the poverty ratio drastically reduced during the time. Second, the policy stance for both donor countries and the Thai government changed during the late 1970s and the beginning of the 1980s, and foreign aid was mainly donated as aid loans for constructing social infrastructure or social capital. Third, social capital constructed by foreign aid has been used efficiently since the foreign direct investment was enlarged drastically by the yen appreciation after 1985. Overall, these are the main reasons practitioners believe foreign aid has a positive effect on Thailand as a recipient country. Keywords Thailand · Economic growth · Poverty · Official development assistance

1.1

Economic Growth and Poverty Reduction in Thailand

Thailand has experienced relatively rapid economic growth for more than half a century under the system of relatively free economic activity. Figure 1.1 shows the economic growth and per capita GNI in Thailand from 1961, which indicates that the economic growth in Thailand records around 5% in most years. During this time, economic growth was especially high in the late 1980s and the beginning of the 1990s. In contrast, soon after the 1997 financial crisis and the beginning of the 1980s, the growth rate was low. Although there has been some economic boom and recession, the per capita GDP in Thailand has increased and has now reached more than 140 thousand Thai baht (around 5000 USD) as a middle-income country. In addition, the average income is more than doubled if limited to the Bangkok vicinity area, almost the level of developed countries. Considering the lower wage and land price level, the real living standards feel much better. For example, compared with Japan, because one Thai baht is converted into 10 Japanese Yen, it looks as if 1.5 million yen is one-third of Japan’s GDP per capita. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 H. Sakurai, Effects of Foreign Aid, New Frontiers in Regional Science: Asian Perspectives 50, https://doi.org/10.1007/978-981-16-2482-7_1

1

2

1 Foreign Aid, Poverty Reduction, and Economic Growth in Thailand

Fig. 1.1 Economic growth in Thailand. (Source: World Development Indicators)

Fig. 1.2 Ratio of poor people living under 1.9 USD per day in Thailand. (Source: World Development Indicators)

Due to this relatively rapid economic growth in the long term, poverty in Thailand has also been reduced. Figure 1.2 shows the poverty line defined by the World Bank as spending 1.9 US Dollar or below per day. This figure indicates that the poverty ratio in Thailand records around 5% in 1981, below 1% in 1994, and under 0.1%, almost zero after 2007. In addition, from the aspect of the standard of living in Thai

1.2 Official Development Assistance (ODA) to Thailand

3

Fig. 1.3 Poverty line and poverty ratio according to the definition by Thai government. (Source: NESDC)

people, Thai government also calculates poverty line and poverty ratio using the socio-economic survey held by Thai government. Poverty line is defined by the total expenses for necessities, such as foods, clothes, and shelter, in normal Thai people. Poverty ratio shows the people living below the poverty line. The results of poverty line and poverty ratio are shown in Fig. 1.3, which indicates that the poverty ratio has decreased since 1988 and nearly the same at around 10% in the last 5 years. Now, the poor people in Thailand mainly consist of elderly people living in the Northeast area.

1.2

Official Development Assistance (ODA) to Thailand

In this section, fundamental knowledge about ODA in Thailand is summarized based on the following three aspects: definition, budget, and policy stance of the Thai government. By considering these aspects, the following three points are inferred. First, the policy stance of both donor countries and the Thai government as a recipient country changed during the late 1970s and the beginning of the 1980s, meaning that the budget for foreign aid of the donor countries was increased under the cold war and that the policy stance of Thai government as a recipient country has welcomed foreign aid since the 1980. Second, the main portion of the ODA or foreign aid is dominated by foreign aid loans for constructing the social infrastructure, meaning that most of them were used to strengthen the productivity of the country with careful consideration about cost and benefit analysis just in case of returning default risk. Third, foreign direct investment (FDI) increased drastically after the 1985 Plaza Agreement, meaning that the social capital of the ODA has been

4

1 Foreign Aid, Poverty Reduction, and Economic Growth in Thailand

used efficiently because many manufacturing industries were in Thailand soon after the yen appreciation in Japan.

1.2.1

Definition of ODA

Supporting developing countries is generally called economic cooperation and is divided into official development assistance (ODA), other official flows (OOF), private flows (PF), and donations by NGOs. Most studies use the word “foreign aid” as “ODA,” and this book also uses this term. According to the Organisation for Economic Co-operation and Development (OECD, n.d.) webpage, the Development Assistance Committee (DAC) defined ODA up to 2017 data1 as follows: Those flows to countries and territories on the DAC list of ODA recipients and to multilateral institutions, which are: 1. provided by official agencies, including state and local governments, or by their executive agencies; and 2. each transaction of which: – is administered with the promotion of the economic development and welfare of developing countries as its main objective. – is concessional in character and conveys a grant element of at least 25% (calculated at a rate of discount of 10%). The aid loan is divided into the ODA and OOF using Grant Element as the following equation (Development Assistance Committee 2013):  GE ¼ 100 1 


r  a

d

1

n

1 ð1þdÞaG

 ð1þd1 ÞaM

o

dðaM  aGÞ

where GE ¼ Grant element M ¼ Maturity G ¼ Grace period A ¼ Number of repayments per year INT ¼ Interval between the commitment date and the first repayment date minus the interval between two successive repayments ¼ G  1/A DR ¼ Repayment duration ¼ M  INT I ¼ Discount rate (10% ¼ 0.1) R1 ¼ Interest rate during grace period R2 ¼ Interest rate during repayment period

1

This book uses only data up to 2017, although the definition is altered after 2018.

1.2 Official Development Assistance (ODA) to Thailand

5

Fig. 1.4 Official development assistance to Thailand per GDP. (Source: DAC)

D ¼ Discount rate per period ¼ ((1 + I)(1/A))  1 NR ¼ Total number of repayments ¼ A * DR C1 ¼ (1 + I)INT C2 ¼ (1 + I)DR This book concentrates on ODA as a foreign aid, although private funds and NGOs are also important for the country to develop.

1.2.2

History of ODA to Thailand

Next, the amount of ODA and the elements of the decision of ODA are summarized. The official development assistance to Thailand per GDP is shown in Fig. 1.4. From the graph, ODA/GDP ratio was over 1% form the late 1970s and the beginning of the 1980s. Since a low loan is returned after borrowing, the offset from the gross is ODA. Figure 1.4 shows that the ODA was more than 1% during the 1980s, where the ODA has been negative in recent years. This fact shows that the importance of ODA toward Thailand has almost finished. In addition, Japan is one of the primary donors to Thailand. Around 55% was dominated by ODA from Japan in total between 1968 and 2016. Since foreign aid from Japan to Thailand is summarized well in the Institute for International Cooperation, Japan International Cooperation Agency (2003) and Ishii (2016), this chapter uses this definition. According to this definition, the history of ODA from Japan to Thailand is summarized based on the following five stages (Table 1.1):

6

1 Foreign Aid, Poverty Reduction, and Economic Growth in Thailand

Table 1.1 Socio-economic movement and foreign aid to Thailand

Notes: Author’s summary from the Institute for International Cooperation, JICA (2003), Kato (2016), and Ishii (2016). Aid amount was obtained from the DAC

1.2 Official Development Assistance (ODA) to Thailand

1. 2. 3. 4. 5.

7

Start-up (from 1968 to 1977) Strategic extension (from 1977 to 1989) Conversion of values (from 1989 to 1996) Financial crisis (from 1997 to 2000) Thailand as a donor country (from 2000)

The ODA history is also discussed in Kato (2016). The first stage shows the starting in 1968 as one way of compensating the World War II. During the first stage, the Vietnam War continued, and Saigon fell in 1975. In the same year, Khumer Rouge established Democratic Kampuchea in Cambodia, and Lao PDR was established as a community country. As a background of concern devoted to the war, Japanese products began to be rejected in Thailand in 1977. As a result, Fukuda doctrine came out in 1977, and the budget of ODA was doubled. This was the beginning of the second stage. In 1980, an internal war occurred in Cambodia, and the importance of Thailand increased as a base camp. A big project like the Eastern Sea Board was established and extended during this period. The second stage was classified during the Cold War until 1989. The third stage began at the end of the Cold War in the world. During the same time, the Cambodia War was over and the new movement happened under the slogan “peaceful peninsula” or “from the war to the peace in Indo-china.” In 1997, the currency crisis occurred in Thailand, and the struggle for calm down lasted for 3 years, which is called the fourth stage. After 2000, Thailand changed its position from the recipient country to the donor toward neighboring countries, which was the fifth stage. The Thailand International Cooperation Agency (TICA) was founded and donated to Laos, Cambodia, and Myanmar. As a whole, ODA does not necessarily have a limited purpose and is used widely for social infrastructure. In addition, foreign aid from Japan is used for economic reasons, not for political purposes. Next, we see how the budget of the ODA to Thailand is decided. Although some elements are claimed from previous studies, it is possible to estimate the use of these five stages. Reasons for deciding the amount of foreign aid, that is, budget of the foreign aid, can be considered by these elements. From a practical point of view, it is considered that the amount of the budget is decided based on the previous year. From this assumption, the amount of foreign aid is regressed by foreign aid in five stages as dummy variables. The result is shown in Eqs. (1.1) and (1.2). Equation (1.1) is depicted in Fig. 1.5, and Eq. (2) is depicted in Fig. 1.6. Equation (1.1): ODA from DAC countries NET ODAt ¼ 154:00 þ 567:72aid1t þ 1012:04aid2t þ 1359:37aid3t þ 1139:78aid4t þ ut ð68:48Þ** ð114:13Þ*** ð104:60Þ*** ð118:60Þ*** ð153:12Þ*** Adj: R2 ¼ 0:784 D:W: 1:873 ð1:1Þ

Equation (1.2): ODA from Japan

8

1 Foreign Aid, Poverty Reduction, and Economic Growth in Thailand

Fig. 1.5 Net ODA from DAC countries to Thailand and estimation. (Source: DAC. real base in 2010)

Fig. 1.6 Net ODA from Japan to Thailand and estimation. (Source: DAC. real base in 2010)

NET JODAt ¼ 225:32 þ 351:15aid1t þ 800:75aid2t þ 1066:56aid3t þ 1070:04aid4t þ ut ð64:61Þ*** ð107:68Þ*** ð98:69Þ*** ð111:90Þ*** ð144:47Þ*** Adj: R2 ¼ 0:734 D:W: 1:951 ð1:2Þ

Note: Standard error in parenthesis, ***: significant at 1%, **: significant at 5%, *: significant at 10% (Statistics from DAC, World Bank, era classification in Table 1.1).

1.2 Official Development Assistance (ODA) to Thailand

9

In each equation, the net ODA to Thailand is estimated by dummy variables categorized by Japan’s ODA from the first to the fifth stage. This equation means that the budget of the ODA from DAC countries or from Japan is nearly the same until international reputation is changed. In contrast, estimation by other variables does not necessarily fit the ODA. Budget of the ODA to Thailand has been distributed based on the three points. First, the aid amount was stable and changed drastically, reflecting the international evaluation of Thailand. Since the whole world and Japan have similar trends, this judgment is similar in both DAC countries and Japan. Second, other elements such as trade, foreign direct investment, and internal reserves do not necessarily have a significant impact on the ODA. Third, aid budgets in donor countries are not necessarily significant. For example, the aid budget in Japan increased until 1998 and decreased thereafter. However, it is ineffectively estimated. This is partly because the source of the aid budget includes a low-interest loan.

1.2.3

Thai Government as a Recipient Country

Finally, this section summarizes the policy stance of the Thai government and the main portion of the ODA. Since Thailand has maintained independence before World War II, the bureaucrat system in the government facilitated well for governing inside the country. In addition, due to advice from the World Bank, the Thai government made an institution for developing the country and increasing people’s living standards inside the country. The initial organization was called the National Economic Council (NEC) in 1950. NEC was renamed as National Economic Development Board (NEDB) in 1959, and the First Economic Development Plan started in 1961. Subsequently, the economic development plan has been decided every 5 years. The development plan includes not only economic development but also redistribution. It also includes projects inside Thailand, together with other ministries. Budget is also decided with Ministry of Finance, NESDB, Bank of Thailand, and Bureau of Budget. The important thing is that economic bureaucrat systems are normally independent of the military. Although not all projects are promoted as the plan shows, it contributes to economic growth in Thailand. From the perspective of the ODA, many projects are together with the economic development plan, such as the Eastern Sea Board and redistribution policies in rural areas. Recipient countries have also worked very hard. In addition, Prime Minister Prem began to encourage investment from foreign countries at the beginning of the 1980s. Finally, the usage of ODA can be seen. ODA to Thailand is mainly used for each project. Specifically, aid toward Thailand focused on constructing social infrastruc-

10

1 Foreign Aid, Poverty Reduction, and Economic Growth in Thailand

Fig. 1.7 ODA in Bangkok area. (Source: JICA)

ture. Some of them are in the Bangkok area, such as main bridges in the Chao Phraya River and highways (Fig. 1.7). Others are constructed in the Eastern Sea Board, located 80–200 km from Bangkok. In this area, natural gas was found in the 1970s, and the Bangkok Sea port was so crowded that a deep-sea port was necessary at the same time. Around 20% of yen loans, ODA from Japan, between 1982 and 1993 was used for constructing deep-sea ports, power plants, highways, and industrial areas (Fig. 1.8). These social infrastructures have helped Thai economy as social capital, as shown in Table 1.2. The background is Japan’s style. First, Japan’s ODA does not use political purposes or matters by reflecting on the experience of World War II. Consequently, Japan’s ODA is concentrated on economic issues, which consist of the fundamentals of better lifestyles through economic growth. Second, the ODA is used for the project base by the low-interest loan of the Japanese yen (Yen Loan), not general budget support. By the obligation of repayment, recipient countries consider selecting projects well. It is also possible to consider the ideas by using benefit cost analysis. Third, accumulating human capital is also important. As a

1.2 Official Development Assistance (ODA) to Thailand

11

Fig. 1.8 ODA in Eastern Seaboard. (Source: OECF 1999) Table 1.2 Examples of output by the ODA to Thailand Transportation Electricity Irrigation

– 14 bridges in Chao Phraya River have been constructed – More than half a million employees are established in Eastern Seaboard Around 15% electricity was constructed or refined at the head of twenty-first century Around 30% of water supply to farmers are operated by the ODA

Source: Institute for International Cooperation, JICA 2003, and Embassy of Thailand in Japan et al. 2008

12

1 Foreign Aid, Poverty Reduction, and Economic Growth in Thailand

result, together with loans, social infrastructure and technical assistance are donated at the same time. For example, power plants are constructed by low-interest loans and the operators of the new power plant are fostered by technical assistance.

1.2.4

Plaza Agreement in 1985 and FDI

Additionally, this section mentions the FDI toward Thailand, since it is natural to feel whether the social capital made by ODA is used efficiently or not. In September 1985, yen appreciation began due to the Plaza Agreement. Many Japanese manufacturing companies are annoyed regarding export from Japan and have changed their location to avoid the yen appreciation. Thailand is a good place, since social capital is arranged and wages are relatively low. As a result, FDI toward Thailand was drastically larger in the late 1980s, as shown in Fig. 1.9. Industrial areas were filled with factories and many people began to work there.

1.3

Purpose of This Book

Considering the above points, this book shows the impact of the ODA to Thailand using semi-macroeconomic data. The fundamental idea is classified into the following three points.

Fig. 1.9 Yen exchange rate and FDI in Thailand. (Source: World Development Indicators and Bank of Japan)

References

13

First, statistical data are used mainly from public data. As pointed out in Chap. 3, most statistics have been arranged after the 1997 crisis, although the ODA has lasted since 1968. In addition, most statistics are only inside the ministry or libraries, not in the Internet. In this context, acquiring long-term data is one of the main operations. Second, they show evidence of practitioners’ beliefs, such as aid agencies and NGOs, from an economic point of view. In particular, Thailand and Japan have corporation, including social infrastructure and foreign direct investment. Now, the people’s standard of life in Thailand normally satisfies their living needs, although research on the process of development is necessary. This book selects the economic model based on aid schemes. Currently, some under-developed countries are in the process of developing. This method is also useful for such countries. Third, this study contributes to the interdisciplinary studies. Thai studies include several approaches, such as politics and socialism, and new findings. However, this book is based on the macroeconomic approach, considering the connection of other fields.

