The growth effect of international financial integration evidence from cross country analysis

UNIVERSITY OF ECONOMICS ERAMUS UNIVERSITY ROTTERDAM HO CHI MINH CITY INSTITUTE OF SOCIAL STUDIES VIETNAM THE NETHERLANDS VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMIC

UNIVERSITY OF ECONOMICS ERAMUS UNIVERSITY ROTTERDAM

HO CHI MINH CITY INSTITUTE OF SOCIAL STUDIES VIETNAM THE NETHERLANDS

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

THE GROWTH EFFECT OF INTERNATIONAL FINANCIAL INTEGRATION: EVIDENCE FROM

CROSS COUNTRY ANALYSIS

BY

HO THY DUNG

Academic Supervisors:

Dr DUONG NHU HUNG

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, December 2016

UNIVERSITY OF ECONOMICS ERAMUS UNIVERSITY ROTTERDAM

HO CHI MINH CITY INSTITUTE OF SOCIAL STUDIES VIETNAM THE NETHERLANDS

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

THE GROWTH EFFECT OF INTERNATIONAL FINANCIAL INTEGRATION: EVIDENCE FROM

CROSS COUNTRY ANALYSIS

By

HO THY DUNG

Academic Supervisors:

Dr DUONG NHU HUNG

HO CHI MINH CITY, December 2016

ACKNOWLEDGMENTS

First of all, I would like to express my heartfelt gratitude to Doctor Duong Nhu Hung, my supervisor His inspiration together with supports has encouraged

me to complete this thesis

I also take this chance to convey my sincere thanks to Doctor Truong Dang Thuy I am so appreciate to his help and enthusiasm

Finally, I am deeply grateful to my family, friends, Ms Tang Thi Xuan Hong (Programme Secretary) and my colleagues for their help and sharing of their resources to complete this thesis successfully and in time Thanks so much for everything!

ABSTRACT

Currently, international financial integration (IFI) is a trend of the world However, there is little evidence of its positive impact on the economic growth The reasons can be error in measurement of IFI indicators, the multicollinearity among variables in model (Quinn and Toyoda, 2008) They may lead to the inconsistent results and accounts in part for larger standard errors and biased estimator Different data might also contribute to these inconsistent results The targets of this paper is to study the role of IFI on economic growth with impact of macroeconomic instability using FDI and portfolio investment capital with 2 weighted ways to proxy for IFI With the latest annual data in the period of 1982-

2014, the paper employs GMM method for estimating the panel-data We start by employing VIF indicators and normal distribution assumption to choose conditional variables, dropping outliers and collinearity We also provide a global picture about the economic performance of countries in the world

In empirical analysis, my research finds a robust and positive statistically significant association between flow of FDI and the real GDP per capita growth The finding emphasizes the important role of IFI in enhancing economic growth

We also find a weak evidence that macroeconomic instability has negative effect

on the impact of IFI on economic growth Interestingly, my thesis asserts the existence of inverse linkage between money supply and economic growth This finding is sharply contrasts with conventional wisdom of many people who think that financial deepening will benefit for economic growth

