The Macroeconomic Analysis of Foreign Capital Inflows in Pakistan

Contents

Preface

Executive Summary (Abstract)

1.1 Introduction and Background
1.2 Statement of Problem
1.3 Objectives of the Study
1.4 Significant Contributions of the Study
1.5 Organisation of the Study

2.1 FCI and Economic Growth: A Review of Literature
2.2 FCI and Domestic Savings: A Review of Literature
2.3 FCI and Investment: A Review of Literature
2.4 FCI and Exchange Rate: A Review of Literature
2.5 Some Other Aspects of FCI: A Review of Literature

3.1 Foreign Capital Inflows: A Theoretical Framework
3.2 Hypothesisto be tested
3.3 Data Specifications
3.3.1 Types and Measurement of FCI
3.3.2 Variables Description
3.3.3 Sources of the Data and Sample Period
3.4 Model Specifications
3.4.1 FCI and GDP: Vector Error-Correction Model (3.1)
3.4.2 FCI and Investment: Vector Error-Correction Model (3.2)
3.4.3 FCI and Savings: Vector Error-Correction Model (3.3)
3.4.4 FCI and CAD: Vector Error-Correction Model (3.4)
3.4.5 FCI and Exchange Rate: Vector Error-Correction Model (3.5)
3.4.6 FCI and Imports: Vector Error-Correction Model (3.6)
3.5 Research Methodology

4.1 Trends, Composition and Sector-Wise Distribution of FCI in Pakistan
4.1.1 Foreign Capital Inflows to Pakistan: Trends 1972-2006
4.1.2 Composition of FCI in Pakistan (19976-2006)
4.1.3 Sector-wise Distribution of FCI in Pakistan
4.2 A Need-Based Analysis of FCI in Pakistan
4.2.1 Savings-Investment Gap in Pakistan
4.2.2 Export-Import Gap and FCI in Pakistan
4.2.3 Current Account Deficit and FCI in Pakistan

5.1 FCI and GDP: Econometric Analysis and Results
5.1.1 FCI and GDP: Unit Root Test& Order of Integration
5.1.2 FCI and GDP: VAR Lag-Order Selection
5.1.3 FCI and GDP: Johansen Cointegration Test Results
5.1.4 FCI and GDP: VECM Estimations and Results
5.1.5 FCI and GDP: Impulse Response Function
5.1.6 FCI and GDP: Results and Findings
5.2 FCI and Investment: Econometric Analysis and Results
5.2.1 FCI and Investment: Unit Root Test & Order of Integration
5.2.2 FCI and Investment: VAR Lag-Order Selection
5.2.3 FCI and Investment: Johansen Cointegration Test
5.2.4 FCI and Investment: VECM Estimation and Results
5.2.5 FCI and GCF: Impulse Response Function
5.2.6 FCI and GCF: Results and Findings
5.3 FCI and Savings: Econometric Analysis and Results
5.3.1 FCI and Savings: Unit Root Test & Order of Integration
5.3.2 FCI and Savings: VAR Lag-Order Selection
5.3.3 FCI and Savings: Johansen Cointegration Test
5.3.4 FCI and Savings: VECM Estimation and Results
5.3.5 FCI and GDS: Impulse Response Function
5.3.6 FCI and GDS: Results and Findings
5.4 FCI and Current Account Deficit: Econometric Analysis and Results
5.4.1 FCI and CAD: Unit Root Test & Order of Integration
5.4.2 FCI and CAD: VAR Lag-Order Selection
5.4.3 FCI and CAD: Johansen Cointegration Test
5.4.4 FCI and CAD: VECM Estimation and Results
5.4.5 FCI and CAD: Impulse Response Function
5.4.6 FCI and CAD: Results and Findings
5.5 FCI and Exchange Rate: Econometric Analysis and Results
5.5.1 FCI and Exchange Rate: Unit Root Test & Order of Integration
5.5.2 FCI and Exchange Rate: VAR Lag-Order Selection
5.5.3 FCI and Exchange Rate: Johansen Cointegration Test
5.5.4 FCI and Exchange Rate: VECM Estimation and Results
5.5.5 FCI and Exchange Rate: Impulse Response Function
5.5.6 FCI and Exchange Rate: Results and Findings
5.6 FCI and Imports: Econometric Analysis and Results
5.6.1 FCI and Imports: Unit Root Test & Order of Integration
5.6.2 FCI and Imports: VAR Lag-Order Selection
5.6.3 FCI and Imports: Johansen Cointegration Test
5.6.4 FCI and Imports: VECM Estimation and Results
5.6.5 FCI and IMP: Impulse Response Function
5.6.6 FCI and IMP: Results and Findings

6.1 Conclusion
6.2 Policy Recommendations

A Data and Statistics

B Time Series Properties of the Variables

C Estimates of Lag Length Selection

D Estimates of Johansen Cointegration Test

E Estimates of Vector Error-Correction Model

F Impulse Response Functions

List of Tables

Table ‎0.1: Foreign Capital Inflows and Components (in Millions of US$)

Table ‎0.2: GDP and Other Macroeconomic Indicators (in Million of USD and %)

Table ‎0.3: Sector-wise Disbursement of Foreign Private Loans (2002-2007)

Table ‎0.4: Sector-wise Disbursement of Foreign Direct Investments (2003-2007)

Table ‎0.5: Definitions of the Variables

Table ‎0.1 - LNFCI: ADF Test at level (with Intercept)

Table ‎0.2 - LNFCI: ADF Test at level (with Intercept & Trend)

Table ‎0.3 - LNFCI: ADF Test at 1st Difference (with Intercept)

Table ‎0.4 - LNFCI: ADF Test at 1st Difference (with Intercept & Trend)

Table ‎0.5 – LNGDP: ADF Test at level (with Intercept)

Table ‎0.6 – LNGDP: ADF Test at level (with Intercept & Trend)

Table ‎0.7 – LNGDP: ADF Test at 1st difference (with Intercept)

Table ‎0.8 – LNGDP: ADF Test at 1st difference (with Intercept & Trend)

Table ‎0.9 – LNGDS: ADF Test at Level (with Intercept)

Table ‎0.10 – LNGDS: ADF Test at Level (with Intercept & Trend)

Table ‎0.11 – LNGDS: ADF Test at 1st Difference (with Intercept)

