Foreign direct investment, financial development and economic growth in asian developing countries

INSTUTION OF SOCIAL STUDIES UNIVERSITY OF ECONOMICS HO CHI MINH CITY THE HAGUE VIETNAM THE NETHERLANDS VIETNAM- NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS FOREIGN DIRECT INVESTMENT, FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH IN ASIAN DEVELOPING COUNTRIES A thesis submitted in partial fulfilment of the requirements for the degree of MASTER OF ARTS IN DEVELOPMENT ECONOMICS By NGUYEN THUC DUY ANH Academic Supervisor: DR NGUYEN VAN PHUC HO CHI MINH CITY, DECEMBER 2011 CERTIFICATION I hereby certify that the substance of the thesis has not already been submitted for any degree and is not being currently submitted for any other degree I also certify that, to the best of my knowledge, and help received in preparing the thesis and all sources used have been acknowledged in the thesis Signature Nguyen Thuc Duy Anh Date: ACKNOWLEDGMENTS I would like to express my gratitude to all those who gave me the possibility to complete this thesis I am deeply grateful to my supervisor Dr Nguyen Van Phuc from Ho Chi Minh City Open University whose support, stimulating suggestions and encouragement • helped me in all the time of research for and writing of this thesis I am also very grateful to all lecturers of the Vietnam-Netherlands Programme for giving me knowledge and guidance to fulfill the M.A Programme I would like to thank all the members of the Vietnam-Netherlands Program, especially, VNP Library for helping me to have necessary documents and research papers during my completion of the thesis Finally, I am indebted to my parents and my dear husband whose love, sympathy and encouragement enabled me to complete this thesis I am also thankful to my classmates for their warm encouragement ; 11 ACRONYMS AND ABBREVIATIONS • FDI Foreign Direct Investment FINANCE Financial Development WDI World Development Indicators WGI Worldwide Governance Indicators UNCTAD United Nations Conference on Trade and Development OECD Organization for Economic Cooperation and Development • MNCs Multinational Enterprises M&A Mergers and Acquisitions R&D Research and Development POOLED Pooled Regression Model FEM Fixed Effects Regression Model REM Random Effects Regression Model OLS Ordinary Least Square • 111 ABSTRACT This research investigates the role of domestic financial development plays in enhancing FDI's positive effects on economic growth in Asian developing countries In other words, we examine whether countries with better domestic financial system can utilize FDI more efficiently The empirical analysis uses balanced panel data of24 Asian developing countries in the period 1995-2009 This research applies the various models and techniques in panel data regression Linear static models for panel data named constant coefficients model or pooled regression model (POOLED), fixed effects regression model (FEM) and random effects regression model (REM) are employed We analyze all models and employ many kinds of test including poolability test, Hausman test, LM test, fixed effects tests and Wald tests to select the most appropriated estimated model The research findings show that FDI alone does not have directly effect on economic growth but does have when combined with financial development Well-developed domestic financial