ABBREVIATIONS

i IFI International financial integration

ii FDI Foreign Direct Investment

iii FE Fixed Effect Model

iv RE Random effect model

v GMM Generalized moment of method

vi VIF Variance inflation factor

vii OLS Ordinary least squares

TABLE OF CONTENTS

CHAPTER 1 1

INTRODUCTION 1

1.1 PROBLEM STATEMENT 1

1.2 Research objectives 2

1.3 Research questions 2

1.4 Research Methodology 2

1.5 Organization of the study 3

CHAPTER 2 4

LITERATURE REVIEW 4

2.1 International financial integration 4

2.2 Theoretical Background underlying IFI and economic growth 4

2.2.1 Classical Theory 5

2.2.2 Exogenous Growth Theory 5

2.2.3 Endogenous Growth Theory 6

2.2.3.1 Learning by doing model 6

2.2.3.2 Research and development model 6

2.3 Theoretical Background relate to relationship between IFI and growth 7

2.3.1 Overview of relationship between IFI and growth 7

2.3.2 Empirical framework 11

2.4 Conceptual framework 16

CHAPTER 3 18

RESEARCH METHODOLOGY 18

3.1 Determinants of international financial integration 18

3.1.1 De jure measures 18

3.1.2 De factor measures 19

3.2 Data source 20

3.2.1 Measures of international financial integration 20

3.2.2 Data for variables in model 20

3.3 Model Specification 21

3.3.1 Model 21

3.3.2 Variables specification: 22

3.3.2.1 The real GDP per capita growth 23

3.3.2.2 The international financial integration 23

3.4 Methodology 26

CHAPTER 4 28

EMPIRICAL RESULTS 28

4.1 Overview of real GDP per capita growth in The World 28

4.2 Preliminary statistics 29

4.2.1 Descriptive statistics 29

4.2.2 Correlation of coefficients 31

4.3 Empirical result for panel data 32

CHAPTER 5 37

CONCLUSION AND POLICY RECOMMENDATIONS 37

5.1 Conclusion 37

5.2 Policy Implications 37

5.3 Limitation and further research 38

REFERENCE 39

APPENDIX: 41

List of 48 countries in research 41

Regression Result from Panel Data 41

Result with IFI1 and all control variables 41

Result with IFI1, adding and subtracting variables 42

Result with IFI2 and all control variables 44

Result with IFI2, adding and subtracting variables 44

Fisher test for independent variables 47

LIST OF TABLES

Table 2.1: Effect of IFI on growth in 11 aspects……….8

Table 2.2: Summary of the empirical papers relate to relationship between IFI and growth

……….15

Table 3.1: Data for international financial integration ……… 20

Table 3.2: Summary of variables in the model ……… 25

Table 4.1 and 4.2: Descriptive statistics……… …….30

Table 4.3: Correlation matrix……….31

Table 4.4: Fisher test with lag (1)……… 33

Table 4.5: Panel regression………34

LIST OF FIGURES Figure 2.1: The relationship between international financial integration and economic growth……… 17 Figure 4.1: Trend of real GDP per capita growth in the World from 1982 to 2014….… 28

Chapter 1 INTRODUCTION

1.1 PROBLEM STATEMENT

Capital have played an important role in the economic theory.Adam Smith and David Ricardo have perspectives that stress on the important of capital accumulation, social capital, free trade, etc in which the economy is self-regulating and always at full-employment (Letiche, 1960) Exogenous growth theory of Slow and Swan stated that capital accumulation is one of input factors of the economy growth in long run (Solow, 1956, Swan, 1956) Arrow (1962) pointed out that specialization would affect the comparative advantage of a country And flow of capital or new investment as a source stimulate the learning by doing process Hence, both physical and human capital will increase economic growth In Cobb-Douglas function, we see that capital

is also main input of the economic growth (Hall and Jones, 1996)

Since the late 1980s, the world has witnessed a trend towards the deepening integration of the domestic financial markets There are many research on growth benefit of international financial integration (IFI) but they cannot provide strong evidence on this subject Then researchers tend to find out the threshold effects in the process of IFI According to their view, financial globalization will have better impacts with a certain level of some threshold conditions (Kose et al., 2006) However, majority of them show a mixed effects or no robust result while only a number of papers find a positive benefit of financial globalization (Kose et al., 2006, Kose et al., 2010)

Recently, it is important to understand the impacts of IFI on growth when the trend of opening financial market generally continued over the world IFI not only lead

to large benefits but can also come with crises and contagion (Schmukler, 2004) Many critics argued that IFIs’ benefits were intangible and undocumented after the Asian crisis of 1997–98 (Obstfeld, 2009) So where is the real result of the financial integration process that many countries are pursuing? In addition, are there certain conditions for the better impacts of IFI on growth? Although there are many debates, empirical research on this issue have just exploded in recent time and along many

For these reasons, I would like to present some new evidences on the impacts of IFI on economic growth Whether international financial integration could be a reason for an increasing in economic growth or not? This study concerned about the differences in economic growth across countries in the world from 1982-2014 and possible factors that affect economic growth We use a large data set, which includes more countries and additional years In particular, I look at capital flow as channel through which international financial integration affect economic growth

1.2 Research objectives

My main research objectives are listed below:

 Examine whether IFI has taken part in increasing economic growth in the world

 Examine whether the growth effect of IFI is influenced by macroeconomic instability and financial deepening

 To suggest policies to speed up economics growth associated with IFI process of countries in the world

1.3 Research questions

In order to achieve these objectives, my thesis is in an effort to find out answers for the two following research questions:

a Does IFI have a positive impact on economics growth at country level?

b Does the growth effect of IFI vary under different development characteristics of countries?