Table ‎0.12 – LNGDS: ADF Test at 1st Difference (with Intercept & Trend)

Table ‎0‎0.13 – LNGCF : ADF Test at Level (with Intercept)

Table ‎0.14 – LNGCF : ADF Test at Level (with Intercept & Trend)

Table ‎0.15 – LNGCF: ADF at 1st difference (with Intercept)

Table ‎0.16 – LNGCF: ADF at 1st difference (with Intercept & Trend)

Table ‎0.17 – CAD/GDP: ADF at level (with Intercept )

Table ‎0.18 – CAD/GDP: ADF at level (with Intercept & Trend)

Table ‎0.19 – CAD/GDP: ADF at 1st Difference (with Intercept)

Table ‎0.20 – CAD/GDP: ADF at 1st Difference (with Intercept & Trend)

Table ‎0.21 – LNOER: ADF at level (with Intercept)

Table ‎0.22 – LNOER: ADF at level (with Intercept & Trend)

Table ‎0.23 – LNOER: ADF at 1st Difference (with Intercept)

Table ‎0.24 – LNOER: ADF at 1st Difference (with Intercept & Trend)

Table ‎0.25 – IMP/GDP: ADF at level (with Intercept)

Table ‎0.26 – IMP/GDP: ADF at level (with Intercept & Trend)

Table ‎0.27 – IMP/GDP: ADF at 1st Difference (with Intercept)

Table ‎0.28 – IMP/GDP: ADF at 1st Difference (with Intercept & Trend)

Table ‎0.1 - LNFCI and LNGDP – VAR Lag-Order Selection Results

Table ‎0.2 LNFCI and LNGDS: VAR Lag-Order Selection Results

Table ‎0.3 – LNFCI and LNGCF: VAR Lag-Order Selection Results

Table ‎0.4 – LNFCI and CAD/GDP: VAR Lag-Order Selection Results

Table ‎0.5 – LNFCI and LNOER: VAR Lag-Order Selection Results

Table ‎0.6 – LNFCI and IMP/GDP: VAR Lag-Order Selection Results

Table ‎0.1 – LNFCI and LNGDP: Cointergration model Selection

Table ‎0.2 - LNFCI and LNGDP: Results of Johansen Cointegration Test

Table ‎0.3 – LNFCI and LNGDS: Cointegration Model Selection

Table ‎0.4 – LNFCI and LNGDS: Results of Johansen Cointegration Test

Table ‎0.5 – LNFCI and LNGCF: Cointegration model selection

Table ‎0.6 – LNFCI and LNGCF: Results of Johansen Cointegration Test

Table ‎0.7 – LNFCI and CAD/GDP: Cointegration Model Selection

Table ‎0.8 – LNFCI and CAD/GDP: Results of Johansen Cointegration Test

Table ‎0.9 – LNFCI and LNOER: Cointegration Model Selection

Table ‎0.10 – LNFCI and LNOER: Results of Johansen Cointegration Test

Table ‎0.11 – LNFCI and IMP/GDP: Cointegration Model Selection

Table ‎0.12 – LNFCI and IMP/GDP: Results of Johansen Cointegration Test

Table ‎0.1: LNFCI and LNGDP: VECM Estimates

Table ‎0.2: LNFCI and LNGDS: VECM Estimates

Table ‎0.3 – LNFCI and LNGCF: VECM Estimates

Table ‎0.4 – LNFCI and CAD/GDP: VECM Estimates

Table ‎0.5 – LNFCI and LNOER: VECM Estimates

Table ‎0.6 – LNFCI and IMP/GDP: VECM Estimates

Table ‎0.7 – LNGDS and LNGDP: VECM Estimates

List of Figures

Figure ‎2.1 - FCI and Macroeconomic Indicators: A Diagrammatic Representation of Variables Under-Study

Figure ‎2.2 - Measurement of FCI by Adding Up the Sub-Components

Figure 4.3: Savings-Investment Gap in Pakistan (1972-2006)

Figure 4.4: Import-Export Gap in Pakistan (1976-2006)

Figure ‎4.5: Foreign Capital Inflows and Current Account Deficit (1976-2005)

Figure ‎0.1: Composition of FCI in Pakistan (Average 1972-1983)

Figure .‎0.2: Composition of FCI in Pakistan (Average 1984-1995)

Figure ‎0.3: Composition of FCI in Pakistan (Average 1996-2006)

Figure ‎0.1: Correlogram of LNFCI at level

Figure ‎0.2 – Correlogram of LNFCI at 1st Diff

Figure ‎0.3 – Correlogram of LNGDP at level

Figure ‎0.4Correlogram of LNGDP at 1st Diff

Figure ‎0.5 – Correlogram of LNGDS at level

Figure ‎0.6 – Correlogram of LNGDS at 1st diff

Figure ‎0.7 – Correlogram of LNGCF at level

Figure ‎0.8 – Correlogram of LNGCF at 1st diff

Figure ‎0.9: Correlogram of CAD/GDP at level

Figure ‎0.10: Correlogram of CAD/GDP at 1st Difference

Figure ‎0.11: Correlogram of LNOER at level

Figure ‎0.12: Correlogram of LNOER at 1st difference

Figure‎0.13: Correlogram of IMP/GDP at level

Figure ‎0.14: Correlogram of IMP/GDP at 1st Difference

Figure ‎0.1 - FCI and GDP: Impulse Response Function

Figure ‎0.2 - FCI and Investment: Impulse Response Function

Figure ‎0.3 - FCI and Savings: Impulse Response Function

Figure ‎0.4 - FCI and CAD: Impulse Response Function

Figure ‎0.5 - FCI and OER: Impulse Response Function

Figure ‎0.6 - FCI and IMP: Impulse Response Function

Preface

All praises and thanks to Allah Almighty, Who bestowed me, the abilities, potential and opportunity to work on this book.

The Main objective of this book is to share the research findings of my M. Phil dissertation with the economists, economics students, economics scholars, faculty members and other interested readers. This book is mainly based on my M. Phil thesis and it is an edited and improved version of my Thesis.