References Development Assistance Committee (2013) Converged statistical reporting directives for the creditor reporting system (CRS) and the annual DAC questionnaire addendum 2. DCD/DAC (2013)/15/ADD2/FINAL Embassy of Japan in Thailand, Japan Bank for International Cooperation, and Japan International Cooperation Agency Thailand Office (2008) Nihon to Thai no keizai kaihatsu kyouryoku (in Japanese) Institute for International Cooperation, Japan International Cooperation Agency (2003) Country study for Japan's official development assistance to the Kingdom of Thailand - From “Development Assistance” to a “New Cooperation” (in Japanese) Ishii R (2016) An analysis on Japanese cooperation towards Thailand as an emerging' donor. Fukuoka Univ Rev Commerc Sci 60(3):525–546. (in Japanese) Kato H (2016) Chapter 1: Japan’s ODA 1954–2014: changes and continuities in a central instrument in Japan’s foreign policy. In: Kato H, Page J, Shimomura Y (eds) Japan’s development assistance. Palgrave Macmillan, London Organisation for Economic Co-operation and Development (n.d.) Official development assistanceassistance – definition and coverage. http://www.oecd.org/dac/financing-sustainable-develop ment/development-finance-standards/officialdevelopmentassistancedefinitionandcoverage.htm. Accessed 15 Jan 2021 Overseas Economic Cooperation Fund (1999) Eastern Seaboard development plan impact evaluation

Chapter 2

Literature Review

Abstract This chapter summarizes previous studies on the effects of foreign aid based on three fields: economic growth, fiscal budget, and the relationship among aid agencies. The lessons from previous studies are summarized as follows. First, finding out the way to economic growth using foreign aid is quite important, since it is clear that economic growth leads to poverty reduction as a robust result. In addition, inviting foreign direct investment is also important because it is easier to promote economic growth, whereas foreign aid itself causes economic growth. Second, maintaining fiscal discipline is important to have an impact from foreign aid to the budget, since many studies show that government consumption increases when foreign aid is increased. Third, the cooperation among aid agencies is important because in-cooperation among aid agencies sometimes causes the concentration of aid in a certain field. Overall, the major opinion is that the governance of recipient countries decides the size of the impact of foreign aid. Keywords Foreign aid · Productivity effect · Economic growth · Fungibility · Fiscal budget · Aid cooperation

2.1

Introduction

One of the dreams of human beings is enjoying a long-time healthy life without any fear, such as infectious diseases and famine. To achieve this dream, improving the living standard by increasing income is essential. Similarly, economic growth is a solution at the country level. In the individual level, food, clothes, and living must be affordable; in the country level, sanitary system, such as water and sewage systems, vaccines and medical care system, should be introduced. Through this process, life expectancy is expected to be extended due to the improvement in infant mortality and nutrition. The source of improving the living standard is economic growth at the country level. In the beginning of this chapter, the impact of economic growth on living standards must be summarized. Previous studies robustly show that economic growth reduces poverty (Ravallion 2001; Dollar and Kraay 2002; Besley and © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 H. Sakurai, Effects of Foreign Aid, New Frontiers in Regional Science: Asian Perspectives 50, https://doi.org/10.1007/978-981-16-2482-7_2

15

16

2 Literature Review

Burgess 2003). In this regard, the next important point is the policy to promote economic growth. If concentrating on the relationship with foreign capital, previous studies show that foreign direct investment (FDI) is one way to strengthen productivity (Borensztein et al. 1998; Hsiao and Hsiao 2006). In addition, research shows that the development of transportation systems extends economic growth (Canning and Pedroni 2008).

2.2

Foreign Aid and Economic Growth

As indicated in Sect. 2.1, economic growth is one of the solutions for improving living standard. In addition, the policy for introducing FDI is also a good policy for reducing poverty. However, foreign aid is not as easy as introducing FDI. In this section, the relationship between foreign aid and economic growth is surveyed by the previous studies. Compared with the impact of FDI, foreign aid is difficult to grasp the effect. In Sect. 2.2.1, the impact of foreign aid mainly using cross-country data is summarized as the main stream of the impact of foreign aid. In Sect. 2.2.2, the productivity effect of public capital is summarized. Since productivity effect is so simple, it is useful even if the data is limited in developing countries using timeseries data.

2.2.1

Impacts of Foreign Aid

One of the characteristics of the relationship between foreign aid and economic growth is widely diversified in this century and is not a unified idea. Since the literature review during the twentieth century is summarized in some studies (McGillivray et al. 2006; Doucouliagos and Paldam 2009), this chapter mainly focuses on the discussion after 2000. At the beginning of the twenty-first century, the main idea about the impact of foreign aid was that “foreign aid is effective as long as the governance of the recipient country is well organized” (World Bank 1998; Alesina and Dollar 2000; Burnside and Dollar 2000; Collier and Dollar 2002). Some conditions may be added, such as more difficulty in tropical zones (Dalgaard et al. 2004) and offset effect by human capital as an instrumental variable (Hansen and Tarp 2001). In contrast, some scholars claim that foreign aid has almost no effect on recipient countries. Reasons for stating are that the estimated results are fragile if terms and explanatory variables are changed, and that sometimes a large part of the aid is not for the purpose directly, such as conference fees (Easterly 2003, 2006, 2007; Easterly et al. 2004). Nwaogu and Ryan (2015) show that foreign aid has no effect by using the error correction model in African and Latin American countries. Recently, the relationship between foreign aid and economic growth in the short term is easy to find relatively, although it is more difficult in the long term

2.3 Foreign Aid and Fiscal Condition

17

(Nowak-Lehmann et al. 2012). In addition, only the second and third industries have the effect of foreign aid by using industrial data (Selaya and Thiele 2010). Overall, the industry is easier to help find out the impact inside the country (Addison et al. 2017). The impact of foreign aid on economic growth using time-series data analysis is summarized by Addison and Tarp (2015). Arndt et al. (2015) show a positive relationship between foreign aid and economic growth from 1970 to 2007 using panel data from 78 countries. In addition, Juselius et al. (2014) examine the VAR model and Granger causality tests using data from 36 countries in Africa from the 1960s to 2007 and show that foreign aid has a positive effect. In addition, it is also insisted that the ODA is used for foreign countries, that is, ODA helps FDI contribute to economic growth. This is called the vanguard effect. Kimura and Todo (2010) checked the Southeast Asian region and found this effect. In contrast, Nowak-Lehmann et al. (2012) insisted that this vanguard effect is not shown, although the effect of foreign aid is seen in the short term.

2.2.2

Productivity of Social Infrastructure

The productivity of social infrastructure is discussed during the twentieth century. This part is summarized based on Iwamoto (2005) and Miyagawa et al. (2013). The productivity effect of social capital is measured in the USA. In the United States, social capital productivity was lowered after the construction of social infrastructure, such as highways and schools in the 1970s, and the lowered productivity was like that of total factor productivity. Aschauer (1989) estimates the production function from macroeconomic data in the USA and shows the productivity effect in the social infrastructure. In Japan’s case, it is considered that social infrastructure may be supplied at an adequate level after the period of rapid economic growth since social infrastructure had been constructed in the 1990s, and most of them found productivity (Iwamoto 1990; Mitsui and Inoue 1997). From a regional point of view, social infrastructure in urban areas is too small (Yoshino and Nakano 1994), and from an industrial point of view, productivity in the first industry is small (Yoshino et al. 1999). Recently, Japan has also faced a deterioration problem regarding social infrastructure because some of them were constructed until the 1970s. Miyagawa et al. (2013) surveyed the productivity effect in Japan.

2.3

Foreign Aid and Fiscal Condition

If the recipient country wastes foreign aid, the effect of aid will be offset. An example is that when the donor country constructed the power plant, the recipient country constructed an Opera house using saved money (Nurkse 1953). This

18

2 Literature Review

problem occurs when the concept of donor and recipient country is different, and from the donor side, the effect is different from the plan. This problem is called fungibility. Fungibility is divided into effects in each field, such as education and health, and the fiscal budget response itself. The fiscal response includes domestic borrowings and governmental revenue as well as governmental expenditure (Morrissey 2015). This chapter focuses on the fiscal budget response. From a theoretical point of view, Fiscal Response Models (FRMs) are used. FRMs maximize the policymaker’s utility function under the governmental budget constraint. Calibrating the model and solving the optimal solution will lead to the understanding of the impact of foreign aid (Franco-Rodriguez et al. 1998). Another idea is to use the VAR model because utility function and effective way may be more complicated. In this regard, foreign aid and fiscal budget (i.e., domestic borrowing, governmental expenditure, and governmental revenue). Specifically, the cointegrated VAR (CVAR) model can be used when these variables are in cointegration, and both short- and long-term relationships can be examined at a time. Therefore, the impact of foreign aid on the fiscal budget (i.e., domestic borrowings, governmental expenditure, and governmental revenue) is divided into short-term and long-term relationships by using the CVAR model. Osei et al. (2005) examine the relationship between foreign aid and government budget from 1966 to 1998 using the CVAR model and impulse response and shows that it looks meaningless to foreign aid due to the increase in the short term, but that it restricts domestic borrowings and governmental capital expenditure in the long term. Examples of similar analyses are Lloyd et al. (2009), Morrissey et al. (2007), and Martins (2010). Most literatures are examined in African countries since foreign aid is thought not to be used efficiently. In contrast, few studies have been conducted on Asian countries partly because the fungibility is not necessarily found. Khan and Hoshino (1992) is an example, which shows that at least a part of the foreign aid flowed into governmental consumption from India, Pakistan, Bangladesh, Sri Lanka, and Malaysia from 1955 to 1976. McGillivray and Ahmed’s (1999) findings also show an increase in government consumption due to an increase in foreign aid from the Philippines (1960–1992).

2.4

Relationship Among Aid Facilities

Aid Cooperation promotes the project under the cooperation of some aid agencies. From the aid agency, the effect can be acquired at a lower price. Promoting international cooperation will provide efficient aid and attain the recipient country. Torsvik (2005) makes the two donors and one developing country model, and solves the amount of aid and redistribution in the developing country in the case of cooperation game and in-cooperation game. The results show that the amount of aid is larger in the cooperation game than in the in-cooperation game. In addition,

References

19

redistribution is smaller in the cooperation game than in the in-cooperation game (Kihara 2011). Each research compares with the cooperation and in-cooperation game Nash equilibrium and shows that the cooperation game is more efficient than the in-cooperation game. One of the elements to offset the aid effect is the aid flood, and a huge amount of aid is concentrated (Arimoto and Kono 2009; Kimura et al. 2007). Based on Arimoto and Kono (2009), aid flood raises the cost and scramble for the current cost, foreign currency, and specialists of the recipient country. The relationship between aid agencies began in the 1980s during the Cold War, and distinguished with either the Cournot equilibrium or the Lindahl equilibrium in the 1990s (Sandler and Murdoch 1990; Sandler 1992; Chiaki and Fukao 1993). The political approach is the war with terrorism and aid cooperation (Mascarenhas and Sandler 2006; Kihara 2011). In addition, fluctuation of aid amount shows the relationship among each aid agency under the assumption that aid agencies decide the aid budget endogenously. From the experience that the aid amount changed significantly due to the 2008 financial crisis, the way to make up for the gap between the long-term goal and annual budget should be researched (Jones 2015; De Matteis 2018).

2.5

Concluding Remarks

In this chapter, previous studies on the effect of foreign aid have been summarized. First, it is widely believed that the impact of foreign aid on economic growth is mainly dependent on the governance of the recipient countries. However, some insist that foreign aid has almost no effect. Specifically, it is easier to determine the impact relatively in the short term, but not in the long term. Second, the impact of foreign aid is normally offset by the governmental organization by increasing the governmental consumption, which is called “fungibility.” The fungibility in the country level is mainly seen in African countries. Third, the corporation among the aid agencies will lead to a greater impact efficiently. In contrast, the following issues have still not been studied enough. First, the effect of the aid is different by aid schemes, such as aid loans, grant aid, and technical assistance. Second, it is difficult to judge whether it is a corporation game or in-corporation game when applying the game theory to the empirics. More research and extension are expected in this field.

References Addison T, Tarp F (2015) Aid policy and the macroeconomic management of aid. World Dev 69:1–5

20

2 Literature Review

Addison T, Morrissey O, Tarp F (2017) The macroeconomics of aid: overview. J Dev Stud 53(7):987–997 Alesina A, Dollar D (2000) Who gives foreign aid to whom and why? J Econ Growth 5(1):33–63 Arimoto Y, Kono H (2009) Foreign aid and recurrent cost: donor competition, aid proliferation and budget support. Rev Dev Econ 13(2):276–287 Arndt C, Jones S, Tarp F (2015) Assessing foreign aid’s long-run contribution to growth and development. World Dev 69:6–18 Aschauer DA (1989) Is public expenditure productive? J Monet Econ 23(2):177–200 Besley T, Burgess R (2003) Halving global poverty. J Econ Perspect 17(3):3–22 Borensztein E, De Gregorio J, Lee J (1998) How does foreign direct investment affect economic growth? J Int Econ 45(1):115–135 Burnside C, Dollar D (2000) Aid, policies, and growth. Am Econ Rev 90(4):847–868 Canning D, Pedroni P (2008) Infrastructure, long-run economic growth and causality tests for cointegrated panels. Manchester School 76(5):504–527 Chiaki M, Fukao K (1993) Foreign aid as a provision of impure public goods. Keizai Kenkyu 44 (1):1–14. (in Japanese) Collier P, Dollar D (2002) Aid allocation and poverty reduction. Eur Econ Rev 46(8):1475–1500 Dalgaard C, Hansen H, Tarp F (2004) On the empirics of foreign aid and growth. Econ J 114(496): F191–F216 De Matteis A (2018) Follow the leader! The peer effect in aid supply decisions. Dev Policy Rev 36 (6):631–648 Dollar D, Kraay A (2002) Growth is good for the poor. J Econ Growth 7(3):195–225 Doucouliagos H, Paldam M (2009) The aid effectiveness literature: the sad results of 40 years of research. J Econ Surv 23(3):433–461 Easterly W (2003) Can foreign aid buy growth? J Econ Perspect 17(3):23–48 Easterly W (2006) The white man’s burden: why the West’s efforts to aid the rest have done so much ill and so little good. The Penguin Press, New York Easterly W (2007) Was development assistance a mistake? Am Econ Rev 97(2):328–332 Easterly W, Levine R, Roodman D (2004) Aid, policies and growth: comment. Am Econ Rev 94 (3):774–780 Franco-Rodriguez S, Morrissey O, McGillivray M (1998) Aid and the public sector in Pakistan: evidence with endogenous aid. World Dev 26(7):1241–1250 Hansen H, Tarp F (2001) Aid and growth regressions. J Dev Econ 64(2):547–570 Hsiao FST, Hsiao MW (2006) FDI, exports, and GDP in East and Southeast Asia—panel data versus time series causality analysis. J Asian Econ 17(6):1082–1106 Iwamoto Y (1990) An evaluation of public investment policy in postwar Japan. Econ Rev 41 (3):250–261. (in Japanese) Iwamoto Y (2005) Kokyotoshi ha yakuni tatteirunoka. In: Ohtake F (ed) Ohyo Keizaigaku Heno Sasoi. Nippon Hyoron Sya., pp 115–136 (in Japanese) Jones S (2015) Aid supplies over time: addressing heterogeneity, trends, and dynamics. World Dev 69:31–43 Juselius K, Møller NF, Tarp F (2014) The long-run impact of foreign aid in 36 African countries: insights from multivariate time series analysis. Oxf Bull Econ Stat 76(2):153–184 Khan HA, Hoshino E (1992) Impact of foreign aid on the fiscal behavior of LDC governments. World Dev 20(10):1481–1488 Kihara T (2011) Enjyo donor no keizaigaku. Nippon Hyoron Sya (in Japanese) Kimura H, Todo Y (2010) Is foreign aid a vanguard of foreign direct investment? A gravityequation approach. World Dev 38(4):482–497 Kimura H, Sawada Y, Mori Y (2007) Aid proliferation and economic growth: a cross-country analysis. RIETI discussion paper series 07-E-044 Lloyd T, McGillivray M, Morrissey O, Opoku-Afari M (2009) The fiscal effects of aid in developing countries: a comparative dynamic analysis. In: Mavrotas G, McGillivray M (eds)

References

21

Development aid: a fresh look. Palgrave Macmillan UNU-WIDER Studies, Basingstoke, pp 158–179 Martins MG (2010) Fiscal dynamics in Ethiopia: the cointegrated VAR model with quarterly data. CREDIT research paper, 10/05 Mascarenhas R, Sandler T (2006) Do donors cooperatively fund foreign aid? Rev Int Org 1 (4):337–357 McGillivray M, Ahmed A (1999) Aid, adjustment and public sector fiscal behaviour in the Philippines. J Asia-Pacific Econ 4:381–391 McGillivray M, Feeny S, Hermes N, Lensink R (2006) Controversies over the impact of development aid: it works; it doesn’t; it can, but that depends. J Int Dev 18(7):1031–1050 Mitsui K, Inoue J (1997) The productivity effect of social capital. Jpn Econ 25(3):3–29 Miyagawa T, Kawasaki K, Edamura K (2013) Reconsideration of the social infrastructure production effect. Keizai Kenkyu 64(3):240–255. (in Japanese) Morrissey O (2015) Aid and government fiscal behavior: assessing recent evidence. World Dev 69:98–105 Morrissey O, M’Amanja D, Lloyd T (2007) Aid and government in Kenya: a time series analysis. In: Lahiri S (ed) Theory and practice of foreign aid. Elsevier, Amsterdam, pp 313–332 Nowak-Lehmann F, Dreher A, Herzer D, Klasen S, Martínez-Zarzoso I (2012) Does foreign aid really raise per capita income? A time series perspective. Can J Econ 45(1):288–313 Nurkse R (1953) Problems of capital formation in underdeveloped countries. Basil Blackwell, Oxford Nwaogu UG, Ryan MJ (2015) FDI, foreign aid, remittance and economic growth in developing countries. Rev Dev Econ 19(1):100–115 Osei R, Morrissey O, Lloyd T (2005) The fiscal effects of aid in Ghana. J Int Dev 17(8):1037–1053 Ravallion M (2001) Growth, inequality and poverty: looking beyond averages. World Dev 29 (11):1803–1815 Sandler T (1992) Collective action: theory and applications. University of Michigan Press, Ann Arbor Sandler T, Murdoch JC (1990) Nash-Cournot or Lindahl behavior?: an empirical test for the NATO allies. Q J Econ 105(4):875–894 Selaya P, Thiele R (2010) Aid and sectoral growth: evidence from panel data. J Dev Stud 46 (10):1749–1766 Torsvik G (2005) Foreign economic aid; should donors cooperate? J Dev Econ 77(2):503–515 World Bank (1998) Assessing aid – what works, what doesn’t, and why. World Bank, Washington, DC Yoshino N, Nakano H (1994) Allocation of public Investment in the Metropolitan Area. In: Hatta T (ed) Economic analysis of the concentration in Tokyo. Nihon Keizai Shimbun sha, Japan. (in Japanese) Yoshino N, Nakajima T, Nakahigashi M (1999) Productivity effect of public capital. In: Yoshino N, Nakajima T (eds) Economic effect of public investment. Nippon Hyoron sha, Japan. (in Japanese)