markets promote the process of technological diffusion associated with FDI in Asian developing countries Therefore, FDI and domestic financial development are complementary in increasing the rate of economic growth in the region There is a threshold level of domestic financial development above which FDI starts to have positive impacts on economic growth • lV I - TABLE OF CONTENTS CHAPTER INTRODUCTION 1.1 Problem statement 1.2 Research objectives 1.3 Research questions 1.4 Research scope and data 1.5 Organization ofthe thesis CHAPTER LITERATURE REVIEW 2.1 The impact ofFDI on economic growth 2.2 FDI, financial development and economic growth: theoretical framework and • empirical studies 10 2.2.1 Theoretical framework 10 2.2.1.1 Neoclassical growth model 10 2.2.1.2 Endogenous growth model 11 2.2.1.3 Technological change model 11 2.2.2 Empirical studies about the role of financial development in FDIGrowth nexus 15 2.3 The measures of financial development 21 2.4 The determinants of economic growth 23 Chapter remark CHAPTER RESEARCH METHODOLOGY 27 • 3.1 Variables measurements 27 3.1.1 Dependent variable: Economic growth (GROWTH) 27 v - - 1.2 Independent variables 27 3.1.3 Control variables (CONTROLS) 31 3.2 Data collection 33 3.3 The estimation strategy 34 3.3 Panel regression models 3.3.1.1 Constant coefficients model (POOLED) 35 3.3 1.2 Fixed effects regression model (FEM) 3.3.1.3 Random effects regression model (REM) 37 3.3.2 Choice of panel regression model (Testing panel models) 38 3.3.2.1 The F-Test for individual and time effects in FEM models: 38 3.3.2.2 The Chow Test for the POOLED against the FEM (Test for poolability): 39 3.3.2.3 The Hausman Specification Test for the FEM against the REM 39 3.3.2.4 The Breusch-Pagan (1980)- Lagrange Multiplier (LM) Test for the POOLED against the REM 40 3.3.3 Testing assumptions of the error term 41 3.3.4 Model specification 41 3.4 Summary of the analytical methods 43 3.5 Hypothesis statements 43 CHAPTER RESEARCH RESULTS 45 4.1 The descriptive statistics of the sample 45 4.2 Empirical results 46 4.2.1 Examining the impact ofFDI, on its own, on economic growth 46 4.2.1.1 Choosing the appropriate model 47 Vl 4.2.1.2 Test for heteroskedasticity 48 4.2.1.3 Results discussion 50 4.2.2 Examining the role of financial development in FDI-Growth nexus 51 4.2.2.1 Choosing the appropriate model 52 4.2.2.2 Test for heteroskedasticity 53 4.2.2.3 Results discussion 55 4.3 Chapter remark 62 CHAPTER CONCLUSION AND RECOMMENDATION 64 5.1 Conclusion 64 5.2 Recommendation 66 5.3 Research limitation and suggestion for further study 68 REFERENCES 70 APPENDIX A LIST OF ASIAN DEVELOPING COUNTRIES A-1 APPENDIX B DATA SOURCES AND DESCRIPTIVE STATISTICS B-2 APPENDIX C PANEL MODEL TESTS C-1 APPENDIX D REGRESSION MODEL RESULTS D-11 Vll LIST OF TABLES Table 3.5-1: Summary of expected signs of all variables .44 Table 4.2-1: Summary of testing panel model results for model (4.1) 47 Table 4.2-2: Summary the result of Testing Heteroskedasticity for Model (4.1) 48 Table 4.2-3: REM regression model (4.1) result .49 Table 4.2-4: Summary of testing panel model results for model (4 2) 52 Table 4.2-5: Summary the results of Testing Heteroskedasticity 53 Table 4.2-6: Two-way FEM regression model results 54 Table 4.2-7: The Wald Test on joint significance ofPOP 55 Table 4.2-8: The Wald Test on joint