This dissertation re-examines the macroeconomic role of Foreign Capital Inflows (FCI) in Pakistan applying vector error correction model (VECM) on annual time-series data for the period of 1972-2006. The topic of Foreign Capital Inflows to Pakistan got a lot of attention in empirical literature, but the existing literature on FCI about Pakistan mostly uses the traditional econometric tools like OLS, 2SLS, FIML and 3SLS for analysis. However, we know that most of the macroeconomic variables are non-stationary, which mandates the re-examination of the past studies using new time-series tools like cointegartion and ECM. Therefore, the present study not only re-examines these linkages (of FCI with domestic savings, investment and GDP) using Johansen Cointegartion Technique and Vector Error-Correction (VECM) Approach, but also adds some new macroeconomic variables like exchange rate, current account deficit and imports.

A large number people have supported and reinforced me in conducting this research, so, I would use this preface as an opportunity to thank all of them. First of all, I would like to acknowledge and appreciate the guidance and support of my all esteemed teachers, who cooperated and helped me to accomplish this task. It is a matter of immense pleasure to express my cordial gratitude to my supervisor, Ms. Nadia Saleem for her generous cooperation, suggestions, constant encouragement and kind supervision throughout this research work. I owe a special thanks to Prof. AsifSaeed, Chairperson Department of Economics, Prof. Hamid Dar and Prof. Salahuddin Sheikh.

Furthermore, I respectfully offer my thanks to my parents who devoted their lives for my disciplined training and always prayed for my success in every sphere of life. I would like to express my special gratitude for all of my family members, as without their encouragement, prayers and moral support the present study would might have been a mere dream.

Here I would also like to thank the whole staff of the Library of GCU for their cooperation. In addition, I also offer a special thanks to the Higher Education Commission (HEC) as well as the Librarian, Post Graduate Library, GCU for the provision of the access to the HEC digital library at Post Graduate Library.

I would like to thank all of my friends including Rashid Sattar, Qasim Ali, Sami Ullah and Nauman Aslam for their cooperation, suggestions and productive comments about my research topic. At last but not least, I owe a very special thanks to my lovely wife, Mrs. Ammara Mohey-ud-din, for her encouragement, moral support and sacrifice of time during adapting my thesis into this book format and for her prayers for my success in academic career.

Ghulam Mohey-ud-din

Executive Summary (Abstract)

This book re-examines the macroeconomic role of Foreign Capital Inflows (FCI) in Pakistan through applying vector error correction model (VECM) on annual time-series data for the period of 1972-2006.

The topic of Foreign Capital Inflows to Pakistan (and other developing countries) got a lot of consideration in empirical literature, but the existing literature on FCI about Pakistan mostly uses the customary econometric tools like OLS, 2SLS, FIML and 3SLS for analysis. However, we know that most of the macroeconomic variables are non-stationary, which mandates the re-examination of the past studies using new time-series tools like cointegartion and ECM. Thus, the current study not only re-examines these linkages (of FCI with domestic savings, investment and GDP) using Johansen Cointegartion Technique and Vector Error-Correction (VECM) Approach, but also adds some new macroeconomic variables like exchange rate, current account deficit and imports.

The present study does not find any evidence for direct positive impact of aggregate FCI on GDP growth and Investment (capital formation). However, the study finds the positive (complementary) relationship between FCI and domestic saving, thus suggesting an indirect positive impact of FCI on GDP through supplementing domestic resources. These results seem contradicting i.e. positive relation with domestic savings but negative linkages with investment and growth. However, we can interpret it as that FCI is supplementing the domestic resources and there is a need and justification for FCI in Pakistan due the shortage of domestic savings. But, these inflows of foreign capital are not transforming in the productive investment and thus not boosting economic growth. As this study shows that most of FCI are of non-investment (non FDI) type and are concentrated in the selected non-export-oriented and less-employment-generating sectors. In addition, the present study finds that exchange rate depreciation and current account deficit causes more inflows of foreign capital in Pakistan. While FCI also results in increasing the import of goods and services in Pakistan.

Subsequently, the present study suggests some policy recommendations like: (i) to target and identify the potential sectors for inviting the inflows of foreign capital, (ii) to change in composition of existing FCI, from non-FDI to FDI (investment) forms of FCI, (iii) to diversify the existing FCI from non-tradable and less job-oriented sectors to the tradable (export-oriented) and specially in agricultural-related manufacturing sectors, (iv) to mobilize the domestic resources, that will reduce the reliance on foreign assistance, and (v) to control the current account deficit and to stabilize the exchange rate, which will be helpful in reducing the reliance on foreign capital.

Chapter 1: Introduction

1.1 Introduction and Background

A man is poor because he is poor is an oft-quoted maxim which is also true in case of a poor country because mostly the poor countries (or underdeveloped countries) are entrapped in the ‘vicious circle of poverty’. They already lack the capital resources and at the same time the domestic saving ratios also remain low due to low income, resulting in low investment levels. Simultaneously, exports always remain lower than the imports in such countries. In such situations, the underdeveloped countries (UDCs) have to face saving-investment deficit as well as the deficit in balance of payments (BOP). The Two-Gap Model suggests that the developing countries have to rely on the foreign capital inflows (FCI) to fill these two gaps: the import-export gap and the saving-investment gap. Along with these government earnings of the UDCs governments, also remains lower because of low tax-paying capacity of poor people of the country. Thus these three deficits (saving deficit, fiscal deficit and BOP deficit) provide a sound rationale for the inflows of foreign capital in UDCs.

A large body of literature both empirical and theoretical suggests that Foreign Capital Inflows to poor countries are significant for a number of reasons. Foreign capital inflows, as discussed earlier, can finance required investment and stimulate economic growth. FCI may increase welfare by enabling households to smooth out their consumption over time and achieve higher levels of consumption. FCI helps in filling the saving-investment and export-import gaps. FCI also helps in structural transformation of economy through technology transfer. However, some other researchers found negative impacts of FCI, that is, large capital inflows can also have unfavourable macroeconomic effects, like rapid monetary expansion, higher inflationary pressures, and real exchange rate appreciation and widening current account deficits. FCI may also distort (substitute) the domestic savings. Hence, a surge in inflows of the foreign capital seen in recent years may create problems for economic policy, especially in the present environment of globalization and free capital mobility.