Chapter 3

Statistics in Thailand

Abstract This chapter summarizes Thailand’s statistics from the macroeconomic analysis viewpoint. After the 1997 crisis, although Thailand has been promoting statistics, some data have occasionally been missed, specifically in the long term. This book mainly uses a system of national accounts and indicates the following points. First, stock statistics are not bifurcated province wise. Second, the gross provincial product (GPP) is published only from the production side, not expenditure side. Third, official development assistance (ODA) data are not necessarily divided into provinces or fields. Keywords National accounts · Labor statistics · Capital stock · Population

3.1

Introduction

Needless to say, to use and analyze statistics, it is important to grasp its characteristics. In addition, the kind of statistics used would reflect on the characteristics. When analyzing time-series data, one needs to be careful about the sequence. Some statistics are compiled by Thailand’s National Statistical Office (NSO), an agency of the Ministry of Digital Economy and Society, which prior to 2016 was known as the Ministry of Information and Communication Technology. In addition, most statistics are generated by individual supervised ministries. With regard to the system of national accounts, the National Economic and Social Development Council (NESDC) is the central planning agency. Most contents are published only in books, not on the internet. Even if a small amount of information exists, it sometimes changes. In addition, before 1997, Thailand’s statistics sometimes had insufficient samples due to a lack of budget. Caution is necessary while using data from the 1970s or 1980s. Moreover, the Thai study as a regional research is so sophisticated that Japanese articles are also available. This chapter refers to Suehiro (1998) for summarizing statistics in Thailand prior to the 1997 crisis; Ishii (2016) for summarizing recent movements; and Kumagai (2011a, b, 2012a, b, 2014) for summarizing statistics as well as recent movements. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 H. Sakurai, Effects of Foreign Aid, New Frontiers in Regional Science: Asian Perspectives 50, https://doi.org/10.1007/978-981-16-2482-7_3

23

24

3.2

3

Statistics in Thailand

Region and Population Statistics in Thailand

Thailand is divided into four major areas: northeast, north, central, and south. Its borders have not changed, ever since it was formed by watersheds in mountainous areas (Fig. 3.1). Long-term data normally consists of these four areas, but recent data consist of seven areas, since the central area has been further subdivided into east, west, and Bangkok vicinity. At any rate, it is necessary to use the four areas when using data from the 1970s. In Thailand, there are now 77 provinces, but their number has increased after the 1970s. Hence, caution is necessary when using data related to provinces. In Thailand, population censuses have been conducted every 10 years. This survey began in 1909, but was suspended during World War II, and has been undertaken in years having “0” (zero) as its last digit, e.g., 1970 and 1980. Since its coverage included all those residing in Thailand, it was called the “Population and Housing Census.” The results of the 2010 survey are shown in Table 3.1. From Thailand’s total population of around 66 million, more than half live in rural areas. Considering each area, nearly 10% reside in Bangkok, and more than 25% in Central, where agriculture (more than biannual cropping) has been developed along the Chao Phraya River, as well as second and third industries due to its proximity to Bangkok. In contrast, nearly 30% are living in the northeast, where it is single cropping due to the lack of water. Given that the Population and Housing Census has a long history of surveying all samples, its historical data is easy to grasp. Figure 3.2 shows the population and annual growth rate, beginning from the first survey in 1909. The data shows that Thailand’s population has been increasing, except during World War II, but this trend changed from the 1990s, and it now faces depopulation due to its aging and declining birth rate.

3.3

National Accounts of Thailand

Thailand’s national accounts data are managed by the NESDC.1 One of its main purposes is formulating development plans every 5 years, along with other economic, social, and land policies. In addition, important policies involving decisions beyond a single ministry are made by using inter-ministry conferences. Thailand’s national accounts are divided into: national income, capital stock, GPP, flow of funds, and input-output tables. However, this book has used only data on national income, GPP, and capital stock available on the internet since the 1970s,

1 This organization was known as the National Economic and Social Development Board (NESDB) until 2018.

3.3 National Accounts of Thailand

Fig. 3.1 Four areas and 77 provinces in Thailand. (Source: National Statistical Office 2015)

25

26 Table 3.1 Number and percentage of population by sex, area, and region

3

Region and area Whole Kingdom Municipality Non-municipality Region Bangkok Central Northern Northeastern Southern

Population Total Male 65.98 32.35 29.13 14.12 36.85 18.23 8.3 18.18 11.66 18.97 8.87

4.03 8.93 5.72 9.28 4.39

Statistics in Thailand

Female 33.63 15.01 18.62 4.27 9.25 5.94 9.69 4.48

Percent 100 44.2 55.8 12.6 27.6 17.7 28.7 13.4

Source: National Statistical Office 2013

Fig. 3.2 Total population in Thailand and annual population growth rate, 1909–2010. (Source: National Statistical Office 2013)

under different versions of system of national account. Although national accounts are convenient, some caution is necessary due to some traits of original statistics. NESDC is in the process of changing its estimation method to Chain Volume Measures, although this book uses the old estimation method. Therefore, some statistics are different from this book, although it is assumed that the trend has not changed.

3.3 National Accounts of Thailand

27

Fig. 3.3 Gap between the production side and expenditure side in Thailand. (Source: Kumagai 2011b)

3.3.1

National Income in Thailand

The annual survey of Thailand’s national accounts commenced in 1951. In addition, soon after the 1997 crisis, quarterly data began in 2001, and data can now be obtained from 1993 onwards. Until the 1970s, national income was measured only from the production side based on the industrial values added at each stage of production. However, after 1980, national accounts comprised not only the production side, but also the expenditure and distribution sides. Thailand’s national accounts are mainly examined based on the production side, partly because grasping production activity is relatively easy, and the produce directly connects to the profit. Conversely, in Japan, for instance, it is mainly examined from the expenditure side, partly because household surveys are relatively large. Figure 3.3 shows the gap between the gross domestic product (GDP) from the production and expenditure sides, as posited by Kumagai (2011b). It shows that the GDP from the production side is estimated to be larger. GDP as seen in Fig. 1.1 indicates that for more than half a century around 5% growth has continued, and that the gross national product per capita has become the level of a middle-income country. In addition, NESDC which is introducing the Chain Volume Method needs to be careful while taking long-term data since it is estimated in a different way.

28

3.3.2

3

Statistics in Thailand

GPP and Socio-economic Survey in Thailand

The GPP in Thailand compiled by the NESDC is available for nominal and real values, added in every province. The value added by several main industries can also be separated since 1981. Although GPP is convenient for comparisons among provinces, it is compiled only from the production or inter-provincial side. Since the transfer or trade between provinces is not surveyed, the expenditure side, or residential concept, is not estimated. Therefore, GPP is not necessarily reflected in each province’s living standards. For example, the young generation who are working and sending remittances to their parents in rural areas are not reflected in the GPP statistics. Under this assumption, the value added in each area for each of the three industries in 2009 is shown in Fig. 3.4. This figure indicates the following two points. First, from a comparison of areas, the central region, located in the Eastern Sea Board and Bangkok Vicinity, is the largest. In addition, in southern areas, plantations such as rubber and fisheries are prosperous because of their tropical weather, including heavy rainfall. Second, from the industry perspective, since the third industry comprises the service sector, such as restaurants and travel, it is the largest. Next in weightage is the second industry, mainly comprising the manufacturing industry. In contrast, the first industry is not necessarily given much weightage. Next, the living standards and GPP in each region were examined. Thailand has surveyed the revenue and expenditure of 4000 households all over the nation almost

Fig. 3.4 Value added of every industry in each area in 2009. (Source: NESDC)

3.4 Foreign Aid

29

Table 3.2 Comparison between the socio-economic survey and gross provincial product (2009)

Source: National Economic and Social Development Board (NESDB) National Accounts and National Statistics Office (NSO) Household Socio-Economic Survey Note: Gross Provincial Product is the annual value added per person, and the Household SocioEconomic Survey is the monthly revenue of households

every 2 years, which is called the “Household Socio-Economic Survey.” This survey can be divided into every region. Table 3.2 shows the comparison between the Household Socio-Economic Survey and GPP in each region. From the table, as compared with Bangkok vicinity the northeastern region scored 14% and 41% in the GPP and Household Socio-Economic Survey, respectively.

3.3.3

Capital Stock in Thailand

Capital stock statistics are available from 1970 by industry, throughout the country. Although it is easy to acquire data from the internet, it is not bifurcated into provinces or areas. Figure 3.5 shows the capital stock statistics for three industries. The third industry has a higher rate than that of the other two, partly because real estate, transportation, and telecommunications dominate at a certain rate. In addition, the movement was steeper after the 1980s, which shows that capital stock had accumulated with rapid economic growth.

3.4

Foreign Aid

Since ODA to Thailand has already been stated in Chap. 1, this chapter mainly summarizes from the data source viewpoint. The ODA data were divided into three fields. First, international facilities such as the Organisation for Economic Co-operation and Development (OECD) and World Bank show statistics on foreign aid. Second, evaluation form of the donor country. In the case of Thailand, yen loans

30

3

Statistics in Thailand

Fig. 3.5 Capital stock data by industry. (Source: NESDC)

are the main source of ODA. Hence, the evaluation of the yen loan is available as statistics. Third, the fiscal budget statistics in Thailand, a recipient country. Details, including some notes, are described below. First, Development Assistance Committee (DAC), one organization in OECD, statistics form the main part of ODA data. In addition, the World Bank often shows foreign aid data. Such data from international facilities sum up the foreign aid from DAC countries. DAC statistics including the amount of net ODA and loan amount, as well as the gross ODA, is available on the internet. Although this data covers a long-term period, it is not necessarily divided into categories. ODA by fields includes only the recent several years. In addition, aid by non-OECD countries, e.g., China, are not included. Second, donor countries’ data are also available. In the case of Japan, one of the major donors in Thailand compiles the evaluation form relating to yen loans—the main part of foreign aid from Japan—with which the evaluation for 200 million yen and over began from the middle of the 1980s. Hence, it is possible to see the trends in the field and area by analyzing the yen loan evaluation. Table 3.3 shows big projects in 1993, when the Laem Chabang deep-sea port project was completed. From the table, it is seen that most big projects were dominated by construction and transportation. Since foreign aid has a history of evaluation, creating a dataset is possible by collecting data from the evaluation. Third, from the side of the recipient country, fiscal budget statistics published by the Thai government show the acceptance of foreign aid. Since the amount of foreign aid was quite different in the 1950s and 1980s, for these decades, the template of the table is different. The Fiscal Budget in Brief, published by the Thai government’s

3.5 Fiscal Data in Thailand

31

Table 3.3 Main projects in 1993 Project Bangkok-Chonburi Highway Construction Project Agricultural Credit for Rural Development Project (BAAC loan) Rural Electrification Project (P) Rama Sixth Bridge (P) Public Telecommunications Map Ta Put port Leam Chabang port Mekuwan port Total

Amount (bil. Yen) 13 5 7 4 1 3 8 7 48

Ministry of Finance, shows the aid amount by some aid agencies from 1977 onwards.

3.5

Fiscal Data in Thailand

Thailand’s fiscal data are available from two facilities. First, the Government of Thailand’s Ministry of Finance (MOF), which has been issuing the fiscal budget since the middle of the 1970s. Its recent version available on its webpage summarizes expenditure, revenue, and borrowing, and also includes how it is used by each ministry. Second, Bank of Thailand (BOT)—the central bank—issues the Economic Bulletin, which includes fiscal statistics since the middle of the 1950s, although the terms used for the statistics have slightly changed over the years. Although the time length is longer in the Economics Bulletin by the BOT, most statistics were altered soon after 1997. In addition, budget details are shown in the fiscal budget issued by MOF. Since the Royal Thai government is centrally administered, most expenditures are directly distributed by each ministry. Therefore, execution of the budget in each province is often included in the budget of each ministry, and it is difficult to know the contents of the budget at the provincial level. In addition, the provincial-level budgets do not always include the total expenditure in the respective provinces. Finally, we depict the movement of the fiscal budget shown in Fig. 3.6. Originally, the Thai government did not like to increase the number of its borrowings. However, it received a loan from the World Bank at the start of the 1980s. During the economic expansion of the 1980s, it did not increase its borrowings. However, after the 1990s, the economic boom became a bubble economy, and the 1997 crisis occurred. During this time, borrowing was larger. In the twenty-first century, fiscal discipline was relatively severe.

32

3

Statistics in Thailand

Fig. 3.6 Domestic borrowing, governmental expenditure, and governmental revenue in Thailand. (Source: BOT until 1996 and MOF after 1997)

3.6

Labor Statistics

Labor statistics in Thailand are divided into three parts: (1) Labor Force Survey held by the NSO, (2) Structure of Labor Force and Employment (name differs by years) from the Ministry of Labour, and (3) some other materials, but it is impossible to use them in the twentieth century due to sample bias. Therefore, this section summarizes two statistics from the perspectives of characteristics and traits of labor statistics in Thailand. First, the Labor Force Survey includes numbers of the labor force, employees, and unemployment rates. It is published every quarter as a book, and has been periodically conducted since 1971. Second, the Structure of Labor Force and Employment includes the number of employees and offices and is managed by the Ministry of Labour by gathering data from its local branches in each province. It has been available for a relatively long time, beginning in the 1980s, with a stable output, and easy connection over the period. The trends are summarized for the following three points. First, yearly drastic seasonal movements because many farmers are coming to the urban area as seasonal workers during the free season of farming. Second, more people came to work in manufacturing industries during the 1980s and 1990s. Third, the tertiary industry is the most common.

3.7 Price

3.7

33

Price

Price statistics comprise the Consumer Price Index (CPI), Producer Price Index (PPI), and GDP deflator. The CPI is formulated by the Ministry of Commerce (MOC), PPI by the Bank of Thailand, and the GDP deflator by the NESDC. Since CPI and the GDP deflator are used in this book, this section also focuses on these two statistics based on the methods and trends. We shall summarize the rough methods of devising the CPI and GDP deflator in Thailand, and their trends since the 1960s. First of all, the mode of compiling CPI statistics is surveyed by the staff in the MOC branch located in each province. They ask shops about the weekly price levels, and the central office of the MOC collects this information. Since the data is collected by the branches themselves, the compilation of the statistics is quite fast. The CPI of the previous month can be seen on the MOC webpage at the beginning of each month (third working day). Next, the method for calculating the GDP deflator is introduced. The benchmark year for Thailand’s GDP deflator was 1988 in the former version used in this book. Since the GDP deflator is habitually twisted far from its benchmark year, NESDC is trying to replace the GDP deflator with the Chain Volume Measure, and since its benchmark year is the previous year it would solve the problem of the year getting twisted, far from the benchmark year. After 1990, the Chain Volume Measure was once again changed in 2018. This book uses the old version since the time-series method is suitable for analyzing long periods. Finally, the inflation trend over a long period is shown in Fig. 3.7, indicating the following three points. First, the early 1970s and 1980s recorded high inflation (20%

Fig. 3.7 Inflation rate from CPI and GDP deflator (Source: MOC and NESDC). Note: GDP deflator is different before 1996, and after 1998

34

3

Statistics in Thailand

or more). At the beginning of the 1980s, Thailand experienced recession, together with inflation. Second, after the 1997 financial crisis, both inflation and deflation were experienced. Specifically, negative inflation of the GDP deflator in 1999 means that the domestic demand shrank rapidly. Third, except for the descripted above two exclusions, the records were stable at less than 5%. Since the CPI includes prices of energy and perishable foods, it shows relatively high rates. In addition, the Bank of Thailand introduced an inflation target of 0% to 3.5% in the core inflation of the CPI after 2000.

3.8

Ways of Acquiring Statistics

Some basic recent data were collected from the Bank of Thailand’s English webpage. In this case, there remained some problems as follows. First, it is relatively difficult to acquire the same statistics in the long-term period since the contents and available terms of the statistics have sometimes changed. Second, the method used for compilation of the statistics is not described in the webpage. In contrast, most statistics in Thailand are published as books and reports. The NSO sells its statistics in the library of the Ministry of Digital Economy and Society. Similarly, the labor statistics yearbook is available by the Ministry of Labour. In addition, libraries in some universities permit entry on payment (for instance, Chulalongkorn University charges 20 baht for entrance).

3.9

Conclusion

Lack of data is annoying when analyzing developing countries. Thailand is not an exception, even though its “Population and Housing Census” began more than 100 years ago, and became sophisticated, especially after the 1997 crisis. Collecting and analyzing long-term data is essential for Thailand, since its economic growth and foreign aid have continued for more than half a century. This book summarizes the lack of analysis due to data constraints in the following three points. First, only industry wise capital stocks are available for the whole country, which means that it is difficult to use the growth accounts for regional or provincial panel data. Second, the GPP is based on the production side, and does not consider the expenditure side because there are no statistics for adjusting. Given this measuring methodology, the gap between the Household Socio-Economic Survey is large. When considering the living standards in each area or province, GPP traits are also considered. Third, foreign aid statistics by field and area are nonexistent, which means that it is difficult to estimate the impact of foreign aid, especially technical assistance, and small donations. Even when concentrating on Yen Loans in big projects, it is

References

35

difficult, because most of these projects are included in the social infrastructure, such as transportation or construction. Thus, it is not easy to consider the impact based on industries. Under these restrictions, the impact of foreign aid on Thailand is measured from the points of view of growth, fiscal, and agency aid in Chaps. 4, 5, and 6, respectively.