significance ofEDUC 56 Table 4.2-9: Final two-way FEM regression model results 57 Table 4.2-10: Asian developing countries have financial system reaching a sufficient level of development so that FDI has positive effect on economic growth 62 Table A-1: List of 24 Asian developing countries in data sample A-1 Table B-1: Variables definitions and Data sources B-2 Table B-2: Summary statistics of GROWTH B-3 Table B-3: Summary statistics ofFDI B-3 Table B-4: Summary statistics of Financial Development Indicators B-4 Table B-5: Summary Statistics ofFDixFINANCE B-4 Table B-6: Summary statistics of control variables B-5 Table B-7: Correlation matrix of the explanatory variables for GROWTH B-5 vm ·· LIST OF FIGURES Figure 3.4-1: Steps ofthe analytical methods 43 Figure C-1: The POOLED model1 C-1 Figure C-2: The FEM model1 C-1 Figure C-3: The REM model1 C-2 Figure C-4: The POOLED model 2a with LLY used as financial development indicator C-2 Figure C-5: The POOLED model 2b with PRIVATECREDIT used as financial development indicator C-3 Figure C-6: The FEM model2a with LL Y used as financial development indicator C-3 Figure C-7: The FEM model 2b with PRIVATECREDIT used as financial development indicator C-4 Figure C-8: The REM model2a with LL Y used as financial development indicator C-4 Figure C-9: The REM model 2b with PRVATECREDIT used as financial development indicator C-5 Figure C-1 0: Hausman Test for FEM and REM of model C-7 Figure C-11: Hausman Test for FEM and REM ofmodel2a- using LLY C-8 Figure C-12: Hausman Test for FEM and REM of model 2b with PRIVATECREDIT C-8 Figure C-13: Test for cross-section fixed effects in the FEM model 2b - using PRIVATECREDIT C-9 Figure C-14: Test for cross-section fixed effects in the FEM model2a- using LLY C-9 Figure C-15: Test for time fixed effects in the FEM model2a- using LLY C-10 IX Figure C-5: The POOLED model 2b with PRIVATECREDIT used as financial development indicator Source ss df MS Model Residual 1766.82905 5873.72233 11 348 160.620823 16.8785124 Total 7640.55138 359 21.2828729 GROWTH Coef Std Err FDI 1036667 -.0376755 002469 -.4527161 1819173 9155976 -.1467556 -1.056212 -.0285924 0132741 8277557 3.213051 0607473 0126789 0015272 3710273 0356748 5386704 0634403 281296 0064249 0178878 7461389 3.471162 PRIVATECRE~T FDixPRIVAT~T iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF _cons Number of obs F( 11, 348) Prob > F R-squar·ed Adj R-squared Root MSE t 0.089 0.003 0.107 0.223 0.000 0.090 0.021 0.000 0.000 0.459 0.268 0.355 360 9.52 0.0000 = = 0.2312 = 0.2069 = 4.1083 [95% Conf Interval] P>ltl 1.71 -2.97 1.62 -1.22 5.10 1.70 -2.31 -3.75 -4.45 0.74 1.11 0.93 = = -.0158113 -.0626125 -.0005347 -1.182454 111752 -.1438616 -.2715302 -1.609467 -.0412289 -.0219077 -.6397536 -3.614044 2231448 -.0127385 0054727 277022 2520827 1.975057 -.0219809 -.5029582 -.0159559 048456 2.295265 10.04015 Figure C-6: The FEM model 2a with LL Y used as financial development indicator Fixed-effects (within) regression Gr·oup variable: country R-sq: Number· of obs Number of groups within = 0.2774 between = 0.0270 overall = 0.0721 corr·(u_i, Xb) = = 11.341 0.0000 F (11, 325) Prob > F GROWTH Coef Std Er·r FDI LLY FDixLLY iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF _cons -.2538103 -.0756504 0047607 -3.159241 0902749 7.21494 -.2604997 4604176 -.0407632 0695304 2.614612 -3.761044 0610522 024399 