The literature, as discussed earlier, shows both positive as well as the negative aspects of FCI. Similarly, the existing literature about FCI on Pakistan also found the mixed (positive/negative) results. In Pakistan, the relationship of FCI with savings, investment and GDP got much attention in empirical literature, but the existing literature on FCI about Pakistan mostly uses the traditional econometric tools like OLS, 2SLS, FIML and 3SLS for analysis. However, we know that most of the macroeconomic variables are non-stationary, which mandates the re-examination of the past studies using new time-series tools like cointegartion and ECM. Therefore, the present study not only re-examines these linkages (of FCI with domestic savings, investment and GDP) using Johansen Cointegartion Technique and Vector Error-Correction (VECM) Approach, but also adds some new macroeconomic variables like exchange rate, current account deficit and imports. Accordingly the major objective of the present study is to analyse the macroeconomic effects of FCI in Pakistan. Thus we can define our research question as under:

1.2 Statement of Problem

The present study re-examines the relationship of FCI with GDP growth, domestic savings, investment and exchange rate in Pakistan. In addition, it also adds two other new macroeconomic variables like current account deficit and imports, that is, it also examines the impact of the FCI on current account deficit and imports in Pakistan. Thus, we can define our research problem as, “to find out the long-run causal relationship of FCI with the Macroeconomic Variables in Pakistan.” For this purpose, the present study constructs the six bivariate Vector Error-Correction models to find out long-run causal links between these variables. The major objectives of the study are given as under.

1.3 Objectives of the Study

The major objective of the study is to re-examine the macroeconomic role of foreign capital inflows in the economic development of Pakistan, using modern time-series econometric tools like cointegration and vector error-correction. The definite objectives of the present research are:

- To analyse the past trends, composition and sector-wise analysis of the Foreign Capital Inflows (FCI) in the recent past in Pakistan.
- To explore the linkages between Foreign Capital Inflows (FCI) and GDP Growth in Pakistan.
- To analyse the linkages between the Foreign Capital Inflows (FCI) and domestic savings in Pakistan.
- To examine the linkages between the Foreign Capital Inflows (FCI) and capital formation (investment) in Pakistan.
- To investigate the linkages between the Foreign Capital Inflows (FCI) and current account deficit in Pakistan.
- To study the linkages between the Foreign Capital Inflows (FCI) and exchange rate in Pakistan.
- To explore the linkages between the Foreign Capital Inflows (FCI) and the imports of goods and services in Pakistan.
- To suggest the policy recommendations for effective utilization and the management of foreign capital inflows in Pakistan.

1.4 Significant Contributions of the Study

The contribution and significance of the present study can be explained as under:

- There are so many form of the FCI but, according to the existing literature, some researchers included the only the FDI, and Aid , some used FDI, Aid and Portfolio Investment, some used only the FDI or Aid, some other used loan and grants only etc. While the present study uses a comprehensive definition of FCI and includes all form of FCI.
- According to the available literature on Pakistan, the relationship of the FCI with GDP, domestic savings and the investment is addressed a lot. While, the present study adds the linkages of FCI with additional macroeconomic variables like Current Account Deficit, Exchange Rate and Imports.
- All of the studies on Pakistan, except Ahmad and Ahmed[2002], use econometric tools like OLS, 2SLS, FIML or 3SLS to analyze the role of FCI in Pakistan. But, we know that most of the macroeconomic variables are non-stationary in nature, which mandates the re-examination of the past studies using new time-series tools like cointegration and ECM. Therefore, the present study uses the Johansen cointegration technique and vector error-correction model (VECM) for the macroeconomic analysis of the FCI in Pakistan.
- In addition to econometric analysis, the present study also presents the trends, composition, sector-wise distribution and need-based analysis of the FCI in Pakistan.

1.5 Organisation of the Study

The organisation of the study follows as; the next chapter (chapter 2) critically reviews the empirical literature on the foreign capital inflows, while the Chapter 3 describes the complete research design and methodology for econometric analysis. The Chapter 4 presents the trend, composition, sector-wise analysis and need-based analysis of the FCI in Pakistan. Chapter 5 gives the details of results and findings of the econometric estimations. While, the Chapter 6 concludes the study, presents the summary of findings and also suggests the policy recommendations.

Chapter 2: Review of Selected Literature

The relationship of FCI with domestic savings, investment, growth and other macroeconomic indicators has received much attention and a lot of research has been carried out both for and against the role of FCI in economic growth and development. The debate about the relationship of FCI with macroeconomic variables like domestic savings, investment, exchange rate and growth remained controversial both in theoretical and empirical terms. According to relevant literature, FCI may supplements domestic resources, promotes growth and reduces poverty of the recipient country. On the other hand, FCI may distort domestic savings, increase the debt burden of the recipient country, cause to appreciate the domestic currency and may create the boom in some sectors of the economy.

Therefore, the current chapter aimed at reviewing this controversy in the existing literature on FCI. The chapter is divided into five sub-sections; section 2.1 reviews the literature on FCI and Growth relationship, while section 2.2 reviews the literature on FCI and Domestic Savings relationship, section 2.3 reviews the literature on FCI and Investment relationship, section 2.4 reviews the literature on FCI and Exchange Rate relationship, and the last section (2.5) reviews various other aspects of FCI discussed in literature.

2.1 FCI and Economic Growth: A Review of Literature

The relationship between the FCI and Growth has debated a lot in the empirical literature. There is a controversy about this relationship; some studies found the positive relationship between FCI and GDP, while some others found a negative relationship.

A large body of literature found the positive relationship between FCI and Growth, for example, North[1956] argued that Long-term foreign capital played an important role in meeting the capital requirements during the periods of development. Similarly, some further studies [like Chenery and Strout (1966); cited by Mahmood (1997)] also argued that foreign economic assistance stimulate the economic growth and concluded, on the basis of empirical evidence that foreign capital has a positive effect on the economic growth. Chen[1977] applied simultaneous-equation (2SLS) model on panel data of the seven Asian countries between the domestic savings and capital inflows and found diverse results for the different countries.

Khan and Rahim[1993], using single-equation (OLS) model, in Pakistan during the period of 1960 to 1988 concluded that the aid has accelerated the rate of growth of GDP. Mahmood[1997] found that foreign aid has helped to promote the economic growth, helped in structural transformation of economy in Pakistan. Burnside and Dollar[2000] using panel growth regressions for 56 developing countries (1970-93) found that aid has a positive impact on growth in developing countries with good fiscal, monetary, and trade policies. Berthélemy and Démurger[2000] using model based on a sample of 24 Chinese provinces, from 1985 to 1996, found the fundamental role of foreign investment in provincial economic growth in China. Hansen and Tarp[2000] run a regression between per capita aid and the real GDP growth for cross-country data and found that aid increases the growth rate through increasing investment, and this result is not conditional on ‘good’ policy.