References Ishii M (2016) Series: the state of statistics in emerging regions. Part 6: Thailand. Johokanri 56 (2):108–115. (in Japanese) Kumagai S (2011a) Thailand from Macroeconomic Statistics (1) Quarterly production side GDP. Tai koku zyouhou 9:63–70. (in Japanese) Kumagai S (2011b) Thailand from Macroeconomic Statistics (2) Quarterly expenditure side GDP. Tai koku zyouhou 11:54–61. (in Japanese) Kumagai S (2012a) Thailand from Macroeconomic Statistics (7) Area and provincial product. Tai koku zyouhou 9:73–81. (in Japanese) Kumagai S (2012b) Thailand from Macroeconomic Statistics (8) Capital stock. Tai koku zyouhou 11:80–90. (in Japanese) Kumagai S (2014) Thailand from Macroeconomic Statistics (17) Price statistics. Tai koku zyouhou 9:63–70. (in Japanese) National Statistical Office (2013) Executive summary, the 2010 population and housing census. Government of Thailand, Bangkok National Statistical Office (2015) Statistical yearbook of Thailand. Government of Thailand, Bangkok Suehiro A (1998) Tai no toukei seido to shuyo keizai, seiji de-ta. Institute of Developing Economies, Taiwan. (in Japanese)

Chapter 4

Foreign Aid Loans and Economic Growth in Thailand

Abstract This chapter examines the productivity of governmental aid loans, the main portion of foreign aid in Thailand, using the economic growth model both in whole-country time-series data from 1971 to 2013 and in regional panel data from 1986 to 2013. The effect of foreign aid remains a point of discussion, and we hope to clarify this issue. Thailand is considered a good example of a country using governmental aid loans to develop social capital, which then fosters manufacturing industries. Three main outcomes exist. First, we determine the impact of governmental loans on economic growth in Thailand using whole-country data. Second, the marginal production effect of governmental loans has recently been reduced, although not to zero. Third, the impact of foreign aid is lower than that of public capital in Thailand in the same period. Fourth, the quantitative impact of yen loans from the Japanese government can be determined using subregional data. Overall, foreign aid has an impact on economic growth, although the magnitude is smaller than that of public capital. Moreover, the impact is large at the beginning of the development, whereas it is smaller in the process of economic growth. Keywords Foreign aid · Official development assistance (ODA) · Economic growth · Productivity effect · Yen loan

4.1

Introduction

The effect of foreign aid remains a point of discussion. Recently, governance in recipient countries is believed to determine the effect of foreign aid (Burnside and Dollar 2000). In contrast, another study has limited focus on foreign aid and is instead based on a macroeconomic perspective (Easterly et al. 2004; Easterly 2006, 2007). In addition, recent studies have demonstrated that foreign aid has a short-term productivity effect in some industries. Nowak-Lehmann et al. (2012) state that a relationship exists between foreign aid and economic growth in the short term. Selaya and Thiele (2010) show that the second and third industry determined the effect of foreign aid. Arndt et al. (2015) present the relationship between economic growth and foreign aid by using panel data from 89 countries from 1970 to 2007. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 H. Sakurai, Effects of Foreign Aid, New Frontiers in Regional Science: Asian Perspectives 50, https://doi.org/10.1007/978-981-16-2482-7_4

37

38

4 Foreign Aid Loans and Economic Growth in Thailand

The history of economic growth in Thailand is as follows. Generally, the real GDP growth rate in Thailand has been around 5% for half a century. In particular, the economic policy was changed to encourage foreign investment at the beginning of the 1980s. This supported the high growth rate in the late 1980s and the beginning of the 1990s. Foreign aid from Japan began in 1968 as compensation for the effects of World War II. In 1977, the Japanese government decided that the foreign aid be doubled. Simultaneously, natural gas was found in the gulf of Thailand; however, the congestion of the Bangkok central district was a concern. The Eastern Seaboard plan was suggested at the beginning of the 1980s and consisted of constructing a deep-sea port, power plant, highways, industrial area, and water supply using foreign aid loans from Japan. Since the late 1980s, many Japanese factories were located in Thailand partly because of the good infrastructure. Thus, Asian countries have experienced rapid economic growth. In particular, Thailand appears to be attaining rapid economic growth using foreign aid mainly for social infrastructure and complementary foreign direct investment. Summarizing and sharing this experience in developing countries is important. Kimura and Todo (2010) demonstrate that foreign aid leads to foreign direct investments and that this relation connects to Asian countries’ economic growth. From the perspective of regional studies, JICA (2003) summarized the effect of foreign aid to Thailand from the input and output perspective. In addition to foreign aid, the productivity of public capital stock has been considered (Aschauer 1989). In the USA, this analysis was conducted to show whether old infrastructure lowered economic growth. In Japan, the productivity of public capital stock was believed to be reduced after the 1990s. Research showed that the effects of social infrastructure remained apparent although the effect was weaker. In addition, less infrastructure is found in rural areas than in urban areas, and public capital in the primary industry was lower than in other areas (Miyagawa et al. 2013). This study examines the productivity of foreign aid in Thailand by measuring the productivity of public capital as one method to develop an economy involves using foreign aid effectively.

4.2

Estimation of the Effect of the Foreign Aid in the Entire Country

In this section, the impact of foreign aid loans is examined in the following three ways: productivity of foreign aid loans by using economic growth from Sects. 4.2.1 to 4.2.3, marginal productivity in Sect. 4.2.4, and productivity of public capital in Sect. 4.2.5.

4.2 Estimation of the Effect of the Foreign Aid in the Entire Country

4.2.1

39

Model

We assume a Cobb–Douglas production function as follows: Y t ¼ At K αt L1α t

ð4:1Þ

where At: total factor productivity in year t Kt: capital in year t Lt: labor in year t Yt: gross domestic product in year t Total factor productivity At depends on foreign aid and the industrial structure similarity index as foreign aid transfers skills from donor to recipient countries, and industrial change leads to changes in the resource allocation of labor and capital. In other words, At ¼ GðAIDt , SIt Þ

ð4:2Þ

where AIDt: accumulated foreign aid until year t SIt: industrial similarity index SIt represents the industrial similarity in 5 years and is known as the similarity index. Originally it is defined as the cosine similarity: Pn AB i¼1 Ai  Bi similarityðA, BÞ ¼ ¼ P   1=2 Pn n 2 2 1=2 kAk  kBk i¼1 Ai i¼1 Bi where values change between 1 and 1, where 1 is perfectly dissimilar and 1 is perfectly similar. Wolff (2002) applied this index to industrial change by measuring the effect of ICT investment. Finally, Miyagawa et al. (2013) used the industrial change effect in the analysis of the effect of public capital. The definition is compared with the industrial construction 5 years ago, which is defined as follows: Pn

i¼1 Y t5,i Y t,i 1=2 Pn 2 1=2 2 i¼1 U t5,i i¼1 Y t,i

SIt ¼ P n

ð4:3Þ

where n cycles through different industries. This index uses values between 0 and 1 because the value added is only positive. If industrial change is quite small, this index approaches 1. If this index is close to zero, the industrial change becomes large.

40

4 Foreign Aid Loans and Economic Growth in Thailand

Table 4.1 Data description of the whole-country data In(Y/L ): GDP per capita In(AID): Governmental Loan to Thailand SI: Similarity Index In(K/L ): Capital per Employee

Mean 11.153 11.164 0.998 12.368

Std 0.453 1.507 0.003 0.503

Min 10.310 7.490 0.990 11.683

Max 11.796 12.664 1.000 13.041

A linear model was adopted for the induced changes in the above model. By taking the logarithm and first derivative of Eq. (4.1), we have the following:     Yt Kt ln ¼ const: þ βðSIt Þ þ γ ln ðAIDt Þ þ α ln þ ut Lt Lt

4.2.2

ð4:4Þ

Data

The variables Yt, Kt, and SIt are determined from the national accounts published by the National Economic and Social Development Board (NESDB) of the Thai Government. In this analysis, we use the following three sectors: the first, second, and third industries. Here, Lt is the number of employees according to the Ministry of Labor (MOL) of the Thai Government, and AIDt denotes DAC statistics from the OECD. Net ODA is conventionally used; however, in this case, we use the accumulated disbursement of the governmental loan from 1971, as most government loans in Thailand are used for the social infrastructure to strengthen the industry, such as constructing power plants, highways, and deep-sea ports. All data periods were from 1971 to 2013. The data descriptions are shown in Table 4.1.

4.2.3

Estimation Results

First, unit root tests are necessary. Thus, all variables are shown as I(1) and I(2), which may generate a spurious regression problem. Checking whether each estimation equation is cointegrated is necessary. Therefore, we should have unit root tests for error terms for level series estimation.1 Table 4.2 presents the results of the estimation in Eq. (4.4) based on the data. SIt is excluded from Eq. (4.1) and in Eq. (4.2). The estimation period runs until 2013, and a dummy variable is added after 2001 in Eq. (4.3). In Eqs. (4.4) and (4.5), the

1 Cointegrated VAR model is insignificantly estimated. Although estimation results are shown as the level series, similar estimated results are obtained by using the first difference.

4.2 Estimation of the Effect of the Foreign Aid in the Entire Country

41

Table 4.2 Estimation results across the whole country Dependent variable: Estimation equation Estimation period In(AID) In(K/L )

In(Y/L) (4.1) 1971–2013 0.066 (0.037)* 0.802 (0.106)***

SI

(4.2) 1971–2013 0.067 (0.031)** 0.787 (0.094)*** 4.625 (4.415)

POST2001 Const. AR(1) MA(1) Adjusted R2 D.W.

0.488 (1.111) 0.838 (0.138)*** 0.336 (0.189)* 0.994 1.902

5.277 (1.165) 0.818 (0.159)*** 0.273 (0.214) 0.995 1.942

(4.3) 1971–2013 0.069 (1.888)* 0.799 (0.107)***

0.016 (0.088) 0.507 (1.138) 0.832 (0.145)*** 0.327 (0.192)* 0.994 1.880

(4.4) 1971–2000 0.095 (0.010)*** 0.590 (0.031)*** 14.371 (3.028)***

(4.5) 1971–2000 0.051 (0.066) 0.859 (0.112)***

17.111 (3.003)*** 0.386 (0.313) 1.000 (67.726) 0.994 1.880

0.068 (1.383) 0.880 (0.110)*** 0.606 (0.181)*** 0.991 1.887

Notes: Standard errors in parentheses. ***: significant at 1%, **: significant at 5%, *: significant at 10%. The error terms in all equations are I(0). That is, these equations are cointegrated POST2001: dummy variable after 2001 is 1

estimation period is limited to 2000. As equations may have serial correlation, we add AR(1) and MA(1). From the results in Table 4.2, we see that the foreign aid terms other than in Eq. (4.5) are estimated to be positive and significant. In Eqs. (4.1) and (4.3), we find significance at the 10% level and in Eq. (4.2) at 5%. This is partly because the standard error of foreign aid is larger because of reductions in foreign aid in the present century. The coefficient of foreign aid in Eq. (4.4) is larger than that in Eqs. (4.1), (4.2), and (4.3), which implies that the effect of foreign aid was greater in the initial period. The coefficient of SIt was negative in that initial period because the industrial structure was changed. Comparing the effects of foreign and per capita capital, the coefficient for per capita capital is nearly ten times larger than that for foreign aid in Eqs. (4.1), (4.2), and (4.3), which indicates that investment is larger than foreign aid. SI is negative, which shows that the industrial change promotes. Thus, Eq. (4.2) is not significant partly because the industrial change stopped during the 1997 crisis in Thailand. Finally, examining the spurious regression problem is necessary. The error terms in all equations in Table 4.2 are I(0). Therefore, we infer that these relations are cointegrated. The Johansen test confirms that cointegration is only one.

42

4 Foreign Aid Loans and Economic Growth in Thailand

Table 4.3 Marginal effect of foreign aid Y/AID γ (coefficient of ln(AID) in Table 4.2) Profit ratio in foreign aid

From 1971 to 2000 21.1 0.095((4), Table 4.2) 200.50%

From 1971 to 2013 16.75 0.066((1), Table 4.2) 110.60%

From these equations, we infer that the productivity effect of the governmental aid loan exists to a certain extent.

4.2.4

Marginal Production Effects

Next, marginal production effects are examined. The coefficient of the production effects shows by what percentage the national income will increase if foreign aid is ∂Y AID ∂Y Y increased by 1%. Hence γ ¼ ∂AID Y . Upon changing to a new variable ∂AID ¼ γ AID, Y it follows that by calculating AID, we can determine the marginal effects of foreign aid. Y is 21.10 between the years 1971 and 2000 and 16.75 between the The value of AID years 1971 and 2013. As γ, which is the coefficient of AIDt in Eq. (4.4) in Table 4.2, is equal to 0.095, the marginal effect of foreign aid between 1971 and 2000 is 2.005. Similarly, as γ in Eq. (4.1) in Table 4.2 is equal to 0.066, the marginal effect of foreign aid between 1971 and 2013 is 1.106. The marginal production effect has been reduced because the marginal effect of aid is also lowered, while capital inside Thailand is relatively static.

4.2.5

Comparison of the Public Capital

We have considered the productivity of foreign aid in this chapter. Next, the productivity of public capital in Thailand is examined as a comparison. We assume the Cobb–Douglas function is the same as in Sect. 4.2.3 and that public capital has direct impact on total factor productivity through technical transfer in the process of making public capital. In addition, previous studies show that the productivity effect is higher in the time-series analysis than in the cross-section analysis. Total factor productivity considers both effects as shown in Eq. (4.5): At ¼ GðPUBLIC K t , SIt Þ

ð4:5Þ

where At shows the total factor productivity, PUBLIC_Kt is the public capital, and SIt presents the industrial construction change in period t, respectively. SIt is shown in Eq. (4.3). The estimation equation is similar to Eq. (4.4), as shown in Eq. (4.6):

4.2 Estimation of the Effect of the Foreign Aid in the Entire Country Table 4.4 Data description about public and private capital

Mean Std Min Max

In (PUBLIC_K ) 14.252 0.943 12.602 15.519

43 In (PRIVATE_K/L ) 12.096 0.461 11.474 12.691

    Yt PRIVATE K t ln ¼ const: þ βSIt þ γ ln ðPUBLIC K t Þ þ α ln Lt Lt þ ut

ð4:6Þ

where PRIVATE_Kt is private capital and PUBLIC_Kt is public capital in period t. Other variables are similar to that in Eq. (4.1), where At is total factor productivity, Kt is capital, Lt is labor, and Yt is gross domestic product in year t, respectively. The description table is shown in Table 4.4, and the results of the unit root tests are I(2). In this chapter, both variables are considered as I(1). Other variables are the same as those shown in Table 4.1. Estimation period is from 1971 to 2013 and 1971 to 2000. In addition, estimation equations are estimated both inclusive and exclusive SIt. Estimation results are shown in Table 4.5, which indicates the following. First, from the estimation period from 1971 to 2013, coefficients of public capital are estimated as sufficiently positive in Eqs. (4.3) and (4.4), which implies that public capital has an impact on economic growth. In addition, the impact on economic growth by public capital seems to be larger than that of foreign aid loans, as the coefficient is larger in public capital. Second, from the estimation results from 1971 to 2000, only Eq. (4.6) is sufficiently positive, and the impact seems larger than the foreign aid loan. Third, SIt, meaning the industrial construction change, is negative in all equations, which is compatible with the theoretical view. In addition, the estimation result was stable by adding SIt.

4.2.6

Summary

We estimate the productivity of foreign aid loans to Thailand using the economic growth model. The following three points are considered. First, foreign aid loans are estimated to contribute to economic growth in Thailand. However, the size of the impact is widely seen as the coefficient is relatively estimated over a wide range. Second, the marginal production effect is wider in the twentieth century than in the whole term partly because the quantity of capital has become larger in recent years. Third, public capital also demonstrates the productivity effect. In addition, this effect is considered greater than that in foreign aid loans because scale merit is easier to facilitate in public capital.

44

4 Foreign Aid Loans and Economic Growth in Thailand

Table 4.5 Productivity of the public capital Dependent variable: ln(Y/L ) (4.1) Estimation period 1971–2013 ln(PUBLIC_K ) 0.102 (0.067) ln(PRIVATE_K/L ) 0.889 (0.123)*** SI Const. AR(1) MA(1) Adjusted R2 D.W. Estimation period ln(PUBLIC_K ) ln(PRIVATE_K/L )

1.079 (1.059) 0.909 (0.088)*** 0.438 (0.194)** 0.994 1.839 (4.5) 1971–2000 0.070 (0.091) 0.892 (0.103)***

SI Const. AR(1) MA(1) Adjusted R2 D.W.

0.718 (1.238)*** 0.924 (0.101)*** 0.632 (0.210)*** 0.990 1.890

(4.2) 1971–2013 0.101 (0.067) 0.892 (0.123)*** 3.430 (3.644) 2.306 (3.704) 0.908 (0.085)*** 0.420 (0.203)** 0.994 1.868 (4.6) 1971–2000 0.285 (0.030)*** 0.384 (0.062)*** 16.391 (4.201)*** 18.794 (4.192)*** 0.430 (0.412) 1.000 (3534.063) 0.986 1.927

(4.3) 1971–2013 0.268 (0.043)*** 0.452 (0.097)***

1.861 (0.613)***

1.000 (2018.188) 0.990 1.751 (4.7) 1971–2000 0.272 (0.055)*** 0.447 (0.117)***

(4.4) 1971–2013 0.284 (0.030)*** 0.421 (0.065)*** 12.415 (3.718)*** 14.390 (3.612)***

1.869 (0.729)**

0.586 (3.395)*** 0.991 1.868 (4.8) 1971–2000 0.278 (0.034)*** 0.400 (0.071)*** 15.865 (3.687)*** 18.174 (3.668)***

1.000 (9834.241) 0.982 1.534

0.425 (0.276) 0.984 1.932

Notes: Standard errors in parentheses. ***: significant at 1%, **: significant at 5%, *: significant at 10%. The error terms in all equations are I(0). That is, these equations are cointegrated. All estimation methods use the fixed effect model

4.3

Estimation of the Effects of the Foreign Aid (Yen Loan) in Regional Panel Data

In the previous section, the productivity effect of the yen loan, or the foreign aid loan from Japan, from 1986 to 2013, is estimated using the panel data under the following data restrictions. First, the gross provincial product is considered only from the

4.3 Estimation of the Effects of the Foreign Aid (Yen Loan) in Regional Panel Data

45

production side and does not reflect the living standard from inside the province. Second, capital stock is not published in each province. Third, the range of foreign aid loans is obscure. Because of these data restrictions, the data on the yen loan and capital in each area and industry are estimated for three industries and four regions. Regardless of these data restrictions, estimating the effect of foreign aid loans is worthwhile, as this is a good example of the development. In this section, the productivity effect is considered by making the panel data in four areas and three industries inside Thailand.