1.4 Research Methodology

My research intended to use quantitative method with panel data of 48 countries from 1982 to 2014 The use of panel data has many advantages such as we could control for unobserved heterogeneity and to rule out the bias of omitted variable

I use VIF indicators and normal distribution assumption to choose conditional variables and dropping outliers Because of the possibility of endogeneity between IFI variables and real GDP per capita, real GDP per capita and lag of it I will use GMM command in Stata We run regression for each period of three years because it helps to

1.5 Organization of the study

My paper consists of four main parts The first part is introduction chapter Chapter 2 is the theoretical framework to make sure that my research is built on a truly

of scientific knowledge It widely discuss about definition of IFI, present theoretical insights as well as empirical works of previous scholars It also present the regression results from panel data using GMM command Chapter 3 is the research methodology

In this chapter, I will set up models to find out the impact of IFI on economics growth with conditional factors Chapter 4 draws conclusion of this research and chapter 5 recommend policies in order to help policy makers have better policies with IFI process

Chapter 2 LITERATURE REVIEW

I begin this chapter by discussing Aspects of IFI Next, the theoretical background about relation of economic growth and international financial integration along is presented Finally, we have an overview of empirical works about this issue and their results

2.1 International financial integration

The concept of market integration is central to research in international finance, international economics and development economics Before defining international financial integration, we look at financial integration Emanuelsson et al (2012) generalized that we have direct financial integration when the risk-adjusted returns are the same between financial markets We have indirect financial integration when the return of investments between different markets are linked indirectly And we have total financial integration when direct and indirect financial integration are coexist Moreover, we also have concept of perfect total financial integration and segmentation

In this case, the markets will have one real interest rate or opposite circumstance, respectively The level of integration is influenced by transaction cost between markets

Based on empirical studies, the researchers used many ways to define financial integration The most common approach focuses on geographical integration This is also the way that our paper research the financial integration It mention international financial integration or financial integration across countries and their economic growth

2.2 Theoretical Background underlying IFI and economic growth

IFI means the way capital flows move across countries, which results a direct change of domestic Savings, technological, management and domestic financial sector development Beside of that IFI also promote for specialization, better economic policies and signaling (Prasad et al., 2003) Hence, discussion about theories that relates to IFI and growth is very useful It helps us a clearer understanding about growth as well as the simplest way IFI affect to growth In other way, economic

growth came from stock of factors of production or efficiency of those factors This laid the basis for model of the research

2.2.1 Classical Theory

It is represented by Adam Smith and David Ricardo Smith stated that the economy is influenced by an invisible hand with the objective economic rules However, it need conditions including commodity production and commodity exchange He also thought that the accumulation of capital and the efficiency of labor are two main factors lead to the economic growth Capital accumulation will enhance the expansion of the market and the specialization of labor With high level of specialization, the wage rise that lead to the increase in the demand and the expansion

of the market

And David Ricardo modified it by including diminishing returns to land The technological changes are highly assessed for impacts on the economic growth (Lanza, 2012)

In general, the theory determines three factors having the effects on the economic growth are capital, labor and technology Which are severed direct effects of IFI IFI will lead to an increase in capital accumulation via domestic saving because the decrease in capital returns And FDI inflows rise during IFI process has generated technology spillovers and better management (Prasad et al., 2003)

2.2.2 Exogenous Growth Theory

Exogenous growth theory is also known as “neoclassical growth theory” The model contributed to economic growth literature in the long run by exploring factors including capital accumulation, the growth of population and productivity (Solow,

This model suggests that IFI will lead to allocation of capital, specialization and sharing of risk Then it lead to the increase in productivity and growth (Kose et al., 2006)

2.2.3 Endogenous Growth Theory

2.2.3.1 Learning by doing model

Learning by doing concept mentions emphasize repetition task and practice of worker It will help workers to increase their ability

This model also concentrate on comparative advantage and growth that are involved to trade A country can get comparative advantage and growth by specialize

in producing goods with higher potential for learning Trade affect specialization of a country and rely on the level of learning externalities