Bailliu[2000], using panel data for 40 developing countries from 1975–95, found that capital inflows promote higher economic growth, but only for economies having improved and developed financial (banking) sector. Reisen and Soto[2001], used panel data analysis covering 44 countries over the period 1986-1997, and suggested that developing countries should not solely depend on the national savings, but rather should encourage the foreign direct investments and portfolio equity inflows to stimulate the long-term growth prospects. Chowdhury and Mavrotas[2005] applied Toda-Yamamoto test for causality to study the direction of causality between FDI and GDP to time-series data of three developing countries. The study suggested that GDP causes the FDI in the case of Chile and not vice versa, while in case of Malaysia and Thailand, there is a strong support of bidirectional causality between FDI and GDP. Hatemi and Irandoust[2005], using panel data co-integration on cross-country panel data of six developing countries, and found that foreign aid has a positive and significant effect on economic activity. Kiong and Jomo[2005] applied OLS technique to the Malaysian data from 1966-1999 and found that FCI augmented domestic investment funds to accelerate the growth rate.

Yasmin[2005] applied the Simultaneous Equation Model on the aggregate time series data for the years 1970–71 to 2000–2001 for FCI, GNP and Savings in Pakistan and found a two-sided positive and statistically significant relationship between FCI and growth. Frimpong and Oteng-Abayie[2006] utilizing Toda-Yamamoto Granger no-causality test on annual time-series data of Ghana and found that FDI caused GDP growth during the post-SAP period. Baharumshah and Thanoon[2006] used dynamic generalized least square model (ECM/co-integration) on the panel data of growth process of eight the East Asian countries, including China and found that FDI is growth enhancing and its impact is felt both in the short and long run. Bhandari, Dhakal, Pradhan, and Upadhyaya[2007] developed an error-correction model is and estimated it by using a fixed-effects estimator. The findings of the study indicated that the inflow of foreign direct investment is significant factors that positively affect economic growth while, foreign aid did not seem to have any significant effect on real GDP.

The studies, discussed above, suggest that FCI has a positive relationship with GDP growth but some other studies argue against these and found the negative relationship between FCI and GDP. For example, Griffith-Jones[2000] found that the private capital flows to Latin America are much more volatile than capital flows to developed countries. This volatility of capital flows, passed on to volatility of other macroeconomic variables, is very negative for both growth and investment. Similarly, Lensink and Morrissey[2001] also found that volatility of FDI has negative impact on growth.

Parai[2003] re-examined implications of foreign capital inflow in a neoclassical model of economic growth and found that, the inflow of foreign capital induced through tax policy, could not be enough to achieve long run per capita growth. Chowdhury and Mavrotas[2005] examined the causal relationship between FDI and economic growth by using a Toda-Yamamoto test for causality and suggested that FDI didn’t cause GDP while GDP causes the FDI in the case of Chile. Burhop[2005] constructed bivariate level VARs, using (1960-1999) data of foreign aid, income per capita and investment for 45 developing countries and found no causal link between foreign aid and economic development. Kamalakanthan and Laurenceson[2005] found that foreign capital can almost not be considered as an important driver of income growth in case of contemporary China and India. Similarly, Carkovic and Levine[2005] used dynamic panel model with data averaged over five-year periods and re-evaluated the relationship between economic growth and FDI for a panel data of 72 countries (1960-95). Carkovic and Levine found that FDI inflows do not exert an independent influence on economic growth.

Likewise, Frimpong and Oteng-Abayie[2006] utilized Toda-Yamamoto Granger causality test and found no causality between FDI and growth for the total sample period and pre-SAP period in Ghana. Prasad, Rajan and Subramanian[2007] also found that there is no any evidence that an increase in foreign capital inflows directly boosts growth. Consequently, the literature remained inconclusive about the relationship between the FCI and GDP growth, most of the studies found positive relationship while some found negative linkages between FCI and GDP.

2.2 FCI and Domestic Savings: A Review of Literature

The relationship between the FCI and domestic savings has been discussed a lot in the literature. There is also a controversy about this relationship, some studies found the positive relationship between FCI and domestic savings, while some other studies found a negative relationship.

Some studies found a positive (complementary) relationship between FCI and domestic savings, like, Islam[1972] found that foreign assistance helped to attain a higher rate of savings and investment in Pakistan. Similarly, Papanek[1972] found diverse results; in some circumstances, foreign inflows undoubtedly stimulated savings, while in other cases, foreign inflows discouraged savings. Aslam[1987], using single-equation regression model (OLS) for the data from 1963-64 to 1984-85 of Pakistan found that the private FCI covered the domestic saving-investment gap.

Yasmin[2005] applied the Simultaneous Equation Model on the aggregate time series data for the years 1970–71 to 2000–2001 for FCI, GNP and Savings in Pakistan and found that FCI helps in filling-in the gap between savings and investment. Moreover, saving has appeared to be a complement of FDI. Hatemi and Irandoust[2005] used a panel data co-integration over cross-country panel data of six developing countries and found that foreign capital flows supplemented the domestic savings.

The studies, discussed above, show the positive or complementary relationship between FCI and savings but there are various other studies who find the negative (substitution) relationship between these variables. For example, [Leff (1969), Griffin (1970) and Weisskopf (1972); cited by Mahmood (1997)], argued that the foreign capital could badly affect the economic growth by substituting the domestic savings. Shabbir and Mahmood[1992] applied simultaneous equation model (2SLS) on the data of Pakistan, ranging from 1959-60 to 1987-88, a negative relationship between the savings and foreign investment and loans & grants. Furthermore, foreign financial inflows may discourage domestic (household or private) and public savings.

Khan, Hasan and Malik[1992] using single-equation (OLS) model, in Pakistan during the period of 1959-60 to 1987-88, found that foreign capital inflows have depressing effect on the national savings. Khan and Rahim[1993], using single-equation (OLS) model, in Pakistan during the period of 1960 to 1988 found negative relationship between foreign aid and domestic resource (savings). Mahmood[1997] concluded that foreign aid may distort domestic savings and may create booming sector. Ahmad and Ahmed[2002] used cointegartion and error-correction model on Pakistan’s data and found that the foreign capital may substitute for the domestic savings.