4.3.1

Model

The fundamental methodologies are the same as in Sect. 4.2.3, which are extended into region r and period t. Estimated production function is shown in Eq. (4.7).     Y rt K rt Ln ¼ const: þ α ln þ β1 SI LOANrt þ β2 SI K rt þ β3 SIrt þ Lrt Ltr γ ln ðYEN LOANrt Þ þ ut

ð4:7Þ

where YEN_LOANrt: accumulated yen loan (governmental aid loan from Japan) SI_LOANrt: effects of the yen loan to region r from other regions in period t SI_Krt: effects of capital to region r from other regions in period t SIrt: industrial construction change index The similarity index is divided into three parts: the aid loan similarity effect with respect to other regions, the similarity effect of other investments with respect to other regions, and the effect of industrial change in the whole region. We define the following variables: SI LOAN rt ¼

X4

SI K rt ¼

j6¼r

ωr,j YEN LOANrt , j 6¼ r ð j and r label the regionÞ

X4 j6¼r

ωr,j K rt , j 6¼ r ð j and r label the regionÞ

where ωr, j captures the similarity in each regional industrial structure and is defined as follows: P3

n¼1 Y j,n Y r,n 1=2 P 1=2 3 2 2 n¼1 Y j,n n¼1 Y r,n

ωr,j ¼  P3

46

4 Foreign Aid Loans and Economic Growth in Thailand

It follows that ωr, j is close to 1 if the industrial structure is similar in region r and region j. The variable n ¼ 1, 2, 3, because the three industries are categorized and estimated in this chapter. We suppose that if the industrial structure is similar, the effects of aid loans or investments in other regions will increase. The variable SIrt shows the industrial change in the year and is derived from Eq. (4.3). It is defined as: P3

SIrt ¼ 

n¼1 Y r,ðt5Þ,n Y r,t,n 1=2 P 1=2 3 2 2 Y Y n¼1 r,ðt5Þ,n n¼1 r,t,n

P3

A linear model is adopted for the induced changes in the above model. Because this method may be too complex, the industrial constriction ratio instead of the similar index as in Eq. (4.8) is also estimated. 

       Y rt K rt SECONDrt SECONDrt ln ¼const:þαln þβ1 þβ2 ln ðYENLOANrt Þ Lrt Ltr Y rt Y rt     THIRDrt THIRDrt þβ4 ln ðYEN LOANrt Þþβ5 SIrt þγln ðYEN LOANrt Þþut þβ3 Y rt Y rt ð4:8Þ

4.3.2

Data

In this work, we divide Thailand into three industries and four regions: Northeast, North, South, and Central. As some data are not published in Thailand, we estimate these data. The variable Yrt is derived from the gross regional product in NESDB, and Lrt is the population in each region, from MOL, as the number of employees in the first industry is not readily found. The variable Krt is the capital from NESDB, allocated according to the cultivated farmland in each region in the first industry, and the number of employees in the second and third industries, as published by MOL. The variable YEN_LOANrt is the loan from the Japanese government, as shown by the final evaluation for projects costing at least 200 million yen. Because some facilities, such as railroads and highways, are allocated to several regions, we divide the data into regions. Although not all projects are evaluated, the gap between the total and summing over regions is approximately 8%. We collected data from 1986 to 2013. The parameters for these data are given in Table 4.6.

4.3 Estimation of the Effects of the Foreign Aid (Yen Loan) in Regional Panel Data

47

Table 4.6 Data description for regional panel data Variable ln(Y/L ): GDP per capita ln(YEN_LOAN): Accumulated Yen Loan ln(K/L ): Capital per capita SI_LOAN: Similarity index in yen loan SI_K: Similarity index in capital SI: Industrial similarity index

Obs 112 112 112 112 112 112

Mean 3.532 12.721 4.724 1.357 7.381 0.997

Std 0.772 0.979 0.843 1.277 8.777 0.004

Min 2.227 9.963 3.176 0.060 0.909 0.980

Max 5.124 14.490 6.382 5.048 32.193 1.000

Table 4.7 Estimation results in panel data Dependent variable Estimation equation Way of estimation Estimation period ln(YEN_LOAN) ln(K/L )

ln(Y/L ) (4.1) FE 1986– 2013 0.061 (0.025)* 0.551 (0.080)***

(4.2) FE 1986– 2013 0.048 (0.045) 0.578 (0.127)**

SI_LOAN

(4.3) FE 1986– 2013 0.037 (0.027) 0.431 (0.058)*** 0.112 (0.005)***

SI_K SI Const.

4.3.3

0.158 (0.398)

3.821 (5.398) 3.999 (5.342)

0.881 (0.010)***

(4.4) FE 1986– 2013 0.071 (0.027)* 0.414 (0.058)***

0.018 (0.002)***

(4.5) FE 1986– 2013 0.023 (0.046) 0.442 (0.072)*** 0.153 (0.101) 0.007 (0.018)

0.544 (0.114)**

0.992847 (0.272)**

(4.6) FE 1986– 2013 0.008 (0.064) 0.472 (0.119)** 0.157 (0.094) 0.008 (0.017) 4.334 (4.040) 5.363 (4.246)

Estimation Results

Table 4.7 shows the results of estimating the parameters in Eq. (4.8). As the results of the F tests indicate the fixed effect model, we focus on that model. For the coefficient of YEN_LOANrt, governmental aid loans from Japan are significant only in Eqs. (4.1) and (4.4). Eq. (4.1) only explains the productivity in governmental loans and capital, and Eq. (4.4) explains the productivity in governmental loans, capital, and SI_K. Both comprise a relatively small number of explanatory variables. In addition, the standard error is large compared to the coefficient. In contrast, the coefficients of ln(K/L ) are significant in all equations, which means that per capita capital contributes significantly to productivity. SL_LOAN is significant in (4.3) instead of YEN_LOAN, which means that the loan effect may

48

4 Foreign Aid Loans and Economic Growth in Thailand

Table 4.8 Estimation results in panel data by the industrial ratio Dependent variable Estimation equation Way of estimation Estimation period ln(YEN_LOAN) ln(K/L ) SECOND/Y

ln(Y/L ) (4.1) FE 1986– 2013 0.024 (0.027) 0.526 (0.065)*** 1.565 (0.926)

SECOND/Y * ln(YEN_LOAN)

(4.2) FE 1986– 2013 0.039 (0.029) 0.461 (0.044)*** 5.575 (1.402)** 0.506 (0.067)***

THIRD/Y

(4.3) FE 1986– 2013 0.052 (0.017)* 0.552 (0.064)***

(4.4) FE 1986– 2013 0.524 (0.290)*** 0.527 (0.049)***

(4.5) FE 1986– 2013 0.024 (0.022) 0.529 (0.066)*** 1.396 (0.928)

1.066 (1.176)

10.470 (6.557) 0.862 (0.540) 5.382 (3.638)

0.475 (1.145)

THIRD/Y * ln(YEN_LOAN) Const.

0.311 (0.331)

1.563 (0.278)**

0.814 (0.449)

0.587 (0.646)

(4.6) FE 1986– 2013 0.049 (0.203) 0.450 (0.058)*** 6.013 (1.833)** 0.554 (0.092)*** 0.589 (4.241) 0.008 (0.299)** 1.336 (2.693)

emerge with industries. This variable in other equations is insignificant, which may mean that selecting according to the effect with other indices is difficult. Coefficients in SI are insignificant in both Eqs. (4.2) and (4.6). This may mean that the financial crisis in 1997 was in the estimated period. In summary, we infer that the productivity effect of the yen loan, or the governmental aid loan from Japan, exists sometimes by itself and sometimes together with industries to a certain extent, as seen in Eqs. (4.1), (4.3), and (4.4). The estimation results show that 10% is significantly positive in the yen loan partly because the dataset itself is relatively small. Because a similar index is more complex, the estimation equation is changed to the ratio of the industrial value added. From the estimation results in Table 4.8, we conclude the following. First, the production function is significantly estimated only in Eq. (4.4), which is quite different from K/L. Second, cross terms of the second and third industries are sufficiently estimated, which implies that the productivity effect is available in the production increase mainly in the second industry. This is suitable for the purpose of the yen loan, which strengthens the manufacturing industry.

References

4.3.4

49

Summary

In this section, the impact of the yen loan, or the foreign aid loan from Japan, to the economic growth in Thailand is examined using the economic growth model. During the process, the panel data are formulated because of data restrictions. The estimated results infer that the yen loan has impact on economic growth. However, the magnitude of the impact is difficult to mention as the coefficient of the estimation results is widely estimated.

4.4

Conclusion

This chapter examines the productivity effect of governmental aid loans, the main portion of foreign aid, in Thailand. The main results comprise the following two points. First, we determined the impact of governmental loans on productivity in Thailand to a certain extent using whole-country data between 1971 and 2013. Second, the marginal production effect of governmental loans has recently reduced, although not to zero. Third, the quantitative effect of yen loans from the Japanese government can be determined to a certain extent in the subregional data in Thailand from 1986 to 2013. Overall, although the impact is difficult to judge because the results of parameter estimation in the regional panel data have a wide range, we infer that foreign aid to Thailand has strengthened production.

References Arndt C, Jones S, Tarp F (2015) Assessing foreign aid’s long-run contribution to growth and development. World Dev 69:6–18 Aschauer DA (1989) Is public expenditure productive? J Monet Econ 23(2):177–200 Burnside C, Dollar D (2000) Aid, policies, and growth. Am Econ Rev 90(4):847–868 Easterly W (2006) The white man’s burden: why the West’s efforts to aid the rest have done so much ill and so little good. The Penguin Press, New York Easterly W (2007) Was development assistance a mistake? Am Econ Rev 97(2):328–332 Easterly W, Levine R, Roodman D (2004) Aid, policies and growth: comment. Am Econ Rev 94 (3):774–780 Institute for International Cooperation, Japan International Cooperation Agency (2003) Country study for Japan’s official development assistance to the Kingdom of Thailand Kimura, Todo (2010) Is foreign aid a vanguard of foreign direct investment? A gravity equation approach. World Dev 38(4):482–497

50

4 Foreign Aid Loans and Economic Growth in Thailand

Miyagawa T, Kawasaki K, Edamura K (2013) Reconsideration of the social infra-structure production effect. Keizai Kenkyu 64(3):240–255. (in Japanese) Nowak-Lehmann F, Dreher A, Herzer D, Klasen S, Martinez-Zaroso I (2012) Does foreign aid really raise per capita income? A time series perspective. Can J Econ 45(1):288–313 Selaya P, Thiele R (2010) Aid and sectoral growth: evidence from panel data. J Dev Stud 46 (10):1749–1766 Wolff EN (2002) Computerization and structural change. Rev Income Wealth 48(1):59–75

Chapter 5

Fiscal Effects of Foreign Aid in Thailand

Abstract This chapter examines impacts of foreign aid on domestic borrowing, expenditure, and revenue, in Thailand from 1961 to 2014 by using VAR model and Granger causality tests. Our main findings are as follows. First, a negative relationship is found between foreign aid and domestic borrowing, which is considered to have an impact on Thailand’s fiscal budget. Second, a clear relationship is not necessarily evident about the relationship between foreign aid and governmental expenditure. Third, no relationship is seen between governmental revenue and foreign aid. Fourth, it is difficult to acquire evidence of the impact of foreign aid on fiscal budget if limited to the 1960s and the 1970s. Overall, foreign aid to Thailand has certain impact on its fiscal budget through diminishing domestic borrowing although this result is different if the period is limited to the 1960s and the 1970s. Keywords Foreign aid · Thailand · Fungibility

5.1

Introduction

An element that offsets the impact of foreign aid is changing the recipient country’s usage of their own budget. This problem is called “fungibility,” which is a longestablished term. Fungibility happens when the donor and recipient countries think differently, and use other methods that do not contribute to strengthening the productivity unless the recipient country receives the aid. More generally, Lloyd et al. (2009) summarize that aid for investment (that promotes growth) may be “redirected” to consumption spending (which does not promote growth), and this undermines aid’s effectiveness. Thailand experienced the economic boom during the 1970s, and Thai economy itself is summarized in Suehiro and Higashi (2000) as follows. First oil crisis in 1973 raised agricultural prices such as rice, maize, rubber, and sugar, and the macroeconomic operation went well. During the 1970s, governmental expenditure was increased. However, the second oil crisis happened in 1979 hit Thai economy. Unlike the first oil crisis in 1973, agricultural prices were not raised. Due to the © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 H. Sakurai, Effects of Foreign Aid, New Frontiers in Regional Science: Asian Perspectives 50, https://doi.org/10.1007/978-981-16-2482-7_5

51

52

5 Fiscal Effects of Foreign Aid in Thailand

world recession, inflation, decreased export and current account deficit happened. At the same time, the fiscal deficit was enlarged by the increased expenditure and diminished revenue. As a result, structural adjustment loan was financed by the World Bank in 1982 and 1983, together with the economic reform including the fiscal policy discipline. During the head of 1980s, Thai government made effort for policy and system change including eliminating the fiscal deficit, and the yen appreciation after 1985 boosted Thai economy until 1997 financial crisis. From the historical point of view, the fiscal policy discipline in Thailand seems to have been worked relatively well especially after 1980s. Hence foreign aid to Thailand also seems to facilitate as substitute of the fiscal budget in Thai government. This chapter is organized as follows. Section 5.2 summarizes the literature review about the fungibility. Section 5.3 describes the methodology and empirical result about foreign aid and fiscal response in Thailand in the whole term and from the 1970s to the 1980s. Section 5.4 concludes the study.

5.2

Literature Review

In the 70-year history of foreign aid, the concept of fungibility has existed from the beginning. Nurkse (1953) introduces the concept of fungibility as an apocryphal story between the Economic Cooperation Administration (ECA) and the Austrian government following World War II, “The Austrian government, so the story goes, asked for the release of counterpart funds to reconstruct the Vienna opera. The E.C.A. is said to have replied that this would not be a productive investment and that the release could not be granted for this purpose. Then the Austrian government remembered that it was itself financing the construction of an electric power plant in the mountains. It went back to the E.C.A. and asked for a release of counterpart funds to pay for this piece of construction, to which the E.C.A. agreed. So all that happened was a switch: the wily Austrians, having got the E.C.A. to take over the financing of the power plant, now financed the reconstruction of the opera from their own resources.” (Nurkse 1953, p. 96). World Bank (1998) also implied that the problem of fungibility is difficult to solve. Normally, fungibility occurs when the donor country and the recipient country have differing opinions, and the project includes the problem of the induce effect as mentioned earlier. The recipient country increases government consumption expenditure or reduces governmental investment expenditure. The analysis of fungibility is divided into two parts. The first part is examining each field such as sanitation and education. The second part pertains to analyzing the macroeconomic point of view by measuring the fiscal response such as the domestic borrowing, governmental expenditure, and governmental revenue. This research examines the effect of fungibility from the macroeconomic perspective. The impact of foreign aid on the fiscal budget of the recipient country is divided into two methods: calibrating the economic model and regressing with time-series data. The first is through the “fiscal response model,” maximizing the governmental

5.2 Literature Review

53

utility function under the budget constraint. Franco-Rodriguez et al. (1998) completes this model and analyzes the fiscal response by the foreign aid in Pakistan from 1956 to 1995. The results show that the increase in foreign aid reduces the governmental consumption income nearly two times, increases governmental investment expenditure at 5%, and reduces the total governmental expenditure. This indicates that foreign aid does not necessarily result in fungibility. Although this approach still exists, there are two differing opinions: how we measure the utility function of the government, and that the way for impacts may be more complicated. The second method is examining the relationship between foreign aid and fiscal budget, such as governmental revenue, expenditure, and domestic borrowing by using the vector autoregression (VAR) model. This methodology is further divided into two: estimating ordinary least squares (OLS) as the long-term relationship and using the VAR model as the short-term relationship. Studies have primarily used the VAR model since the impact of foreign aid may be more complicated. For the practical use of the VAR model, Bwire (2012) and McGillivray and Morrissey (2004) summarized the impact of foreign aid with regard to the governmental budget. Since most of the foreign aid goes into the governmental sector, the effect can be seen in the governmental budget. More precisely, the governmental budget is divided into three parts: domestic borrowing, governmental expenditure, and government revenue. First, the inflow of foreign aid leads to the decrease in domestic borrowings, which is evidenced by several studies. Hence, fungibility can be a result if domestic borrowings does not decrease despite the increase in foreign aid. Second, the impact on governmental expenditure will decrease with the increase in foreign aid. If not, this can be considered as fungibility. Since governmental expenditure is decided by the political situation, previous studies do not always clearly indicate the impact. In McGillivray and Morrissey (2004) mentions as follows: “In other words, the absence/presence of aid did not directly alter public expenditure patterns but rather affected government borrowing from the domestic economy” (McGillivray and Morrissey 2004, p. 88). Governmental expenditure can also be divided into governmental consumption and investment expenditure. Under normal circumstances, the governmental consumption expenditure does not decrease if the recipient country has profits. However, governmental investment expenditure is not considered as fungibility since it contributes to capital accumulation. Third, the impact of increased foreign aid on governmental revenue will be considered, which, in theory, is affected by tax reduction and increased productivity. However, many previous studies do not indicate a relationship between foreign aid and governmental revenue. From regional analysis, studies show that fungibility primarily happens in African countries (Aiyar and Ruthbah 2008; Osei et al. 2005; Martins 2010) as shown in Table 5.1. As regards Asian countries, similar analyses were held prior to the 1990s (Franco-Rodriguez et al. 1998; Khan and Hoshino 1992; McGillivray and Ahmed 1999). However, fungibility has not been examined for Asian countries in recent times because the fiscal budget seems to have been used efficiently in recent years. This study analyzes the effect of Thailand’s foreign aid on its governmental budget. The contributions of this study are as follows. First, the efforts in protecting

54

5 Fiscal Effects of Foreign Aid in Thailand

Table 5.1 Impact of foreign aid per 1% GDP increase

Absorption Spending Reserves Investment

Full sample Short-run 0.30*** 0.56*** 0.05 0.14***

Long-run 0.83*** 1.60*** 0.05 0.26***

Africa Short-run 0.41*** 0.79*** 0.01 0.15***

Long-run 1.11*** 2.14*** 0.00 0.26***

Aid dependent Short-run Long-run 0.45*** 1.13*** *** 0.68 1.48*** 0.06 0.00 0.19*** 0.33***

Notes: Standard error in parenthesis, ***: significant at 1%, **: significant at 5%, *: significant at 10%. (Source: Aiyar and Ruthbah 2008; Martins 2010)

fiscal regulation and development of the recipient country are compatible in the example of Thailand. Second, the governance of the recipient country is important for combining both sides, as compared to the situation prior to the 1970s. Third, the results of this study can be applied to emerging developing countries in southeast Asia, such as Myanmar and Vietnam. Given this background, this study examines fungibility in Thailand from 1961 to 2014 using the VAR model and the Granger causality tests.