Learning by doing was introduced by Arrow in 1962 He state that new knowledge can be produced through the learning by doing process is diminishing And

we need new flows of capital to stimulate this process (Arrow, 1962) This can be seen more clearly in the production function of Cobb – Douglas as follow:



 

L BK Y

Where Y is the output of economy, L is the labor force, K is the capital which is included both physical and human capital B is the level of knowledge increase from learning by doing process

2.2.3.2 Research and development model

Research and development model was introduced firstly by (Romer, 1986) He argued that knowledge is an input in long run growth model In this model, growth rate can be increasing over time because increasing marginal productivity of knowledge

Jones (1995) summarized R & D model in the following equation:





)(

1

Y AL K

Where Y is output, K is capital, A is stock of knowledge or productivity, which is already exist in the economy, L is labor We can simply understand A as stock of

knowledge that were accumulated in the past by researchers Labor can be used to produce output or create new knowledge

Base on this equation, an increase in stock of knowledge would experience an increase in economy input And countries invest more in R&D also has higher economic growth

In general, Endogenous Growth Theory revolves around the elements capital, labor and technology and productivity Hence, we can expect a signal relationship between IFI and Growth when we compare with result of IFI process in economy

Besides the theory basis above, (Kose et al., 2006) point out that output volatility

of IFI not explicit in theory Simultaneously with risk diversification, IFI can be affected by certain features that make countries more vulnerable to external shocks Hence, effect of IFI also base on conditions of the economy

2.3 Theoretical Background relate to relationship between IFI and growth

2.3.1 Overview of relationship between IFI and growth

Based on theories, countries attend IFI process not only get potential benefits but also bear potential cost When a country access to international financial market, its investors have more opportunities to diversify portfolio, achieve higher risk-adjusted rates of return Domestic economy gain from international risk sharing and abundant capital However, the risk of volatility and flow of capital may be a significant cost Evidence are financial crises that comes with instability of cross-border financial markets and the risks of international financial transactions (Agénor, 2003)

Table 2.1: Effect of IFI on growth in 11 aspects

Consumption smoothing

‘counter-cyclical’ role of world capital markets Stabilization in temporary shocks

Domestic investment and growth

complementarity effect play an important role positively relate to economic growth

The efficiency of banking

system and the stability of

financial market

Increase competition of banking system and decrease risk of depositors by foreign institutions

in periods of financial instability

improvement of domestic financial markets and risk diversification of depositors

The allocation and lack of access

distortions and poor supervision

of domestic financial system Limitation in long-run growth Loss of macroeconomic stability flexible exchange rate increase financial volatility

The cyclicality of short-term

capital flows asymmetric resources increasing macroeconomic instability

volatility of capital flows

sensitive movement of short-term capital flows Financial contagion

Participation of foreign banks

liberal lending policies, lower operational costs and sensitive movement of foreign banks

‘too big to fail’ domestic bank and instability of domestic banking system

Source: (Agénor, 2003)

We can see that benefits and costs of IFI surround capital flows Especially, FDI capital play a core role when portfolio investment capital have an important position In addition, final potential gain is higher economic growth rate but difference between countries base on initial conditions

Kose et al (2006) research same issue with (Agénor, 2003) but they summarized impacts of IFI in two views These views are also discussed in (Prasad et al., 2003)

Traditional view mentions on direct channels in which, capital flows will lead

to GDP growth and decrease consumption volatility

Source: (Kose et al., 2006)

And other view focuses on the catalyst effect of IFI for certain collateral benefits

Source: (Kose et al., 2006)

According to Schmukler (2004), IFI tend to be positive in long-run for some countries if they get over challenges He focus on developing countries and state that

with IFI process, capital flows allocate mainly in middle-income countries and not same for different domestic sector

Almost Researchers agree that net benefit of IFI can be positive or negative depend on conditions of countries And (Kose et al., 2011) generalized them into thresholds that countries need to overcome if they want to gain from IFI

Source: (Kose et al., 2011)Base on existing theoretical literature, he choose thresholds including financial market development, the overall level of development and other national index such

as trade openness and labor They find that conditions are much lower when IFI is proxied by stocks of FDI and portfolio equity liabilities The thresholds are also different with data samples Most of developed countries and only some emerging markets are above threshold levels while almost developing countries achieve them