Paul and Sakthivel[2002], using Johansen’s Maximum Likelihood Tests for co-integration and error correction model on time-series data for 50 years (1950-2000) of India, suggested that foreign capital is negatively related to domestic savings. Kiong and Jomo[2005] applied OLS technique to the Malaysian data from 1966-1999 and found that FCI had negative influences on the savings rate as well as on the balance of payments. Whereas Areskoug[1976], Chen[1977], Reinhart and Talvi[1998] have found diverse (mixed) result from a panel of countries. In some countries, FCI complements (crowds in) savings and in others, it substitutes (crowds out) the domestic savings. Thus the review of the above studies shows a diverse relationship between FCI and savings, some studies found complementary relationship and some found that FCI substitutes the domestic savings.

2.3 FCI and Investment: A Review of Literature

The existing literature on the relationship between the FCI and Investment shows both positive as well as negative links. As most of the studies found positive relationship like North[1956] argued that Long-term foreign capital played an important role in meeting the capital requirements by directing real resources towards the needed social overhead investment and making possible an import surplus of consumer and capital goods during the periods of development. Similarly, Islam[1972] found that foreign assistance helped to attain a higher rate of savings and investment in Pakistan.

Walt and Wets[1993], by employing both a single equation regression and macro-econometric model, found that foreign capital inflow complemented domestic resources to finance investment and generating growth. Khan[1993] concluded that foreign aid has played an extremely important role in influencing the pace of development in Pakistan, especially investments and imports largely depended upon the amount of foreign aid. Bosworth, Collins and Reinhart[1999] applied a regression analysis on sample of developing economies andfound that that FDI is strongly linked to aggregate investment, and had a more positive impact, on domestic savings and investments, than any other form of FCI like loans, portfolio investment and borrowings. Similarly, Hansen and Tarp[2000] run a regression and found that aid continues to impact on growth via investment.

Ulengin and Yentürk[2001] applying the vector autoregressive (VAR) models found that foreign savings has an increasing effect on consumption and an increase of investment arises from the accelerator effect of consumption, which results in an upward trend in investment in non-tradable sectors. Verma and Wilson[2004], using ARDL co-integration technique and Granger causality test on time-series data of India for the period of 1950-2000, found that there are significant and complicated relationships between the components measures of savings, investment and foreign inflows. Similarly, Kiong and Jomo[2005] found that FCI augmented domestic investment funds to accelerate the growth rate.

The above studies show a positive relationship between FCI and investment but some other studies found negative relationship. Like, Griffith-Jones[2000] found that volatility of capital flows is very negative for both growth and investment. Similarly, Burhop[2005] claimed that there is no causal link between foreign aid and economic development, measured by income per capita and investment. Aslam[1987] also found that the public forms of FCI did not affect the domestic investment significantly, while the private forms of FCI covered the domestic saving-investment gap. Areskoug[1976] also found that FCI partially substitute role in capital formation in developing countries. Thus, after reviewing literature, we come to know that most of the studies found the positive relationship between FCI and Investment while, some studies also found negative relationship.

2.4 FCI and Exchange Rate: A Review of Literature

In recent past, the relationship between exchange rate and FCI got some attention and a few researchers addressed this issue. Like, Ag´enor[1998] found that a permanent fall in world interest rate leads to a steady-state reduction in net capital inflows and a real depreciation. Ag´enor also found that the real exchange rate appreciates in the net debtor case, but may either appreciate or depreciate in the net creditor case. Kosteletou and Liargovas[2000], applying causality tests on data for EU countries as well as for the USA and Japan over the period 1960–1997, found that in large countries, with freely floating currencies such as the USA, the UK and Japan, causality runs from the real exchange rate to FDI, while Causality runs two-ways in small countries with fixed or quasi-fixed currencies, such as the EU countries.

Mafusire[2003] developed a computable general equilibrium (CGE) model for Zimbabwe and found that foreign capital inflows resulted in the appreciation of the exchange rate. Athukorala and Rajapatirana[2003] found that the degree of appreciation in real exchange rate associated with capital inflow is uniformly much higher in Latin American countries as compared to Asian countries, despite the fact that the latter experienced far greater foreign capital inflows relative to the size of the economy. Thus, according to the most of literature the causality runs from FCI to exchange rate, that is, FCI causes exchange rate appreciation. Whereas, the Kosteletou and Liargovas[2000] found that, in small countries with fixed or quasi-fixed currencies, causality may run in both directions, i.e. exchange rate may also cause the FCI.

2.5 Some Other Aspects of FCI: A Review of Literature

The studies discussed above explain the relationship of FCI with domestic savings, investment, economic growth and exchange rate, but some studies also find the other aspects of FCI in literature. Thus, the present section presents the mixed literature on FCI.

Hong[1997] quantified the contributions by various types of foreign capital inflows towards the growth of individual Korean industries during the last 20 years (1990-1990) and suggested that the foreign capitals played a vital role in the success of Korea's manufacturing sector, which served as the engine of economic growth. FDI had more statistically significant effects on the factor productivity than either commercial or public loans. Marjit, Broll and Mitra[1997] using a model with trade and unemployment found that the inflows of foreign capital deliver the expected results when it inflows to protected intermediate-goods sector.

Khan[1998] concluded that, if the countries implement the necessary macroeconomic policies and realize the structural transformations that build up their financial system that will not only be helpful in managing the capital inflows but will also reduce the risks of the reversal of these capital flows. Sin and Leung[2001] used panel data models both with random-effect and fixed-effect models and found that the policy change has an encouraging effect on FDI and confirmed that the governments are successful in absorbing foreign capital inflows through more liberal policies.

Employing the neoclassical growth model, Chow and Zeng[2001] found that an increase in foreign capital inflows in a developing economy would lead to a higher level of domestic capital stock and consumption. Kohli[2004] concluded that despite the fact that the shift in external financing from aid transfers to private capital flows has increased the accessibility of external resources to the Indian economy, but it has also imposed greater restraint through the increased vulnerability to unfavourable capital account shocks, volatility and other risks. Kohli also pointed out the significance of self-protection policies that countries must adopt in order to tone down the risks.