5.3 5.3.1

Methodology and Estimation Results Data

We use annual fiscal data for Thailand since 1977 in “Fiscal Budget in Thailand” by the Ministry of Finance, Royal Thai Government. In addition, fiscal data is also available in the “Monthly Bulletin” since 1961 by the Bank of Thailand. Although data from the Bank of Thailand covers a longer period, it has been changed in its definition after the 1997 crisis. In contrast, data from the Ministry of Finance holds the same definition although it covers a shorter period. In this study, from the Bank of Thailand, we use data prior to 1998, and from the Ministry of Finance, we use data from 1998 onwards. Aid data has been adopted from Development Assistance Committee and Organisation for Economic Co-operation and Development, which includes loan, grant, and technical assistance in the net base. Since this data is in US dollars, it has been converted into Thai baht using International Financial Statistics in International Monetary Fund. Finally, all statistics have been adjusted by Consumer Price Index (2010 price). Data description is shown in Table 5.2. All variables are I(1) from the result of the unit root test.

5.3 Methodology and Estimation Results

55

Table 5.2 Data description (billion Thai Baht, 2010 prices) Title Domestic borrowing Governmental revenue Governmental expenditure Governmental consumption expenditure Governmental investment expenditure Net official development assistance

5.3.2

Name DB R GD

Number 54 54 54

Mean 61.725 738.216 791.436

Std 104.258 589.899 651.327

Maximum 455.493 1978.129 2267.706

Minimum 99.054 68.260 70.483

GDC

54

602.166

508.125

1859.472

56.034

GDK

54

185.178

153.567

516.786

14.449

NETODA

54

12.387

17.013

49.669

49.819

Total Governmental Expenditure

We will first estimate governmental expenditure as one variable. We will assume that domestic borrowing is regressed by foreign aid, governmental revenue, and governmental expenditure. First, we check whether or not there is cointegration. We will then assume that the domestic borrowing is regressed by foreign aid, governmental revenue, and governmental expenditure through OLS, and the residual is tested by the Engle–Granger test. The residual is I(0) and is estimated in Eq. (5.1). DBt ¼ 0:712NETODAt  0:557Rt þ 0:588GDt þ 16606:55 þ ut ð0:392Þ*

ð0:054Þ*** ð0:050Þ*** ð12:075Þ

ð5:1Þ

Adj: R ¼ 0:813 D:W: 1:507 2

Note: DBt: Domestic Borrowing, NETODAt: Net ODA, Rt: Governmental Revenue, GDt: Governmental Expenditure, in period t. Standard error in parentheses. ***: significance at 1%, **: significance at 5%, *: significance at 10%. From Eq. (5.1), Durbin–Watson ratio is relatively low, but foreign aid reduces domestic borrowing at the 10% significance level. In addition, both governmental revenue and governmental expenditure are significant at 1%, and the sign conditions are the same. From this equation, the long-term relationship is not necessarily seen among foreign aid and fiscal budget.1 Next, the VAR model is estimated in Eq. (5.2), and the cointegration is estimated by the level series. The VAR model is used to estimate domestic borrowing (DB), net official development assistance (NETODA), governmental revenue (R), and governmental expenditure (GD) at the previous period (t  1). αi (i ¼ 1, 2, 3, 4) is the constant term, βi, γ i, δi, κi are the endogenous terms, and uit is the error term.

1

Cointegrated VAR model (CVAR model) is insufficiently estimated.

56

5 Fiscal Effects of Foreign Aid in Thailand

Table 5.3 VAR model DB(1) NETODA(1) R(1) GD(1) C Adj. R2

DB 0.271 (0.204) 1.261 (0.561)** 0.244 (0.142)* 0.284 (0.145)* 20.680 (17.153) 0.646

NETODA 0.056 (0.038) 0.609 (0.105)*** 0.008 (0.027) 0.009 (0.027) 7.032 (3.206)** 0.537

R 0.503 (0.236)** 0.179 (0.651) 1.263 (0.164)*** 0.269 (0.168) 18.678 (19.891) 0.985

GD 0.531 (0.233)** 1.574 (0.642)*** 0.852 (0.162)*** 0.196 (0.166) 33.007 (19.613)* 0.988

Note: DB: domestic borrowing, NETODA: net ODA, R: governmental revenue, GD: governmental expenditure in the period t. (1) denotes the previous period Standard deviation in parentheses. ***: significance at 1%, **: significance at 5%, *: significance at 10%

2

3 2 3 2 α1 β1 DBt 6 NETODA 7 6 α 7 6 β t7 6 6 27 6 2 6 7¼6 7þ6 4 5 4 α3 5 4 β 3 Rt GDt

α4

β4

γ1 γ2

δ1 δ2

γ3

δ3

32 3 2 3 DBt1 u1t κ1 7 6 7 6 κ2 76 NETODAt1 7 6 u2t 7 7 76 7 þ 6 7 ð5:2Þ 5 4 5 4 κ3 T t1 u3t 5

γ4

δ4

κ4

GDt1

u4t

The VAR model estimates that each variable in the previous period influences the ones in the present period. In this chapter, we primarily examine the significance, sign condition, and coefficient of NETODA(1), previous period of the net ODA, compared to other variables, DB, R, and GD, present period of domestic borrowing, governmental revenue, and governmental expenditure, since the aim is to decipher the fiscal response of foreign aid. The estimated results from the VAR model are shown in Table 5.3. Examining the impact of foreign aid in the previous period, NETODA(1), on DB, R, and GD, it can be seen that the relationship between foreign aid in the previous period and domestic borrowing in the present period is negatively estimated at the 1% significance level, and the relationship between foreign aid in the previous period and governmental expenditure meets with the sign condition at 1%. This result indicates that increased foreign aid decreases domestic borrowing and governmental expenditure, which is ideal for the substitution of governmental budget. Next, the Granger causality test is estimated. We also see causality from foreign aid (NETODA) in relation to other variables (DB, R, and GD) since the aim of this study is to examine the effect of foreign aid. Table 5.4 shows that the Granger causality for foreign aid to domestic borrowing is significant at the 5% level. In addition, the effect of foreign aid on domestic borrowing is inferred in the negative since the sign condition is minus in the VAR

5.3 Methodology and Estimation Results

57

Table 5.4 Granger causality tests Null hypotheses NETODA ! DB DB ! NETODA R ! DB DB ! R GD ! DB DB ! GD

Obs. 53 53 53 53 53 53

F-statistics 4.796** 6.204** 3.666* 2.328 4.512** 6.160**

Null hypotheses R ! NETODA NETODA ! R GD ! NETODA NETODA ! GD GD ! R R ! GD

obs. 53 53 53 53 53 53

F-statistics 0.491 0.136 1.117 1.351 0.345 24.849***

Note: DB: domestic borrowing, NETODA: net ODA, R: governmental revenue, GD: governmental expenditure Standard deviation in parentheses. ***: significance at 1%, **: significance at 5%, *: significance at 10%

model in Table 5.3. In contrast, governmental expenditure and revenue by foreign aid do not have causality in terms of the Granger causality test. These results show that increased foreign aid affects domestic borrowing relatively in the short term, but not governmental revenue or expenditure, which is consistent with results of previous studies. In addition, these results are also adequate to the effort of protecting the fiscal policy discipline in the Thai government.

5.3.3

Categorizing Consumption Expenditure and Capital Expenditure

The impact of foreign aid on fiscal budget, specifically governmental expenditure, is divided into governmental consumption and capital expenditure. First, we check if the variables have cointegration. OLS is estimated in Eq. (5.3), and the error term is tested by the Engle–Granger test, which is I(0). DBt ¼ 0:354NETODAt  0:524Rt þ 0:661GDCt þ 0:232GDKt þ 11:951 þ ut ð0:333Þ

ð0:047Þ*** ð0:044Þ*** ð0:087Þ***ð10:022Þ Adj: R2 ¼ 0:872 D:W:1: 453 ð5:3Þ

Note: DBt: Domestic Borrowing, NETODAt: Net ODA, Rt: Governmental Revenue, GDCt: Governmental Consumption Expenditure, GDKt: Governmental Capital Expenditure, GDt: Governmental Expenditure, in period t. Standard error in parentheses. ***: significance at 1%, **: significance at 5%, *: significance at 10%. From Eq. (5.3), it can be seen that foreign aid is insignificant and is not satisfied with the sign condition. Other explanatory variables, that is, governmental revenue, governmental consumption expenditure, and governmental capital expenditure, are significant and satisfied with the sign condition.

58

5 Fiscal Effects of Foreign Aid in Thailand

Next, the VAR model as shown in Eq. (5.4) is estimated similar to Eq. (5.2). The only difference is that governmental expenditure is divided into governmental consumption expenditure (GDC) and governmental capital expenditure (GDK). In addition, the Granger causality tests were applied. We can also see the correlation of foreign aid (NETODA) to other variables. 2

DBt

3

2

α1t

3

2

β1

7 6 7 6 6 6 NETODAt 7 6 α2t 7 6 β2 7 6 7 6 6 7 ¼ 6 α3t 7 þ 6 β3 6 Rt 7 6 7 6 6 7 6 7 6 6 4 GDCt 5 4 α4t 5 4 β4 GDKt α5t β4 2 3 u1t 6 7 6 u2t 7 6 7 7 þ6 6 u3t 7 6 7 4 u4t 5 u5t

γ1

δ1

κ1

γ2 γ3

δ2 δ3

κ2 κ3

γ4

δ4

κ4

γ5

δ5

κ5

θ1

32

DBt1

3

76 7 θ2 76 NETODAt1 7 76 7 7 6 θ3 7 Rt1 76 7 76 7 θ4 54 GDCt1 5 θ5 GDKt1

ð5:4Þ

The estimated results of the VAR model and the Granger causality tests are shown in Table 5.5. The results of the VAR model show that the foreign aid in the previous period, NETODA(1), is negatively estimated in the DB, GDC, and GDK at 10%. In contrast, there is no effect on governmental revenue. These results are consistent with the results in governmental consumption as a whole. Next, the Granger causality tests show that only domestic borrowing is significantly estimated. This also estimates the impact of foreign aid from the previous period to the present period in each governmental budget variable. These results indicate that the increase in foreign aid will cause a decrease in domestic borrowing in the short term. In contrast, no causality was found in governmental revenue, governmental consumption expenditure, and governmental capital expenditure. These results are consistent with that of previous studies, where the results connect to the effort of reducing domestic borrowing, but it is difficult to see effects in governmental expenditure due to political conflict. These results are also suitable for the fiscal policy discipline in Thailand.

5.3.4

Impact of Foreign Aid in the 1960s and the 1970s

Although corruption has not been heard of in Thailand of late, the country seems to have experienced it in the 1960s and the 1970s. While foreign aid has been facilitated efficiently for the fiscal budget in Thailand during the whole period, it is still unclear if the different trend may be included before the financial crisis and the introduction of foreign capital in the 1980s. Therefore, this section examines the

5.3 Methodology and Estimation Results

59

Table 5.5 VAR model and Granger causality tests (governmental expenditure divided) (a) VAR model DB 0.018 (0.287) NETODA(1) 1.154 (0.565)* R(1) 0.381 (0.182)** GDC(1) 0.502 (0.223)** GDK(1) 0.223 (0.158) C 19.072 (17.072) Adj. R2 0.648 (b) Granger causality Null hypotheses Obs. NETODA ! DB 53 DB ! NETODA 53 R ! DB 53 DB ! R 53 GDC ! DB 53 DB ! GDC 53 GDK ! DB 53 DB ! GDK 53 R ! NETODA 53 NETODA ! R 53 DB(1)

NETODA 0.061 (0.054) 0.613 (0.107)*** 0.011 (0.034) 0.014 (0.042) 0.005 (0.030) 6.978 (3.237)** 0.527

R 0.091 (0.320) 0.372 (0.630) 1.080 (0.203)*** 0.043 (0.249) 0.428 (0.176)** 19.621 (19.030) 0.986

GDC 0.283 (0.186) 0.643 (0.367)* 0.002 (0.118) 1.127 (0.145)*** 0.175 (0.103)* 17.847 (11.099) 0.994

GDK 0.023 (0.152) 0.574 (0.299)* 0.391 (0.096)*** 0.296 (0.118)*** 0.380 (0.084)*** 16.143 (9.052)* 0.954

F-statistics 4.796** 6.204** 3.666* 2.328 4.617** 4.130** 3.089* 5.158** 0.491 0.136

Null hypotheses GDC ! NETODA NETODA ! GDC GDK ! NETODA NETODA ! GDK GDC ! R R ! GDC GDK ! R R ! GDK GDK ! GDC GDC ! GDK

Obs. 53 53 53 53 53 53 53 53 53 53

F-statistics 1.408 1.529 0.286 0.613 3.790* 1.379 9.335*** 21.445*** 1.015 2.902*

Note DB: domestic borrowing, NETODA: net ODA, R: governmental revenue, GD: governmental expenditure in the period t. (1) denotes the previous period Standard deviation in parentheses. ***: significance at 1%, **: significance at 5%, *: significance at 10%

effect of foreign aid from 1961 to 1979 using the VAR model and Granger causality tests. Since the estimation results do not depend on governmental expenditure divided into governmental consumption and capital, this is examined at a time as comparison. First, estimated equations are shown in (5.5) and (5.6). Both of error terms are I(0), cointegrated. Coefficients of the net ODA are insignificant while other variables are significant. These results indicate that foreign aid at that time was barely useful in reducing domestic borrowing in the long term. DBt ¼ 0:335NETODAt  0:678Rt þ 0:274GDt  4:011 þ ut ð0:850Þ

ð0:164Þ*** ð0:128Þ*** ð6:952Þ

Adj: R2 ¼ 0:776 D:W: 2:295

ð5:5Þ

60

5 Fiscal Effects of Foreign Aid in Thailand

Table 5.6 VAR model in the 1960s and the 1970s (a) governmental expenditure unified DB NETODA DB(1) 0.387 0.107 (0.386) (0.057)* NETODA(1) 1.426 1.086 (2.031) (0.301)*** R(1) 0.628 0.137 (0.375) (0.055)** GD(1) 0.738 0.111 (0.351)* (0.052)** C 10.692 3.236 (16.603) (2.458) Adj. R2 0.496 0.610 (b) governmental expenditure divided DB NETODA DB(1) 0.403 0.109 (0.401) (0.059)* NETOD(1) 0.724 1.189 (2.796) (0.414)*** R(1) 0.502 0.118 (0.509) (0.075) GDC(1) 0.547 0.083 (0.620) (0.092) GDK(1) 0.973515 0.145927 (0.719) (0.107) C 9.393 3.427 (17.514) (2.594) Adj. R2 0.460 0.583

R 0.043 (0.190) 1.589 (0.999) 0.820 (0.184)*** 0.224 (0.173) 11.859 (8.163) 0.986

GD 0.319 (0.410) 0.178 (2.157) 0.357 (0.398) 0.831 (0.373)** 2.441 (17.633) 0.958

R 0.049 (0.198) 1.333 (1.378) 0.774 (0.251)*** 0.293 (0.306) 0.137618 (0.355) 11.385 (8.635) 0.985

GDC -–0.208 (0.209) 1.178 (1.459) 0.433 (0.266) 0.537 (0.324) 0.543009 (0.375) 3.265 (9.139) 0.982

GDK 0.155 (0.253) 0.606 (1.762) 0.276 (0.321) 0.240 (0.391) 0.948072 (0.453)* 4.460 (11.039) 0.744

Note: DB: domestic borrowing, NETODA: net ODA, R: governmental revenue, GDC: governmental consumption expenditure, GDK: governmental capital expenditure, GD: governmental expenditure in the period t. (1) denotes the previous period Standard deviation in parentheses. ***: significance at 1%, **: significance at 5%, *: significance at 10%

DBt ¼ 0:336NETODAt  0:680Rt þ 0:726GDCt  0:722GDKt  3:982 þ ut ð0:904Þ

ð0:279Þ** ð0:330Þ** ð0:299Þ** ð8:339Þ Adj: R2 ¼ 0:760 D:W: 2:295 ð5:6Þ

Note: DBt:Domestic Borrowing, NETODAt: net ODA, Rt: Governmental Revenue, GDCt: Governmental Consumption Expenditure, GDKt: Governmental Capital Expenditure, GDt: Governmental Expenditure, in period t.