0009831 1.367625 0420891 1.275207 1039408 5524606 0059672 0294632 1.13188 8.705588 sigma_u sigma_e rho 4.7721797 5488675 64390399 (fr·acti on of val'i ance due to u_ i) F test that all u_i=O: F(23, 325) = C-3 360 24 15 15.0 15 = -0.7866 t P>itl -4.16 -3.10 4.84 -2.31 2.14 5.66 -2.51 0.83 -6.83 2.36 2.31 -0.43 = = obs per group: = avg = max = 0.000 0.002 0.000 0.022 0.033 0.000 0.013 0.405 0.000 0.019 0.022 0.666 6.29 [95% Conf Interval] -.3739177 -.1236503 0028266 -5.849756 0074734 4.706239 -.4649814 -.6264327 -.0525023 0115677 3878767 -20.88746 Prob > F -.1337028 -.0276504 0066947 -.4687248 1730764 9.723641 -.0560179 547268 -.0290241 127493 4.841348 13.36537 = 0.0000 ··!1-''"""" , ''''"" -··· ··· -· ···-,·· ·· ·· ~···-··-·-· R-sq: within = 0.2790 between = 0.0193 overall = 0.0721 corr(u_i, Xb) = ·-·-,-···-··- ··· · = 360 24 Obs per group: = avg = max = 15 15.0 15 Number of obs Number of groups Fixed-effects (within) regression Group variable: country F (11, 325) Prob > F -0.7620 = [95% Conf Interval] Coef Std Err FDI iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF _cons -.2049658 -.0929483 0063478 -3.320912 1320121 6.876593 -.2867758 421572 -.0393022 0603487 2.70417 -1.65494 0624349 0213825 0015613 1.281806 0426741 1.250158 1034691 5518614 0059961 0287075 1.081869 8.004326 sigma_u sigma_e rho 4.5572391 3.5450764 6230031 (fraction of variance due to u_i) FDixPRIVAT~T F test that all u_i=O: F(23, 325) t P>ltl -3.28 -4.35 4.07 -2.59 3.09 5.50 -2.77 0.76 -6.55 2.10 2.50 -0.21 0.001 0.000 0.000 0.010 0.002 0.000 0.006 0.445 0.000 0.036 0.013 0.836 -.3277933 -.135014 0032764 -5.842596 0480597 4.41717 -.4903294 -.6640996 -.0510984 0038728 5758198 -17.40177 6.19 = 11.43 0.0000 = GROWTH PRIVATECRE~T = -.0821382 i -.0508827 0094193 -.7992284 2159645 9.336016 -.0832221 1.507243 -.0275061 1168246 4.83252 14.09189 Prob > F = 0.0000 Figure C-8: The REM model 2a with LL Y used as financial development indicator Random-effects GLS regression Group variable: country R-sq: Number of obs Number of groups within = 0.1941 between = 0.2872 overall = 0.2116 corr(u_i, X) GROWTH = 360 24 Obs per group: = avg = max = 15 15.0 15 wald chi2(11) Prob > chi2 (assumed) Coef = = Std Err z 0.530 0.003 0.002 0.192 0.000 0.061 0.021 0.031 0.000 0.110 0.289 0.388 88.47 0.0000 [95% Conf Interval] FDI LLY FDixLLY iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF _cons -.0377077 -.0366365 0029801 -.6554857 1450978 1.198514 -.1703203 -.7185275 -.0329545 0314662 8192321 3.588859 0600834 0122408 0009537 5021886 0358732 6394918 0738955 3323059 0062887 0196696 7728631 4.159343 sigma_u sigma_e rho 93168516 3.5488675 06447817 (fraction of variance due to u_i) C-4 -0.63 -2.99 3.12 -1.31 4.04 L87 -2.30 -2.16 -5.24 L60 L06 0.86 P>lzl = = -.1554691 -.0606282 0011109 -1.639757 0747877 -.0548667 -.3151529 -1.369835 -.0452801 -.0070855 -.6955517 -4.563304 0800536 -.012644g 0048494 328786 2154079 2.451895 -.0254878 -.0672199 -.020629 0700179 2.334016 11.74102 Figure C-9: The REM modellb with PRVATECREDIT used as financial development indicator Number' of obs Number of groups Random-effects GLS regression Group var·iable: country within = 0.1990 between = 0.3009 overall = 0.2191 R-sq: cor·r(u_i, X) = 360 24 Obs per gr·oup: = avg = max = 15 15.0 15 Wald chi2(11) Pr·ob > chi2 = (assumed) GROWTH Coef Std Err FDI iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF _cons 0001549 -.0504558 0038229 -.5562332 1666654 1.454345 -.1909601 -.9404587 -.0320367 0224833 1.337667 597387 0614316 0138409 0015435 4578016 0370785 6432101 0719135 3263708 0062383 0196867 8191257 4.052907 sigma_u si gma_e rho 87417743 3.5450764 05732076 (fraction of variance due to u_i) PRIVATECRE~T FDixPRIVAT~T = z [95% Conf Interval] P>lzl 0.00 -3.65 2.48 -1.22 4.49 2.26 -2.66 -2.88 -5.14 1.14 1.63 0.64 0.998 0.000 0.013 0.224 0.000 0.024 0.008 0.004 0.000 0.253 0.102 0.522 92.41 0.0000 = = -.1202489 -.0775835 0007976 -1.453508 093993 1936766 -.3319079 -1.580134 -.0442636 -.0161019 - 2677903 -5.346164 1205586 -.0233281 0068481 3410414 2393378 2.715014 -.0500123 -.3007837 -.0198097 0610684 943123 10.54094 Poolability Tests Poolability Tests across Countries (Chow Test): Firstly, we ran country by country OLS regressions and obtained the value of Residual Sum of Squares (RSSi) for each country i Then, we calculated sum of all RSSi, denoted as URSS Secondly, we ran POOLED OLS regression and obtained the value of RSS, denotes as RRSS Finally, the obtained values were replaced in the equation in section 3.3.2.2 • F= (~ RRSS Modell: pooled - {;t ~URSs.Jf[(N -l)(K +1)] _(6024.6527-531.99554)j(23xlO)_ f.uRSS; /(NT -N(K +1)) - ( ) 531.99554 I 24x i=l C-5 -5.3868 >- FOJ -X 10 24 , X _ -1.4654 • Model 2a with LL Y used as financial development indicator: (RRSSpooled- ±uRSS;jf[(N-1)(K +1)] F= ~ i=1 fuRSS;j(NT-N(K+1)) = (5915.9088-290.27631)j(23x12) = 5.0557 >-F, 290.27631j(24x3) , 31 24 =1 5908 -'-· ,_ i=1 • F= Model 2b with PRIVATECREDIT used as financial development indicator: (RRSSpooled- fuRSS;J![(N -1){K +1)] ~ i= IURSsi/(Nr-N{K+1)) = (5873.7223-311.2680)j(23x12) = 4.6618 >-F, 311.2680j(24x3) _ 31214 _x =1 5908 ,-XO i=l Poolability Tests over Time (Chow Test): Similar to poolability test across countries, but this time we ran time over time OLS regressions and obtained the value of Residual Sum of Squares (RSSi) for each year t Then, we calculated sum of all RSS() denoted as URSS Secondly, we ran POOLED OLS regression and obtained the value of RSS, denotes as RRSS Finally, the obtained values were replaced in the equation in section 3.3.2.2 • F = Modell: (~ RRSSpooled - ±uRSS jf[(r-1)(K +1)] 1= 1 T "f.URSS /(TN-T(K+ 1)) = (4879.4911-2479.8337)j(14x10) ( 2479.8337/ 15x14 ) = 1.9243 >- f 14x 10 15 x 14 =1.42558 · 1=1 • Model 2a with LL Y used as financial development indicator: ((RRSSpooled - {;t fuRSS jf[(r-1)(K +1)] F= ±uRssl/(rN-T(K+1)) _ {5915.9088-1715.9515)j{14x12) _ ( ) -1.9243 >- F14 1115 1715.9515/ 15x12 x_, 1=1 C-6 12 X _ -1.4239 • Model 2b with PRIVATECREDIT used as financial development indicator: ( RRSSpooled F= ~URSS )f[(T-1)(K +1)] f:J ±uRss~/(rN-T(K+ )) = (5873.7223-1475.2982) j{14x12) ( ) =3.1943>-FI4121512=1.4 1475.2982/ 15x12 X· X t=l Hausman tests Figure C-10: Hausman Test for FEM and REM of mode/1 FDI iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF (b-B) Difference sqr·t (di ag (V_b-V_B)) S.E .044383 -.921372 1109555 1.164496 -.2014226 -.7634579 -.0315076 0436342 2446693 -.1398848 -2.388675 -.0440697 5.135294 -.1258721 1.03389 -.0071527 0075406 2.281482 0137738 1.178305 0221048 1.095407 0744605 4495972 -.0955018 -3.310047 0668858 6.29979 -.3272947 270432 -.0386603 0511748 2.526151 B Test: - - Coefficients - (b) (B) fern rem Ho: = - 0216673 7968183 b = consistent