In general, the literature points out two views on the impacts of IFI on growth The composition hypothesis focuses on all capital flows and shows their effects on growth Researchers find out the role of each kind of capital flows on growth, compare positive and negative effects The threshold hypothesis focuses on certain minimum conditions that each country need to meet to benefit from IFI However, according to (Wei, 2006), these two hypothesis can be “the two sides of the same coin” He found that nations have low development of financial sector attract inward portfolio equity flows more than inward foreign direct investment

2.3.2 Empirical framework

Until now, many studies research the relationship between IFI and economic growth Most of papers do research based on a “large n” sample with various methodologies and variables The majority of researcher can not to find clear evidence

on the growth effect of IFI while only a number of papers can get systematic evidence

to support this point (Kose et al., 2006, Rodrik and Subramanian, 2009) In addition, they could not find enough systematic evidence to explain the impacts of IFI on growth based on threshold effects

For the first time, de Jure measures were mainly use to evaluate IFI Quinn (1997) has set an assessment by himself He developed coding rules based on IMFs’ AREAER to calculate changes in international financial regulation He adopted a scale for each dimension of goods, exchange restriction and invisibles payments and receipts, international agreements and laws Data include 64 countries for 1958 to

1989 Firstly, a basic model of long run growth in which the measure of change in IFI was added to as an independent variable Then the model was added and subtracted variables to see the changes He found that capital account liberalization (CAL) robustly linked with economic growth In 2008, Quinn and Toyoda reexamine the role

of CAL on economic growth with data of 94 nations from 1950 to 2004, using pooled time-series, cross-sectional OLS and system GMM estimators GMM estimators allow for the endogeneity of CAL to growth This study have shown robust evidence for a positive benefit of financial liberalization to economic growth and stated that conflicting results may derive either from measurement error or differing periods of data (Quinn and Toyoda, 2008)

The later stages, there is a trend of using de factors measures or various measures and methods to research this relation because the availability of data and the extent of the growing international financial integration increasing ability to reflect IFI

of de factors measures

De Gregorio Rebeco (1999) emphasize the role of financial development in finding the effect of IFI on growth Although the results are weak, he found a positive relation between IFI and the development of domestic financial system However, no

welfare effects might allow a greater degree of consumption smoothing It explain portfolio diversification effect is small compare with theories

Edison et al (2002) is a pioneer in using mixed measures, n-large data and various methods to research the impact of IFI on economic growth and to evaluate whether the growth effect of IFI depends on the development of the domestic economy The paper exams 57 countries over the 1980–2000 or 1976-2000 period Statistical methodologies include simple ordinary least squares (OLS) regressions, two-stage least squares instrumental variable estimator and dynamic panel procedure (generalized method of moments (GMM)) The dynamic panel approach help author

to control for country specific effects, which may bias the coefficient estimators In order to reduce the cyclicality of the data, five-year averages were chosen to run model Assortment of measures and methodologies can contribute to the robustness of result The reasons can be international differences in barriers to financial transactions and measures cannot fully distinguish them He found no robust evidences on the growth benefit of IFI even when controlling for particular conditions However, the research can be referenced because the generalizability of the study

Similarly to (Edison et al., 2002), Vo (2005) also use many measures and methodologies to research this topic but bigger data sample (79 countries), broader and more detail measures covering the period from 1980 to 2003 In particular aspects, the result is quite same with (Edison et al., 2002) when it point out the relationship between IFI and economic growth not depending on different economic conditions However, he found out evidences indicating a weak relationship between IFI and growth It is consistent with the current literature

Because advantages of panel data and dynamic panel approach, almost of later research use GMM model and find a breakthrough from the different sample data, the measurement of the IFI or the conditions of the economy in the study sample

Masten et al (2008) using macro and industry-level data in Europe with panel estimation to analyze the effect of financial development and IFI on economic growth

He found that countries with higher levels of financial development get more benefits

of IFI than less developed European countries

Lucey and Zhang (2011) using a sample of 4477 public firms from 24 countries and investigate positive benefit negative benefit of corporate leverage to credit market integration and equity market integration respectively And Gamra (2009) used a measure set by himself based on another research of (Kaminsky and Schmukler, 2002)