Paul and Truong[2004] showed that foreign capital inflow could create long-term economic benefits to the host country only when the foreign capital reinvestment rate is adequately greater than the host country’s saving rate. Kaminsky[2005] concluded that, there is no optimal policy to manage the risks of volatility in international capital flow, because the policies that may help in the short-run may have undesirable effects in long run. Sikdar[2006] suggested a Reduction in non-FDI flows will reduce the need for large unproductive reserves and investment in infrastructure in general (and not only in telecom) must increase to boost domestic investment and attract FDI in manufacturing. These studies reviewed above highlighted various positive and negative aspects of foreign capital inflows.

Ultimately, after reviewing the all the available economic literature on FCI, it becomes obvious that most of the studies, mentioned above, have found a mixed (positive/negative) relationship between FCI and other macroeconomic variables.

Chapter 3: Research Design

This chapter describes the complete research design and procedure for empirical analysis. However, before going to the empirical investigation of the linkage among FCI, Savings, Investments, Economic Growth, Current Account Deficit and Exchange Rate it will be notable to review the some basic theoretical framework for linkage between FCI and Macroeconomic Indicators. This framework will help in modelling the relationship between the variables. Thus, in this chapter, first of all, we discuss a theoretical framework in the section 3.1. The next section 3.2 formulates the hypothesis. Section 3.3 describes the data specifications, the variable description and sample period, section 3.4 provides the information about the model specification and the last section (3.5) gives the procedure for estimation (methodology) of the models formulated in section 3.4.

3.1 Foreign Capital Inflows: A Theoretical Framework

Theoretically, the major approaches to the relationship between FCI and growth in developing countries are identified as Neo-classical, Dependencia, Bargaining and Structuralist schools of thoughts. According to the first (neo-classical) approach FCI is good for the growth of an economy. In contrast with the Neo-classical approach the Dependencia School emphasizes the risk and negative effects of FCI on growth and development. The proponents of the Bargaining approach argue that the distributions of gains from FCI are generated by bargaining and negotiation between foreign investors and developing countries. The Structuralist school challenges the Bargaining school’s relative optimism about long-term negotiations. By and large, the bargaining power declines with the passage of time [Yasmin (2005)]. Despite of the other three schools of thought, the neo-classical theory is still considered most important in the literature to model FCI linkage with other macroeconomic variables.

If we start from Adam Smith, who also stressed the importance of the law of capital accumulation as the primary factor, contributing to economic progress or, as he termed it, to the “wealth of nations” [Cypher and Dietz (2005): p. 106]. Similarly, neo-classical (Solow-type) model also suggests that the developing countries, to increase their standard of living, should increase the rate of saving and investment, that is, by accumulating the physical capital at higher rate[ ibid, p: 122].

Rosenstein-Rodan emphasizes the need for creating a “big push” of investment simultaneously in a number of branches of industry, especially in social overhead capital [ ibid, p. 132].Similarly, Walt Whitman Rostow suggests a rise in the rate of productive investment from lower level to over 10 percent or higher of national income to enter in ‘take off’ stage [ ibid, p. 150]. The new growth theory, unlike the Solow model, explains technical change as an endogenous outcome of the public and private investment in human capital [Todaro (2001); p. 100-101]. Thus, the endogenous growth model suggests an active role of direct and indirect foreign private investment in knowledge-intensive industries [ ibid, p. 102].

The all theories discussed above, underlines the need of higher rates of capital accumulation but if we apply this on a developing country, which is already entrapped in the ‘vicious circle of poverty’ and facing the two-gaps, there will be a need of foreign capital.

But this represents the one side of the coin; if we look at the other side the ‘Dependecia School of Thought ’ shows the possible risk and negative effects of FCI on growth and development. For example, the neoclassical dependence model attributes the existence and persistence of underdevelopment largely to highly unequal capitalist system rich-country to poor-country relationship. Whether because the unequal power relationship between the Center (Rich countries) and Periphery (Poor countries) makes, the attempts by the poor nations, to be independent and self-reliant, difficult or even impossible [ ibid, p. 91].

Similarly, the false-paradigm model attributes underdevelopment to the faulty and inappropriate policy-advice provided by the international experts, who are biased and uninformed about the economy of the poor countries [ ibid, p. 92]. Thus the dependence and false-paradigm models stresses that the foreign assistance may result in the dependence on the developed countries as well as in exploitation by these countries. In addition, the technical assistance and policy advice by the developed nations are not applicable in developing countries.

Consequently, after reviewing the theoretical literature as well as the empirical literature, discussed in previous chapter, we found that the FCI have both positive as well as the negative impacts on the economy. On the positive side, the FCI may be helpful in filling the two-gaps, generation of additional savings, transfer of technology, structural transformation, and ultimately may result in the economic growth and development. On the negative side, the FCI may result in the distortion (substitution) of the domestic savings, vulnerability to external shocks, increase in debt burden, increase in foreign dependence, appreciation of exchange rate, and increase in imports and current account deficit.

Now after a reviewing a detailed empirical literature (in chapter 2) as well as theories of development and growth (in this chapter) and defining our research problem, we can develop a theoretical framework for the present study. This theoretical framework enables us to understand the network of relationship between our variables. As, we are interested in finding out an empirical evidence of the relationship of FCI with GDP Growth, Domestic Savings, Capital Formation (Investment), Current Account Deficit, Exchange Rate and Imports of Goods and Services. Since, the present study is aimed at finding the long run causal (unidirectional or bidirectional) relationship between these variables, the relationship between these variable, on the basis of the theoretical and empirical literature may explained by the figure 3.1 as under;

illustration not visible in this excerpt

Figure ‎2.1 - FCI and Macroeconomic Indicators: A Diagrammatic Representation of Variables Under-Study

Abbildung in dieser Leseprobe nicht enthalten

3.2 Hypotheses to be tested

In the light of the objectives, stated earlier, and the theoretical framework presented above, the present study aimed at finding out the long-run causal relationship of The Foreign Capital Inflows with the macroeconomic development (measured by six macro indicators like GDP, Savings, Investment, Current Account Deficit, Exchange rate and Imports) in Pakistan. Thus the specific hypotheses are given as under:

I – Foreign capital inflows contributes positively to GDP growth in Pakistan
II – Foreign capital inflows enhances the capital formation (investment) in Pakistan.
III – Foreign capital inflows have some positive relationship with domestic savings in Pakistan.
IV – Foreign capital inflow fills the current account deficit in Pakistan.
V – Foreign capital inflow appreciates the foreign exchange rate in Pakistan.
VI – Foreign capital inflow increases imports of goods and services in Pakistan.