5.3 Methodology and Estimation Results

61

Table 5.7 Granger causality tests in the 1960s and the 1970s (a) governmental expenditure unified Null hypotheses Obs. F-statistics NETODA ! DB 18 1.577 DB ! NETODA 18 1.421 R ! DB 18 2.996 DB ! R 18 3.689* GD ! DB 18 5.076** DB ! GD 18 4.725** (b) governmental expenditure divided NETODA ! DB 18 1.577 DB ! NETODA 18 1.421 R ! DB 18 2.996 DB ! R 18 3.689* GDC ! DB 18 3.770* DB ! GDC 18 2.383 GDK ! DB 18 8.920*** DB ! GDK 18 2.152 R ! NETODA 18 4.036 NETODA ! R 18 0.771

Null hypotheses R ! NETODA NETODA ! R GD ! NETODA NETODA ! GD GD ! R R ! GD

Obs. 18 18 18 18 18 18

F-statistics 4.036* 0.771 2.579 0.024 4.906** 5.001**

GDC ! NETODA NETODA ! GDC GDK ! NETODA NETODA ! GDK GDC ! R R ! GDC GDK ! R R ! GDK GDK ! GDC GDC ! GDK

18 18 18 18 18 18 18 18 18 18

3.089* 0.094 1.076 0.004 7.102** 5.089** 1.872 2.357 0.003 1.216

Note: DB: domestic borrowing, NETODA: net ODA, R: governmental revenue, GDC: governmental consumption expenditure, GDK: governmental capital expenditure, GD: governmental expenditure. Standard error in parentheses. ***: significance at 1%, **: significance at 5%, *: significance at 10%.

Standard error in parentheses. ***: significance at 1%, **: significance at 5%, *: significance at 10%. Since it includes cointegration, the VAR model applied in the level series. The estimation results are shown in Table 5.6. The column of net ODA shows that in the previous period with regard to domestic borrowing, governmental revenue, and governmental expenditure, no variables are significantly estimated. This result does not change even if governmental expenditure is divided into governmental consumption and capital expenditure. Next, the Granger causality tests are shown in Table 5.7. Observing foreign aid to other variables, no variables are significantly estimated. This result does not change even if governmental expenditure is divided into governmental consumption expenditure and governmental capital expenditure. From these results, we infer that foreign aid did not have an impact on fiscal budget even with domestic borrowing in Thailand during the 1960s and the 1970s.

62

5.4

5 Fiscal Effects of Foreign Aid in Thailand

Conclusion

This chapter examined the impact of foreign aid on fiscal budget, especially with regard to domestic borrowing, governmental expenditure, and governmental revenue in Thailand by using the VAR model and Granger causality tests. Previous studies indicate that foreign aid impacts domestic borrowing while it barely has an influence on governmental revenue. Although the influence of foreign aid on governmental expenditure is often seen, it is difficult to see an impact on domestic borrowing partly because of political influence. The results are the same in the case of Thailand. As for the relationship between foreign aid and domestic borrowing, a negative relationship is found. This result implies that foreign aid acts as a substitution, and is considered to have an impact on Thailand’s fiscal budget. This result is also suitable for the stance of the government of Thailand, which has a severe fiscal deficit. With regard to the effect of foreign aid on governmental revenue, no relationship is seen. Most previous studies also show no relationship, and it is also difficult to show a relationship from a theoretical perspective. Therefore, a clear relationship is not necessarily evident about the relationship between foreign aid and governmental expenditure. This is partly because governmental expenditure is easily influenced by politicians compared to domestic borrowing. In Thailand’s case, for the whole period, that is, the 1960s and the 1970s, it is difficult to acquire evidence of the impact of foreign aid on fiscal budget. This indicates that introducing severe fiscal discipline and implementing policies regarding foreign capital after the 1980s in the government of Thailand can change the impact of foreign aid. Overall, foreign aid to Thailand has certain impact on its fiscal budget through diminishing domestic borrowing. However, this result is different if the period is limited to the 1960s and the 1970s partly because the effort of the fiscal consolidation from the 1980s in Thai government facilitates well.

References Aiyar S, Ruthbah U (2008) Where did all the aid go? An empirical analysis of absorption and spending. IMF working paper WP/08/34 Bwire T (2012) Aid, fiscal policy and macroeconomy of Uganda: a cointegrated vector autoregressive (CVAR) approach. Dissertation, University of Nottingham Franco-Rodriguez S, Morrissey O, McGillivray M (1998) Aid and the public sector in Pakistan: evidence with endogenous aid. World Dev 26(7):1241–1250 Khan HA, Hoshino E (1992) Impact of foreign aid on the fiscal behavior of LDC Governments. World Dev 20(10):1481–1488 Lloyd T, McGillivray M, Morrissey O, Opoku-Afari M (2009) The fiscal effects of aid in developing countries: a comparative dynamic analysis. In: Mavrotas G, McGillivray M (eds) Development aid: a fresh look. Palgrave Macmillan UNU-WIDER Studies, Basingstoke, pp 158–179

References

63

Martins MG (2010) Fiscal dynamics in Ethiopia: the cointegrated VAR model with quarterly data. CREDIT research paper, 10/05 McGillivray M, Ahmed A (1999) Aid, adjustment and public sector fiscal behaviour in the Philippines. J Asia-Pacific Econ 4(2):381–391 McGillivray M, Morrissey O (2004) Fiscal effect of aid. In: Addison T, Roe A (eds) Fiscal policy for development: Poverty, Reconstruction and Growth. Palgrave Macmillan for UNU-WIDER, New York, pp. 72–96 Nurkse R (1953) Problems of capital formation in underdeveloped countries. Basil Blackwell, Oxford Osei R, Morrissey O, Lloyd T (2005) The fiscal effects of aid in Ghana. J Int Dev 17(8):1037–1053 Suehiro A, Higashi S (2000) New trends in Thai studies and the assessment of economic policy. In: Suehiro A, Higashi S (eds) Economic policy in Thailand: the role of institutions and actors. Chiba, Institute of Developing Economies, pp 3–57. (in Japanese) World Bank (1998) Assessing aid –what works, what doesn’t, and why. World Bank, Washington, DC

Chapter 6

Relationships Between Aid Agencies

Abstract This chapter examines the relationships between three primary aid agencies, Japan, the World Bank, and the Asian Development Bank, by scrutinizing the amount of aid they granted to Thailand from 1986 to 2015. The key findings are as follows. First, the aid amounts revealed that the World Bank was more involved after the financial crisis, while Japan was more involved after the big flood in 2011, implying that the World Bank has more interest in financial institutions and Japan has more interest in protecting industrial areas. Second, an OLS of the aid from each agency illustrates the long-term relationship between these three agencies, inferring that Japan looks to act relatively independently, the World Bank and the Asian Development Bank look to act mutually independently, and the Asian Development Bank has the ability to act mutually. Third, after examining their short-term relationships with a VAR model, there was almost no relationship between the three aid agencies except for Japan, which is slightly delayed for the Asian Development Bank. Keywords Foreign aid · Thailand · Aid agency

6.1

Introduction

Aid coordination distributes limited aid resources to the recipient country efficiently, and without duplications. However, not all projects are necessarily coordinated. For example, the Eastern Seaboard project began with opposition from the World Bank during the first half of the 1980s. In contrast, the aid agencies coordinated when the big flood occurred in 2011. Because it is difficult to grasp the characters from a concrete example, it is better to examine the amount of Direct Government Loans from the main aid agencies, Japan, the World Bank (WB), and the Asian Development Bank (ADB), listed in the “Fiscal Budget in Brief” from the Thai government’s Ministry of Finance (MOF). Figure 6.1 provides the data from each agency, and Fig. 6.2 displays the combined annual totals from the three aid agencies. Figures 6.1 and 6.2 highlight the following three points. First, government loans increased during the 1980 and 1997 currency crises in Thailand. During these crises, © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 H. Sakurai, Effects of Foreign Aid, New Frontiers in Regional Science: Asian Perspectives 50, https://doi.org/10.1007/978-981-16-2482-7_6

65

66

6 Relationships Between Aid Agencies

Fig. 6.1 Governmental aid loans from the three primary donors. (Source: MOF, real base in 2010)

Fig. 6.2 Cumulative annual governmental aid loans from the three primary donors. (Source: MOF, real base in 2010)

the WB’s lending increased greatly. Second, the Eastern Seaboard was constructed during the 1980s, and government loans from Japan increased after the large flood in 2011. Third, other aid agencies increased their aid soon after a leader increased its aid first. In summary, the WB and Japan, two primary aid agencies, looked to specialize in, and coexist with, fields they were strong in: finance and manufacturing.

6.2 Methodology and Data

67

These assumptions will be used to examine the relationships between the aid agencies.

6.2 6.2.1

Methodology and Data Introduction

Although there have been many prior studies on the impact of foreign aid, few studies have explained the relationships between these types of aid agencies. Therefore, a better study will consider the general concept from Matsumura (2004). It often happens that others define the optimal solution for an entity. For example, a firm must consider the location of other firms when choosing the location of its factory. Similarly, in the case of global environmental problems, a country’s policy must consider the stance of other countries. This type of relationship is known as “Strategic Interdependence,” when the stance of other entities determines the optimal solution for a given entity. In addition, “Strategic Interdependence” relationships subdivide into “Strategic Substitutes” and “Strategic Complements” relationships. “Strategic Substitutes” means that supply decreases due to other entities’ actions. In contrast, “Strategic Complements” means that supply increases due to the action of other entities. In this context, these two relationships are translated as follows: if a company moving into a developing country causes a tight labor supply or competitive market, that is a “Strategic Substitutes” relationship. In contrast, if a company moving into a developing country contributes to an increase in the number of related companies or skilled laborers, that is a “Strategic Complements” relationship that increases the incentives for other companies to move into that country. Since each aid agency supplies public goods, other entities can also consume the aid they supply. Hence, aid agencies have “Strategic Interdependence” relationships. This fact can be verified using the simple model, and examined using data.

6.2.2

Simple Model

The contents of aid should consist of goods from the manufacturing industry, g1, and financial support, g2. In addition, the aid agency 1 supplies g1, and the aid agency 2 supplies g2. Aid agencies pay a cost of ci (i ¼ 1, 2) for the interest or the investigation. Under this condition, each aid agency maximizes its profit, consisting of the utilities W1 and W2, minus its cost, as shown in Eq. (6.1).

68

6 Relationships Between Aid Agencies



 max π j ðg1 , g2 Þ ¼ W j g j , gi  c j g j fg j g

ð6:1Þ

Suppose W1 and W2 are increasing functions of g1 and g2, and are diminishing the marginal benefit. In this case, the reaction function maximizes profit under the condition that another supply is given. The first condition of utility maximization is that the marginal benefit and marginal cost are equal. ∂π 1 ∂W 1 ∂c1 ðg1 Þ ∂π 2 ∂W 2 ∂c2 ðg2 Þ ¼  ¼ 0, ¼  ¼0 ∂g1 ∂g2 ∂g1 ∂g1 ∂g2 ∂g2

ð6:2Þ

Taking the total deviation of Eq. (6.2) for the agency 1 as an example: f 1 ð g1 , g2 Þ 

∂π 1 ∂W 1 ∂c1 ðg1 Þ ¼  ∂g1 ∂g1 ∂g1

ð6:3Þ

2

∂ W1

∂ f =∂g dg1 ∂g ∂g ¼  1 2 ¼  2 11 22 1 ∂ W ∂ c dg2 ∂ f =∂g 2  2 1

1

Since

2

∂ W1 ∂g21

< 0,

2

∂ C1 ∂C 21

∂g1

ð6:4Þ

∂c1

> 0 from the assumption, the denominator of Eq. (6.4)

depends A “Strategic Substitutes” relationship exists when  1 on the numerator.  2 ∂ ∂W ∂ ∂W < 0 , ∂g ∂g < 0 , meaning that g1 will decrease in response to g2 ∂g1 ∂g2 2 1  1 increasing, and a “Strategic Complements” relationship exists when ∂g∂ ∂W >0 ∂g2 1  2 ∂ ∂W , ∂g ∂g < 0, meaning that g1 will increase in response to an increase in g2. The 2

1

two possibilities can therefore be considered from a practical perspective. For instance, if a large flood causes an increased demand for aid, g1, to stop water from leaking or draining from an industrial area, g2 will also increase. In contrast, if manufacturing industries decline due to the large flood, g2 will also decrease. From this simple model, strategic substitution appears to work well. The next section examines the relationships between aid agencies that increase or decrease aid from other agencies in models with three or five entities.

6.2.3

Data and Methodology

This dataset comes from the “Direct Government Loan” and “Fiscal Budget in Brief” reports from the MOF. The data consists of five facilities: the WB, Japan, the ADB, the Private Financial Institute (PFI), and others. The price level was adjusted in 2010 using the Consumer Price Index from the WB’s World Development Indicators

6.3 Estimation Results

69

Table 6.1 Data description (billions of 2010 Baht)

N Mean Std Maximum Minimum

World Bank (WB) 38 7.829 11.752 59.033 0

JAPAN 38 6.894 10.173 58.267 0

Asian Development Bank (ADB) 38 3.700 7.753 43.728 0

Private Financial Institute (PFI) 38 1.212 2.427 8.183 0

OTHERS 38 3.249 5.899 23.184 0

normalized. Table 6.1 provides a description of the data, showing that the standard deviation is larger for the WB and Japan than for the ADB. This chapter presents loans by gross base, excluding the repayment; since loan repayment begins after the project is complete, it is better to promote development by measuring the gross base. The analysis in this chapter used the VAR model and Granger causality tests. Although the Cointegrated VAR model is also good, it cannot be used this time. Therefore, OLS shows the long-term relationship, while the VAR model shows the short-term relationship.

6.3

Estimation Results

First, unit root tests were examined. The results were I(0) for four variables: the WB, Japan, ADB, and Others. In contrast, the PFI was I(1), partly because the trend for private finance was increasing. Therefore, the estimation for the three primary facilities was in a level series, and adding PFI and Others was in the first difference. First, the long-term relationship between the three primary aid agencies was estimated using OLS: WBt ¼ 3:604  0:026JAPANt þ 1:191ADBt þ ut ð1:537Þ** ð0:135Þ ð0:177Þ***

ð6:5Þ

R2 ¼ 0:580 D:W:1:856 JAPANt ¼ 5:049  0:041WBt þ 0:585ADBt þ ut ð1:887Þ** ð0:212Þ

ð0:321Þ*

R ¼ 0:121 D:W: 2:284 2

ð6:6Þ

70

6 Relationships Between Aid Agencies

ADBt ¼ 1:032 þ 0:474WBt þ 0:148JAPANt þ ut ð1:028Þ ð0:070Þ*** ð0:081Þ*

ð6:7Þ

Rt ¼ 0:616 D:W:2:145 Notes: Standard deviations are in parentheses. *** Significance at 1%, ** significance at 5%, * significance at 10%. Equation (6.5) shows whether the aid loan amount from Japan and the ADB explained the amount of aid loans from the WB. Similarly, Eq. (6.6) shows whether the loans from the WB and ADB explained the amount of aid loans from Japan, while Eq. (6.7) does the same for loans from WB and Japan explaining the amount of aid loans from the ADB. The three estimated equations are as follows. First, Eq. (6.6) was significantly effective, with the ADB and WB significant at the 1% level. In contrast, Eq. (6.5) from ADB and Japan was only estimated to be significant at the 10% level. As for Eq. (6.7) with the WB and Japan, the coefficient was estimated to be statistically insignificant, although its constant term was significant. Second, the sign conditions showed that each aid agency moved in the same direction, because all the estimated variables were positive. In addition, the coefficients of explanatory variables were basically less than 1, meaning that the equations converged. The following elements were assumed about the aid amount from each agency. First, Japan had smaller impacts from agencies other than the WB and the ADB, which implies that Japan was near the donor’s budget. Equation (6.6) reflects the fact that its constant term was significant at the 5% level, whereas it was insignificant for the WB and significant at the 10% level for the ADB. Next, the ADB was relatively more influenced by the other two agencies, since the constant term in Eq. (6.7) was insignificant, whereas the WB and Japan coefficients were significant at the 1% and 10% level, respectively. Finally, the WB had a relatively deep relationship with the ADB, but was relatively independent. Considering the relationships between the aid agencies as a whole, the WB and the ADB were estimated to be effective at the 1% level of significance and were thought to have a deep relationship. In addition, the ADB and Japan had some relationship, since the coefficient was estimated to be significant at the 10% level. In contrast, the WB and Japan did not necessarily have a relationship, since their term was estimated to be insignificant, meaning that there was almost no relationship. Now, consider the short-term effects between aid agencies by examining the VAR model shown in Eq. (6.8). First, the VAR model examined the relationship between the WB, Japan, and the ADB, and then the Granger causality tests were used. The constant terms were αit (i ¼ 1, 2, 3), βi, γ i, and δi were endogenous terms, and uit were error terms.