He concluded, “Full liberalization of the financial sector has been associated with slower growth outcomes while more moderate partial liberalization is associated with more positive outcomes”

Chen and Quang (2014) provided new evidence relying dynamic panel techniques With the panel threshold regression, he found some evidences that domestic financial development and moderate government spending will support potential gains from IFI

Schularick and Steger (2010) run regression covering 24 countries for the years 1880-1914 using both cross-section regression and GMM panel estimation The result show that in the first era of financial globalization, IFI boosted economic growth but

no robust relation today Because employing similar models and techniques with former researcher, author stated that not the model or regression method but data that drive the findings

Focusing on country specific effect, several researchers point out that countries that have higher level of wealth and education tend to have higher level of integration Researchers consider IFI with a set of threshold conditions Empirical studies on the benefit of IFI on economic growth has used the growth rate of real per capita GDP growth as measure of economic growth, measures of IFI plus control variables which proxy growth drivers with dynamic panel model (Schularick and Steger, 2010) Kose

et al (2011) stated that there are levels of country characteristics such as domestic financial market development, institutional quality, labor market rigidities, trade openness, general development that are important determinants of growth benefit of IFI

Broto et al (2011) state that global factors get important role link to country specific drivers Wyplosz (2001) found that developing countries have lower level of financial liberalization than developed countries Furthermore, financial liberalization

Another trend of Inherit to former theoretical framework and research methodology of the IFI, a large number of the researchers in the last few centuries focused explore the impact of the IFI to the developing countries (DCs) Although theoretical basic of growth benefit of IFI in Developing countries is strong, there is little evidence about this issue IFI can offer a major boost to growth or cause an exodus of domestic capital if preconditions is sufficiently weak (Prasad et al., 2003) Majority of researchers using a “large n” sample including to evaluate the impact of IFI on growth They usually do research in general on these two groups after control for some initial conditions (Quinn, 1997, Masten et al., 2008) The papers that focus

on DCs countries are still little and not systematic They focus on different sector through which IFI affect economic growth such as business cycle, domestic development, stock market, CAL Özbilgin (2010) stated that financial integration are associated with higher investment and output volatility

Table 2.2 generalize empirical papers that have presented in this section It can help us have an overview about research of relationship between IFI and economic growth

Table 2.2: Summary of the empirical papers relate to relationship between IFI and growth

Authors Time Period Observations Method Findings

Quinn (1997) 1960 to 1989 64 countries OLS Capital account liberalization

(CAL) robustly associated with economic growth Quinn and

Toyoda (2008)

1955 to 2004 94 countries pooled

time-series, OLS, fixed effect regression and GMM

A positive relation between financial liberalization and economic growth

de Gregorio

Rebeco (1999)

1976 to 1993 24 countries OLS Weak positive link from

financial integration to the development of domestic financial system but no evidence of effect of IFI on growth

Edison et al

(2002)

1980–2000 or 1976-2000

A weak relationship between IFI and growth

Gamra (2009) 1980–2002 6 emerging

East Asian countries

LS, 2SLQ, panel fixed effect, GMM

Moderate partial liberalization will get more positive outcomes than full liberalization

Chen and

Quang (2014)

1984–2007 80 countries Panel

Estimator, GMM

Domestic financial development and moderate government spending will support potential gains from IFI

Some country characteristics are important determinants of growth benefit of IFI

To sum up, classical theory, exogenous growth theory and endogenous growth theory strongly support the role of stock and efficiency of capital factor on economic growth According to these theories, capital lead to resource allocation, technology transfer, innovation and specialization that contribute to economic growth And empirical studies research the relationship between IFI and economic growth based on these literature They function economic growth with IFI is an independent variable because IFI process will affect stock and efficiency of capital factor

Based on literature review, I lead to hypothesis as follow:

H1: IFI has a positive link to economic growth

H2: In country with low inflation rate, the growth effect of IFI tend to be higher

Based on above arguments I try to avoid the possibility that lead to conflict results of empirical studies that are measurement error, different data, the clustering and collinearity With annual data in the period of 1982-2014, we have the latest data

I employ VIF indicators and normal distribution assumption to choose conditional variables, dropping outliers and collinearity Moreover, I also use GMM regression that is the best method in the present