3.3 Data Specifications

3.3.1 Types and Measurement of FCI

There are many forms of the FCI but theoretically, the FCI can two main forms i.e.

a) Official Capital Flows (OCF) and

b) Private Capital Flows (PCF).

These OCF and PCF are measured by summing-up the sub-components of FCI as defined by the OECD Glossary of Statistical Terms. ¹

Figure ‎2.2 - Measurement of FCI by Adding Up the Sub-Components

illustration not visible in this excerpt

The official capital flows (OCF) are further divided into two sub-forms: (a) Official development assistance

and (b) other official flows.

(a) Official Development Assistance and Official Aid (ODA), which includes bilateral and multilateral concessional (and/or developmental) loans & grants etc.

(b) Other Official Flows (OOF), which includes the bilateral and multilateral non-concessional (and/or non-developmental) loans and grants.

Similarly, the Private Capital Flows (PCF) are further divided into two sub-components:

a) Foreign Direct Investment (FDI), which comes in the form of direct investment in the host country.

b) Other Private Flows (OPF), which includes the portfolio (bonds and equity) investments and other private commercial loans and credit.

The present study measures the FCI as the sum of all these four sub-components of FCI, which is shown in the following figure 3.2.²

3.3.2 Variables Description

A brief description the variables, used in the study, are given as under:³

illustration not visible in this excerpt

3.3.3 Sources of the Data and Sample Period

The annual time-series data on all the components of FCI is taken from the “Organization for Economic Cooperation and Development” (OECD)’s Statistical Database (OECD.Stat)⁴, while, the rest of the variables are taken from the World Bank’s World Development Indicators (WDI)⁵, for the period of 1972-2006.

The sample period is selected on basis of data availability because the data on some variable, used in the study, is not available for the years beyond 1968 from any source. Accordingly, the data is available for period from 1968 to 2006 from the said sources but the data for the years 1968-71 is skipped deliberately because the data from 1968 to 1971 includes the data before the separation of Pakistan i.e. the data on West Pakistan and East Pakistan (Bangladesh). Thus to avoid the issue of structural break in the data the sample period is reduced to 1972-2006.⁶

3.4 Model Specifications

A Large number of empirical studies were undertaken to analyse the linkages of the foreign capital inflows with domestic savings, capital formation (investment) and GDP Growth. However, in these studies, as discussed in earlier sections, most of the researchers used various econometric techniques like single-equation regression models [OLS], simultaneous equation regression models [2SLS, FIML, 3SLS] and vector autoregressive [VAR] models. However, we know that almost all of the macroeconomics variable like GDP, exchange rate, trade flows and price level etc. are always non-stationary in nature. So according to the recent research and growing literature these studies should be re-examined using appropriate modern techniques like cointegration and error-correction models (ECM) [Greene (2006)].

There are two approaches for testing the cointegration, these are, Engle-Granger technique & simple ECM, and Johansen Cointegartion & Vector Error-Correction Model (VECM). The Johansen Cointegration and VECM are estimated within VAR framework, which has an advantage that It captures the endogenous / simultaneous effect between the various time-series, because it treats all the variables as endogenous (expect for any exogenous variable, specified in model)

Therefore, the present study uses the Johansen cointegration technique and vector error-correction (VEC) model to re-examine the long run relationship of FCI with savings, investment, GDP growth, current account deficit, exchange rate and imports as well as the long run causal (unidirectional or bidirectional) relationship among these variables. So, this study uses following six different vector error-correction models for empirical analysis. The all six models are the open economy model, which is, based on the assumption of the open market (free flow of capital). To incorporate this assumption (of international capital mobility), an additional variable “Openness” (OPEN) is added as the exogenous variable in all models.⁷ The openness (OPEN) is proxied by the volume of trade (exports + imports).

3.4.1 FCI and GDP: Vector Error-Correction Model (3.1)

The dynamic relationship between the FCI and the GDP Growth is modelled as a vector error-correction model (VECM) in the following form (model 3.1):

illustration not visible in this excerpt

3.4.3 FCI and Savings: Vector Error-Correction Model (3.3)

The dynamic relationship between the FCI and the domestic savings is modelled as a vector error-correction model in the following form (model 3.3):

illustration not visible in this excerpt

illustration not visible in this excerpt

Where;

[Abbildung in dieser Leseprobe nicht enthalten]= 1st Difference Operator

3.4.4 FCI and CAD: Vector Error-Correction Model (3.4)

The dynamic relationship between the FCI and the Current Account Deficit (CAD) is modelled as a vector error-correction model in the following form (model 3.4):

illustration not visible in this excerpt

illustration not visible in this excerpt

3.4.5 FCI and Exchange Rate: Vector Error-Correction Model (3.5)

The dynamic relationship between the FCI and the Exchange Rate (OER) is modelled as a vector error-correction model in the following form (model 3.5):

illustration not visible in this excerpt

[...]

---

¹ OECD.[2008a]. OECD Glossary of Statistical Terms. Organization for Economic Co-operation and Development.<http://stats.oecd.org/glossary/index.htm>.

² The detailed definitions of these sub-components of FCI are given in Annexure A, See Table A.5

³ The detailed definitions of all these variables are given in Annexure A, See Table A.5.

⁴ OECD.[2008b]. OECD.Stat, Online Edition. Organization for Economic Cooperation and Development, <http://stats.oecd.org/wbos/Default.aspx>

⁵ World Bank[2008]>. World Development Indicators, Online Edition. Accessed through Global Development Network (GDN) <http://www.gdnet.org/proxy/wdi.html>

⁶ All the Data sets, used in the present study, are given in Annexure A, See Table A.1 and Table A.2.

⁷ Khan, Hasan and Malik[1992] used OPEN (proxied by Trade) as explanatory variable, similarly, Athukorala and Rajapatirana[2003] also used OPEN (Openness) as an explanatory variable.