6.3 Estimation Results

71

Table 6.2 VAR model for the three primary aid agencies

WB 0.079 (0.248) 0.059 (0.199) 0.533 (0.393) 4.640 (2.447)* 0.119

WB(1) JAPAN(1) ADB(1) C Adj. R2

JAPAN 0.016 (0.145) 0.173 (0.116) 1.142 (0.229)*** 3.889 (1.426)** 0.602

ADB 0.023 (0.171) 0.030 (0.137) 0.328 (0.271) 2.567 (1.687) 0.037

Notes: Standard errors are in parentheses. *** Significant at 1%, ** significant at 5%, * significant at 10% Table 6.3 Granger causality tests for the three primary aid agencies

Null hypotheses JAPAN ! WB WB ! JAPAN ADB ! WB WB ! ADB ADB ! JAPAN JAPAN ! ADB

Obs. 37 37 37 37 37 37

F value 0.518 17.416*** 2.350 0.021 56.147*** 0.050

Notes: Standard errors are in parentheses. *** Significant at 1%, ** significant at 5%, * significant at 10%

2

JAPANt

3

2

α1t

3

2

β1

6 7 6 7 6 4 WBt 5 ¼ 4 α2t 5 þ 4 β2 ADBt α3t β3

γ1 γ2 γ3

δ1

32

JAPANt1

3

2

u1t

3

76 7 6 7 δ2 54 WBt1 5 þ 4 u2t 5 δ3 ADBt1 u3t

ð6:8Þ

Examined VAR model is shown in Table 6.2. Only ADB in previous period is estimated significantly positive to Japan, and other results are insignificant. Table 6.3 displays the results of the Granger causality tests. The Granger causality was significantly effective in two ways: from the WB to Japan, and from the ADB to Japan, which indicates that Japan is a follower. Figure 6.3 illustrates the impulse response from Japan when the amount of aid loans from the WB was increased by one standard deviation. Next, Fig. 6.4 illustrates the impulse response from Japan when the amount of aid loans from the ADB was increased by one standard deviation. Figure 6.3 shows that the impact of increased loans by the WB on Japan was almost inexistent. In addition, the 95% probability range included both positive and negative numbers. Although the Granger causality test revealed a negative impact, the result from the impulse response curve was obscure. In contrast, Fig. 6.4 shows the impact of the ADB on Japan, indicating a steady increase. Both figures reveal that Japan follows the WB and ADB, especially the ADB, but it was not clear. Next, we added the PFI and Others to the WB, ADB, and Japan. First, Eq. (6.9) provides the VAR model, and then the Granger causality tests are provided. The

72

6 Relationships Between Aid Agencies

Fig. 6.3 Impulse response reaction function from Japan to an increase by the WB (Broken lines delineate the 95% probability range)

Fig. 6.4 Impulse reaction function from Japan to an increase by the ADB (Broken lines delineate the 95% probability range)

constant terms are αit (i ¼ 1, 2, 3, 4, 5), βi, γ i, δi, θi, and ρi are endogenous terms, and uit are error terms.

6.3 Estimation Results

73

Table 6.4 VAR model of PFI and OTHERS as well as WB, JAPAN, ADB D(WB) 0.445 (0.213)* 0.049 (0.162) 0.403 (0.308) 2.478 (1.333)* 1.093 (0.767) 0.632 (2.160) 0.151

D(WB(1) D(JAPAN(1) D(ADB(1)) D(PFI(1) D(OTHERS(1)) C Adj. R2

D(JAPAN) 0.207 (0.127) 0.349 (0.097)*** 0.910 (0.185)*** 0.610 (0.798) 0.729 (0.459) 0.155 (1.294) 0.687

D(ADB) 0.092 (0.146) 0.130 (0.111) 0.259 (0.211) 1.722 (0.914)* 1.138 (0.526)** 0.438 (1.481) 0.121

D(PFI) –-0.008 (0.034) 0.006 (0.026) 0.005 (0.049) 0.138 (0.212) 0.077 (0.122) 0.126 (0.343) 0.127

D(OTHERS) 0.007 (0.050) 0.033 (0.038) 0.026 (0.072) 0.669 (0.310)** 0.262 (0.179) 0.675 (0.503) 0.041

Notes: Standard error in parentheses. ***: Significant at 1%, **: significant at 5%, *: significant at 10%

3 2 α1t β1 7 6 7 6 6 DðWBÞ 7 6 α2t 7 6 β2 6 t 7 6 7 6 6 6 DðADBÞt 7 ¼ 6 α3t 7 þ 6 β3 7 6 7 6 6 7 6 7 6 6 4 DðPFIÞt 5 4 α4t 5 4 β4 DðOTHERSÞt α5t β5 2 3 u1t 6 7 6 u2t 7 6 7 7 þ6 6 u3t 7 6 7 4 u4t 5 u5t 2

DðJAPANÞt

3

2

32 3 DðJAPANÞt1 γ 1 δ1 θ1 ρ1 76 7 γ 2 δ2 θ2 ρ2 76 DðWBÞt1 7 76 7 7 6 γ 3 δ3 θ3 ρ3 7 76 DðADBÞt1 7 76 7 γ 5 δ4 θ4 ρ4 54 DðPFIÞt1 5 DðOTHERSÞt1 γ 5 δ5 θ5 ρ5

ð6:9Þ

Table 6.4 presents the VAR model resulting from Eq. (6.9). The estimation results reveal that effects are seen from the WB and ADB to Japan, although the opposite is not seen. In addition, although the Granger causality analysis shows that from PFI to Others is statistically significant and effective, PFI and Others do not affect other entities. For this reason, the impulse response from the WB and ADB to Japan was examined. Table 6.5 provides the results for the Granger causality tests of the five entities. Figure 6.5 presents the results of the impulse response from the WB on Japan, while Fig. 6.6 presents them from the ADB on Japan. Figure 6.5 shows that an increase from the WB will lead to a positive impulse response from Japan, because even the 95% lower bound is positive. Similarly, Fig. 6.6 shows that an increase by the ADB has a positive impact on Japan after two years, so an ADB increase will lead to one from Japan.

74

6 Relationships Between Aid Agencies

Table 6.5 Granger causality of five variables Null hypothesis D(JAPAN) ! D(WB) D(WB) ! D(JAPAN) D(ADB) ! D(WB) D(WB) ! D(ADB) D(PFI) ! D(WB) D(WB) ! D(PFI) D(OTHERS) ! D(WB) D(WB) ! D(OTHERS) D(ADB) ! D(JAPAN) D(JAPAN) ! D(ADB)

Ob. 36 36 36 36 36 36 36 36 36 36

F-statistics 0.156 21.637*** 1.491 0.112 1.718 0.294 0.854 0.190 57.949*** 1.181

Null hypothesis D(PFI) ! D(JAPAN) D(JAPAN) ! D(PFI) D(OTHERS) ! D(JAPAN) D(JAPAN) ! D(OTHERS) D(PFI) ! D(ADB) D(ADB) ! D(PFI) D(OTHERS) ! D(ADB) D(ADB) ! D(OTHERS) D(OTHERS) ! D(PFI) D(PFI) ! D(OTHERS)

Ob. 36 36 36 36 36 36 36 36 36 36

F-statistics 0.458 0.018 0.181 0.718 1.489 0.190 1.989 0.010 0.558 4.955**

Notes: Standard error in parentheses. ***: Significant at 1%, **: significant at 5%, *: significant at 10%

Fig. 6.5 Impulse Response from Japan to an increase by the WB (with five entities, broken lines delineate the 95% probability range)

6.4

Conclusion

This chapter considered the relationships between the primary aid agencies of the World Bank, Japan, and the Asian Development Bank through time-series analysis. The main findings are as follows. First, the governmental aid loans from the primary aid agencies—the WB, Japan, and the ADB—indicate that aid loans from the WB increased during the financial crisis, while those from Japan increased during the big flood.

Reference

75

Fig. 6.6 Impulse Response from Japan to an increase by the ADB (with five entities, broken lines delineate the 95% probability range)

Second, governmental aid loans from the three primary aid agencies were examined through long-term relationships. The results showed that Japan was relatively independent from the other agencies, the WB and the ADB were mutually influenced, and that the ADB and Japan had the possibility of mutually influencing each other. In contrast, there was no relationship between the WB and Japan, which is consistent with their disagreement over the Eastern Seaboard. Third, short-term governmental aid loans were examined using a VAR model, Granger causality, and impulse responses. The results indicate that Japan tends to follow the ADB, although no aid agency was a clear leader or follower in terms of aid amounts. In sum, each aid agency in Thailand appears to act relatively independently and protects the field in which it is strong, although more research is desirable.

Reference Matsumura T (2004) Strategic complementarity in direct investments. Rev Dev Econ 8(4):583–596

Chapter 7

Summary and Conclusion

Abstract This chapter provides a concluding remark for this book. First, a summary of each chapter is presented. Second, public opinion about official development assistance (ODA) in Japan and further research issues are discussed. Third, the results of a recent Japanese survey on impressions about ODA is shown with suggestions about further discussions for future ODA. Keywords Official development assistance · Public opinions survey

7.1

Summary of Research Results

The purpose of this book is to show the outcome of ODA from the perspective of economic growth, fiscal budget, and aid agencies, with Thailand as an example. First, the background to this study is described in Chaps. 1–3. Chapter 1 briefly summarizes economic growth and foreign aid in Thailand. Thailand has recorded an economic growth rate of around 5% for more than 50 years, which has promoted poverty reduction. Thailand’s ODA history shows five stages: start-up (from 1968 to 1977), strategic extension (from 1977 to 1989), conversion of values (from 1989 to 1996), financial crisis (from 1997 to 2000), and Thailand as a donor country (from 2000). The budget for ODA is estimated in each stage and changes when the reputation or stance of donor countries or OECD changes. Chapter 2 summarizes the previous literature on the effect of foreign aid from the aspects of economic growth, government budgets, and aid agencies. The widely accepted opinion is that the effectiveness of foreign aid is dependent on governance in the recipient country, and sometimes foreign aid is not effective. Although the effect of foreign aid is still under discussion, there is some consensus that foreign aid does not necessarily affect economic growth or fiscal budget. Chapter 3 summarizes Thailand’s statistics. Since Thailand promoted the collection of statistical data after the 1997 crisis, the data were limited. This chapter summarizes the statistics used in this book: the System of National Account (SNA), labor statistics, population, and socio-economic household survey. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 H. Sakurai, Effects of Foreign Aid, New Frontiers in Regional Science: Asian Perspectives 50, https://doi.org/10.1007/978-981-16-2482-7_7

77

78

7 Summary and Conclusion

Chapters 4–6 summarize the estimation of foreign aid to Thailand. Chapter 4 examines the effect of foreign aid on economic growth in Thailand using the productivity approach. Since most of the foreign aid from development assistance committee (DAC) countries so far is concerned with social infrastructure, foreign aid is considered to be a form of social capital. In addition, the productivity approach is a useful way to measure whether it contributes to economic growth. Since some of statistics are missing, they have been reconstructed from evaluations or from other materials. The results of Chap. 4 indicate that the ODA contributes to economic growth in Thailand to a certain extent. Chapter 5 examines the effect of foreign aid on fiscal conditions in the Thai government. Although foreign aid will increase the recipient country’s capacity from a theoretical point of view, the actual budget of the recipient country is often undiminished or released by inviting foreign aid. The reason is that most of the budget equivalent to foreign aid is used for government consumption rather than investment expenditure. The comprehensive way is to see this through domestic borrowing in the recipient country, since “the absence/presence of aid did not directly alter public expenditure patterns but rather affected government borrowing from the domestic economy” (McGillivray and Morrissey 2004, p. 88). In Chap. 5, the relationship between foreign aid and the fiscal budget in Thailand is divided into domestic borrowings, government expenditure, and governmental revenue. The results show that ODA contributes to fiscal discipline by decreasing domestic borrowings. Chapter 6 presents the relationships between primary aid agencies. Although the cooperation of aid agencies is generally important, it is sometimes difficult to assess from a practitioner’s point of view. Chapter 6 infers the relationship among aid agencies from the data of each by using the receiving amount from each aid agency. The results show that aid agencies act independently to a certain extent and promote good policies. These results are suitable for practitioners and those in other research fields. In sum, the effect of foreign aid in Thailand is examined from the perspective of economic growth, fiscal budget, and aid agencies. The effect of foreign aid was examined at a certain level in each analysis.

7.2 7.2.1

Research Results and Further Research Research Results and the Purpose of This Book

Based on the research results and purpose of this book, the following points can be mentioned. First, this analysis matches the theory and previous studies by using limited timeseries data. Although data in Thailand are restricted before the 1997 financial crisis, the effect of foreign aid can be shown. Second, the belief of practitioners is shown as correct by economic analysis. Foreign aid practitioners believe that foreign aid to Thailand is available mainly

7.3 ODA Has a Bad Impression in Japan

79

through the construction of social infrastructure and inviting foreign direct investment. This hypothesis was clinched as corrected at a certain level by using the productivity approach. Third, research in other fields as well as economics is co-existent. Thai studies have been popular in Japan and the two countries have enjoyed exchanging research. In addition, many practitioners and residents participated in the study groups. These results are suitable for economic theory and studies using long time-series data. Second, these results are suitable for further studies. Next, these results are understandable from the perspective of practitioners. Third, these results are acceptable for researchers in other fields.

7.2.2

Further Research

In addition to Chap. 2, there are still fields that still need to be studied due to the relatively few studies already conducted. In relation to this book, the following three fields are indicated: First, this methodology can be applied to other countries such as Vietnam and Myanmar. These countries are expected to grow more strongly. Sakurai (2020) shows that Vietnam does not necessarily show that foreign aid to Vietnam contributes to economic growth, partly because most infrastructure received by foreign capital was constructed after the late 1990s, meaning that foreign aid has facilitated relatively in a short term. Second, it is not clear whether ODA has a direct impact on poverty reduction or a reduction in income disparity. ODA has an impact on poverty reduction through economic growth, which means that it needs to be careful about balanced growth. Some research results show different outcomes that need to be researched. This is important for improving the national life of recipient countries. Third, the role of aid agencies can be classified more precisely. This analysis is also applicable to relationships among donor countries. Further research is desirable, together with data collection. Fourth, economic growth is relatively available for low-income countries, but it is not that easy to have an impact on middle-income countries—this is called the middle-income trap. Fifth, the impact may differ by aid schemes such as loans, donations, and technical assistance. Although Iimi and Ojima (2008) provide examples of this, few studies have been conducted.

7.3

ODA Has a Bad Impression in Japan

As shown in Chap. 1, Japan’s ODA increased during the 1980s, partly because of the desire of the United States to maintain power. Since Japan decided that it would not have power in politics from the regret of what happened in the World War II, ODA

80

7 Summary and Conclusion

Fig. 7.1 Public opinion about future of Japan’s development cooperation. (Source: Cabinet Office, government of Japan)

was mainly concentrated in the fields of economy and society. After the ODA increased from 1977 to 1997, it decreased for nearly 10 years. The Government of Japan reviewed Japan’s development cooperation strategy in the “Public Opinion Survey on Diplomacy” in 1977. The results in Fig. 7.1 highlight the following three points. First, cooperation should be promoted more actively, and the current level is appropriate, and is nearly the same until 1990. This is partly because the regret of what happened in World War II was widely shared among the people. People also shared the importance of free trade because at that time, Japan imported raw materials and exported manufactured goods, and the main trade partner was the United States. Second, the opinion of promoting ODA declined during the 1990s. Unfortunately, corruption and the ineffectiveness of ODA in the Philippines and South Korea in the 1980s were widely broadcasted. The word ODA became associated with corruption. In addition, the recession after the bubble economy began in the 1990s. At the same time, there was also a general feeling that “Japan should minimize the level of development cooperation.” Third, the decline of ODA reflected national opinion in Japan through the 2000s, regardless of the terrorism act in 2001, since the opinion of the “current level is appropriate” is around the same level during the diminishing budget of ODA. In addition, the opinion of minimizing ODA increased at the beginning of the 2000s. Reflecting on these opinions and the severe budget constraints, ODA from Japan decreased. Figure 7.2. shows the ODA of major countries with Japan as the top donor until 2000. While other DAC countries increased ODA after the terrorism attack in 2001, Japan decreased ODA. As a result, Japan became the fifth largest donor in 2007. Japan’s government has also declared the importance of international cooperation. Interestingly, the word “ODA” and “international cooperation” are perceived differently: “ODA” is associated with corruption and “international cooperation” with contributing to the poverty reduction. Some agree to increase international cooperation even if they disagree about increasing ODA. Reflecting this impression,

7.4 One Suggestion to the Future Foreign Aid

81

Fig. 7.2 Trends in the net ODA of major DAC countries. (Source: Ministry of Foreign Affairs, Government of Japan)

today many people in Japan use the word “international cooperation,” but not the word “ODA.”

7.4

One Suggestion to the Future Foreign Aid

In this chapter, the research results are presented and implications are considered. This section summarizes the reason why efficient use is attained from practitioners’ point of view. From the perspective of the Thai government, as a recipient country, high management ability is mentioned. The Thai government changed its policy stance at the beginning of the 1980s by restricting foreign capital by strengthening the Board of Investment. Inter-ministry meetings were established to solve problems of new locating companies. Economic and social planning goals has been established for every five years, and foreign aid is located in the macroeconomy of Thailand. From the donor side, project aid financed by low-rate loans dominates. Project aid is relatively easy for arranging the promotion of a project, and it is also relatively easy to distinguish aid from consumption expenditure. In contrast, general budget support is used as a compensation in the recipient country’s budget, which makes it relatively difficult to distinguish between governmental consumption and investment expenditure. However, Japan’s opinion is still against foreign aid, as shown in Fig. 7.1, which is reflected in the decreasing budget allocations in the twenty-first century. Since people’s opinion of agreeing to international cooperation is important

82

7 Summary and Conclusion

for increasing the budget of ODA, more research will be one of the elements for indicating the evidence of the widespread usage of the ODA, which connects to confidence not only in the recipient country but also in the donor country. In this regard, more research and discussion in this field are required. Since this field impacts many people, much discussion is necessary. However, poverty reduction is one of the big goals for humans. A big picture perspective is necessary to attain this dream. In addition, developed countries, including Japan, should discuss foreign aid more frequently.

References Iimi A, Ojima Y (2008) Complementarities between grants and loans. J Jpn Int Econ 22:109–141 McGillivray M, Morrissey O (2004) Fiscal effect of aid. In: Addison T, Roe A (eds) Fiscal policy for development: Poverty, Reconstruction and Growth. Palgrave Macmillan for UNU-WIDER, New York, pp. 72–96 Sakurai H (2020) Foreign aid loans and economic growth in Vietnam. Bull Appl Econ 7(1):83–94