under Ho and Ha; obtained from xtreg inconsistent under Ha, efficient under Ho; obtained from xtreg difference in coefficients not systematic chi2(9) = (b-B)'[(V_b-V_B)A(-l)](b-B) 10.73 = 0.2949 Prob>chi2 = (V_b-V_B is not positive definite) C-7 Figure C-11: Hausman Test for FEM and REM of model2a- using LLY - - Coefficients - (B) (b) fern rem -.2538103 -.0756504 0047607 -3.159241 0902749 7.21494 -.2604997 4604176 -.0407632 0695304 2.614612 FDI LLY FDixLLY iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF -.0377077 -.0366365 0029801 -.6554857 1450978 1.198514 -.1703203 -.7185275 -.0329545 0314662 8192321 (b-B) Difference sqrt(diag(V_b-V_B)) -.2161025 -.0390138 0017805 -2.503755 -.0548229 6.016426 -.0901793 1.178945 -.0078086 0380642 1.79538 0108332 0211063 0002387 1.272087 0220139 1.103269 0730969 4413451 s E .021936 8269426 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(11) = (b-B)'[(V_b-V_B)A(-1)](b-B) 61.84 = Prob>chi2 = 0.0000 (V_b-V_B is not positive definite) Figure C-12: Hausman Test for FEM and REM of model 2b with- PRIVATECREDIT - - Coefficients - (b) (B) fern rem FDI PRIVATECRE~T FDixPRIVAT~T iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF -.2049658 -.0929483 0063478 -3.320912 1320121 6.876593 -.2867758 421572 -.0393022 0603487 2.70417 0001549 -.0504558 0038229 -.5562332 1666654 1.454345 -.1909601 -.9404587 -.0320367 0224833 1.337667 (b-B) Differ·ence sqrt(diag(V_b-V_B)) -.2051206 -.0424925 002525 -2.764679 -.0346533 5.422248 -.0958157 1.362031 -.0072656 0378654 1.366503 0111478 0162985 0002347 1.197265 0211251 1.071996 0743929 4450092 s E .0208939 7067344 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: differ·ence in coefficients not systematic (b-B)'((V_b-V_B)A(-l)](b-B) 50.84 0.0000 (V_b-V_B is not positive definite) chi2(11) = = Prob>chi2 = C-8 - - - - Fixed effects tests Test for cross-section fzxed effects: Figure C-13: Test for cross-section fixed effects in the FEM model 2a- using LL Y ( 1) ( 2) ( 3) c 4) c 5) ( 6) ( 7) ( 8) c 9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) Jcountry_2 = Icoun1:ry = Icoun1:ry_4 = Jcoun1:ry = Icoun1:ry_6 = Jcoun1:ry = Icoun1:ry_8 = Jcoun1:ry_9 = Jcoun1:ry_l.O = Jcoun1:ry_ll = Jcoun1:ry_l2 = Jcoun1:ry_l3 = Jcoun1:ry_l4 = Jcoun1:ry_l.S = Icoun1:ry_l6 = Icoun1:ry_l7 = Jcoun1:ry_l8 = Jcoun1:ry_l9 = Icoun1:ry_20 = Icoun1:ry_21 = Icoun1:ry_22 = Icoun1:ry_23 = Jcoun1:ry_24 = 325) = F( 23, Pt•ob > F = Figure C-14: Test for cross-section fixed effects i the FEM mode/2b- using PRIVATECREDIT ( 1) ( 2) ( 3) ( 4) ( 5) ( 6) ( 7) ( 8) ( 9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) 6.29 0.0000 C-9 Jcoun1:ry_2 = Icoun1:ry_3 = Icoun1:ry_4 = Icoun1:ry_S = Icoun1:ry_6 = Jcoun1:ry_7 = Jcoun1:ry_8 = Jcountry_9 = Jcoun1:ry_l0 = Jcoun1:ry_l1 = Jcoun1:ry_l2 = Jcoun1:ry_l3 = _Icoun1:ry_l4 = _Icoun1:ry_l5 = _Icoun1:ry_l6 = _Icoun1:ry_l7 = _Icoun1:ry_l8 = _Icoun1:ry_l9 = _Icoun1:ry_20 = _Icoun1:ry_21 = _rcoun1:ry_22 = _rcoun1:ry_23 = _rcoun1:ry_24 = F( 23, 325) = Pt·ob > F = 6.19 0.0000 Test for timeftxed effects: Figure C-16: Test for time fixed effects in the FEM modella -using LL Y ( ( ( ( 1) 2) 3) 4) ( 5) ( 6) ( 7) ( 8) ( 9) (10) (11) (12) (13) (14) Iyear_1996 Jyear_l997 Jyear_1998 Jyear_1999 Iyear_.2000 Jyear_.2001 Jyear_.2002 Iyear_.2003 Iyear_.2004 Iyear_.2005 lyear_.2006 Iyear_.2007 Jyear_.2008 Jyear_.2009 F( 14 = = = = = = = = = = = = = = Prob > F c 1) 0 0 0 0 0 0 0 311) = I = Figure C-15: Test for time fixed effects in the FEM model2b- using PRIVATECREDIT ( 2) ( 3) ( 4) c 5) ( 6) ( 7) ( 8) 9) c (10) (11) (12) (13) (14) 6_28 0.0000 Jyear_1996 Jyear_1997 Iyear_1998 Jyear_1999 Iyear_.2000 Jyear_.2001 Jyear_.2002 Iyear_.2003 Iyear_.2004 Iyear_.2005 Jyear_.2006 Jyear_.2007 Jyear_.2008 Iyear_.2009 = =0 = =0 = = =0 = = =0 =0 = =0 =0 311) = F( 14, Prob > F = 5.80 0.0000 Heteroskedasticity Tests for Two-way FEM Regressions Models Figure C-17: HET Test for the REM mode/1 ikelihood-ratio test (Assumption: Restricted nested in Unrestricted) LR chi (23) Prob > chi2 Figure C-18: HET Test for the Two-way FEM Madella - using LL Y Modified Wald test for groupwise heteroskedasticity in fixed effect regression model HO: sigma(i)A2 = sigmaA2 for all i chi2 (24) = Prob>chi2 = 1314_79 0.0000 Figure C-19: HET Test for the Two-way FEM Mode/2b- using PRIVATECREDIT Modified Wald test for groupwise heteroskedasticity in fixed effect regression model HO: sigma(i)A2 = sigmaA2 for all i chi P (24) r'ob>chi 1239.9.1 = o_oooo C-10 249.26 0.0000 - - - APPENDIX D REGRESSION MODEL RESULTS Figure D-1: The REM mode/1 with Robust Standard Errors Number of obs Number of groups Random-effects GLS regression Group variable: country " R-sq: within between overall corr(u_i, X) = = Obs per group: = avg = = 0.1578 = 0.3132 = 0.1939 max = Wald chi2(9) Prob > chi2 = (assumed) = = 360 24 15 15.0 15 83.80 0.0000 (Std Err adjusted for' 24 clusters in country) GROWTH Coef Robust Std Err FDI iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF _cons 044383 -.921372 1109555 1.164496 -.2014226 -.7634579 -.0315076 0436342 2446693 4.056647 096054 5800785 0603592 1.017392 0829336 456565 0070633 0276181 1.18712 7.325578 sigma_u sigma_e rho 1.110122 3.6654267 08401922 (fraction of variance due to u_i) , D-11 z 0.46 -1.59 1.84 1.14 -2.43 -1.67 -4.46 1.58 0.21 0.55 P>lzl 0.644 0.112 0.066 0.252 0.015 0.094 0.000 0.114 0.837 0.580 [95% Conf Interval] -.1438794 -2.058305 -.0073463 -.8295569 -.3639695 -1.658309 -.0453514 -.0104962 -2.082044 -10.30122 2326453 2155609 2292573 3.158548 -.0388757 1313931 -.0176637 0977647 2.571383 18.41452 Figure D-2: The Two-way FEM Regression Model 2a with Robust Standard Errors- using LLY Number of obs = 360 F( 25, 311) = 6.86 Prob > F = 0.0000 R-squal'ed 0.5824 = Adj R-squal'ed = 0.5179 Root MSE 3.203 = Linear regression, absorbing indicators GROWTH Coef Robust Std En FDI LLY FDixLLY iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF _Iyear_1996 _Iyear_1997 _Iyear_1998 _Iyear·_1999 _Iyear_2000 _Iyear_2001 _Iyear_2002 _Iyear_2003 _Iyear·_2004 _Iyear_2005 _Iyeal'_2006 _Iyear_2007 _Iyear_2008 _Iyear_2009 _cons -.2579394 -.060756 0037706 -7.460473 1306783 4.364267 -.2005144 6537246 -.040748 0063268 2.529694 -.0875398 -.3785189 -1.941111 -.5473965 7409255 1010817 1.450658 2.77038 3.560317 3.663711 4.809626 5.752941 3.793493 4679464 37.86814 0923979 0236118 0010214 2.901285 043795 1.747161 1197131 511516 0073417 0299537 1.257938 9490902 9633381 1.050055 1.019555 940828 1.060464 1.026825 1.058166 1.12246 1.318707 1.514717 1.46837 1.704235 738655 17.6363 country absorbed t (95% Conf Interval] P>ltl -2.79 -2.57 3.69 -2.57 2.98 2.50 -1.67 1.28 -5.55 0.21 2.01 -0.09 -0.39 -1.85 -0.54 0.79 0.10 1.41 2.62 3.17 2.78 3.18 3.92 2.23 0.27 2.15 0.006 0.011 0.000 