Figure 2.1: The relationship between international financial

integration and economic growth

International financial integration

Macroeconomic instability Financial deepening

Resource allocation Technology transfer Innovation Specialization

FDI flow

Stock of FDI and portfolio investment

Economic growth

Chapter 3 RESEARCH METHODOLOGY

This chapter is divided into 3 parts Part 1 presents the measurement of international financial integration Part 2 will point out sampling method and combination of data And part 3 will specify models and estimated method

3.1 Determinants of international financial integration

Until now, there are two broad categories of IFIs’ measures

3.1.1 De jure measures

With this measure, IFI refers to an individual country’s linkages to international capital markets The level of IFI will be defined based on financial policies and actual capital flows (Wang, 2004)

The most popular approach is to use measures of capital account liberalization (CAL) And there are two main kinds of these measures Quinn (1997) may be the first person that used IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER) to evaluate CAL He use a range with more than two level to measure legal restrictions on capital account liberalization After Quinn, many authors use similar way but the measures are binary indexes Edison et al (2002) has used IMFs’ restriction index that will equal one and zero to refer restrictions on capital account transactions or no restrictions respectively Gamra (2009) set a binary index that based on the dates that a financial market is full liberalization or partial liberalization, etc These indexes can capture legal capital controls but they cannot provide the intensity of CAL

Beside the ways above, Chinn and Ito has created Kaopen index that measures the extent of openness in capital account transactions The index was initially introduced in Chinn and Ito (2008) Kaopen is based on the binary dummy variables that reported in the IMF's Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER)

3.1.2 De factor measures

The second trend to measure CAL is various variables that control for actual capital flows These flows may partly be caused by capital restrictions but may also respond to many different nongovernment considerations (Quinn and Toyoda, 2008) Hence, these indexes can capture better the intensity of CAL in practice This evaluation is very popular in literature with “large n” research Edison et al (2002) use 2 variables that are stock of capital flows and Flow of capital to assess CAL Total capital inflows and total capital flows are 2 of 4 measures of CAL in (Honig, 2008)

On the other hand, measure of IFI based on the estimated gross stocks of foreign assets and liabilities as shares of GDP provide a better indication because they are more stable from time and are limit error in measurement (Prasad et al., 2003) And to avoid the volatility and proneness in measurement error, it is better to use the total of gross stocks of foreign assets and liabilities as a ratio to GDP (Kose et al., 2006) Edison et al (2002) used two De factor measures that are Inflows of capital and stock of capital inflows

Both measures above have particular advantages However, De Jure measures have limitation in data and they show just one facet of IFI Implementation and enforcement of restrictions on capital flows differs so greatly across countries For example, for some countries, the extent of mobility legally allowed usually be lower than observed capital flows Hence, this research decide to use De factor measures

As literature review, we can see that many measures were used but capital flows is the most frequently index because high availability of data and capital is a main factor in model of almost growth theory Between capital flows, FDI and portfolio investment were valued that have important role in economic growth or in positive effect of IFI on growth (Abbes et al., 2015, Alfaro et al., 2004) Hence, we use flow and stock of FDI with stock of portfolio investment to evaluate relation of IFI and growth in short-run and long run

3.2 Data source

This research use secondary data that compiled by World Bank database and UNCTAD database The sample aggregate data of 48 countries in the world covering the period from 1982 to 2014 The reason for choosing 48 countries from 1982 to

2014 as our sample reflects the availability of data that we can collect 48 countries is the biggest number of countries that we have necessary data Besides of that, I use data from 1982 to 2014 since this is the most recent data, which would give the most up-to-date picture about the effects of IFI on GDP And it have higher quality than previous period For more information about our sample, please refer to appendix

3.2.1 Measures of international financial integration

Because the model use weighted variables to proxy for IFI, we start to skim data source of component variables as below table And the way to weight IFI variable will be presented in table 3.2 (page 24)

Table 3.1: Data for international financial integration

Variable Variables definition Data source

Stock FDI inwards

or outward as share

of GDP

Ratio of Stock FDI inwards or outward to GDP (%)

http://unctadstat.unctad.org

3.2.2 Data for variables in model

To measure economic growth, we use real GDP per capita growth Based on theoretical literature, GDP is represented for output of growth Because different between countries in sample, per capita index is necessary Data of this variable is available from http://databank.worldbank.org