0.011 0.003 0.013 0.095 0.202 0.000 0.833 0.045 0.927 0.695 0.065 0.592 0.432 0.924 0.159 0.009 0.002 0.006 0.002 0.000 0.027 0.788 0.033 -.4397436 -.1072151 0017609 -13.1691 0445064 9265162 -.4360644 -.3527452 -.0551937 -.0526108 0545497 -1.95499 -2.274003 -4.007221 -2.553494 -1.110268 -1.98551 -.5697447 6883111 1.35174 1.068996 1.829237 2.863744 4402045 -2.953068 3.166594 (24 categories) ~ -~ - Figure D-3: The Wald Test on joint significance of POP in the Two-way FEM Model 2a -using LL Y test POP ( 1) POP = F( -.0761353 -.0142969 0057802 -1.7518421 2168502 7.802017 0350356 1.660194: -.0263023 0652644 5.004839 1.77991 516966 1249995 1.458701 592119 2.187673 3.47106 4.852449 5.768894 6.258426 7.790016 8.642137 7.146782 3.888961 72.56969 1, 311) Prob > F • D-12 = = 1.63 0.2022 - Figure D-4: The Two-way FEM Regression Madella with Robust Standard Errors- using LLY after eliminating POP Number of obs F( 24, 312) Prob > F R-squared Adj R-squared Root MSE Linear regression, absorbing indicators , GROWTH Coef Robust Std Err FDI LLY FDixLLY iGDP INVEST OPENNESS GOVEXP INFLATION EDUC GOVEFF _Iyear·_l996 _Iyear_l997 _Iyear_l998 _Iyear_1999 _Iyear_2000 _Iyear_2001 _Iyear_2002 _Iyear_2003 _Iyear_2004 _Iyear_2005 _Iyear_2006 _Iyear_2007 _Iyear_2008 _Iyear_2009 _cons -.2467751 -.0611298 0036947 -6.834686 1329217 4.114813 -.2056224 -.0409213 0060879 2.784116 -.1301899 -.448146 -2.018622 -.6499046 6220635 -.0281213 1.297167 2.61226 3.402253 3.46758 4.520441 5.374294 3.365265 -.0070936 35.85051 090422 0234052 0010048 2.854349 0436247 1.707731 1195829 0070627 0313833 1.247458 934478 9591465 1.037569 1.001266 9100522 1.042639 1.007889 1.036828 1.097374 1.294105 1.472139 1.42028 1.641883 1.675401 17.49612 country absor·bed t -2.73 -2.61 3.68 -2.39 3.05 2.41 -1.72 -5.79 0.19 2.23 -0.14 -0.47 -1.95 -0.65 0.68 -0.03 1.29 2.52 3.10 2.68 3.07 3.78 2.05 -0.00 2.05 -.4246892 -.1071817 0017176 -12.45089 0470858 7546861 -.4409133 -.0548179 -.0556617 3296213 -1.968866 -2.335359 -4.060139 -2.619992 -1.168552 -2.079615 -.6859515 5721997 1.243063 921304 1.623866 2.579756 1347016 -3.303607 1.425206 - Figure D-5: The Wald Test on joint significance of EDUC in the Two-way FEM Model 2a -using LL Y test EDUC ( 1) EDUC F( -.068861 -.0150778 0056718 -1.218479 2187575 7.474939 0296686 -.0270248 0678375 5.238611 1.708486 1.439067 0228948 1.320183 2.412679 2.023372 3.280286 4.652319 5.561442 6.013856 7.417016 8.168831 6.595829 3.28942 70.2758 (24 categories) - • 360 7.18 0.0000 0.5803 5171 3.2059 [95% Conf Interval] P>ltl 0.007 0.009 0.000 0.017 0.003 0.017 0.087 0.000 0.846 0.026 0.889 0.641 0.053 0.517 0.495 0.978 0.199 0.012 0.002 0.008 0.002 0.000 0.041 0.997 0.041 = = = = = = =0 1, 312) = Pr~ob > F D-13 = 0.04 0.8463 Figure D-6: The Two-way FEM Regression Model 2b with Robust Standard Errors-using PRIVATECREDIT Number of obs = 360 F( 25, 311) = 6.63 Prob > F = 0.0000 R-squared 5761 = Adj R-squared = 5107 Root MSE 3.227 = Linear regression, absorbing indicators GROWTH Coef Robust Std Er'r FDI FDixPRIVAT-T iGDP INVEST OPENNESS GOVEXP POP INFLATION EDUC GOVEFF _Iyear_l996 _Iyear_l997 _Iyear_l998 _Iyear_l999 _Iyear_2000 _Iyeat'_2001 _Iyear_2002 _Iyear·_2003 _Iyear_2004 _Iyear·_2005 _Iyear_2006 _Iyear·_2007 _Iyear_2008 _Iyear·_2009 _cons -.2265723 -.0528343 0047063 -7.410809 144302 4.065401 -.2247779 5503388 -.0411492 0027775 2.486295 -.1314264 -.3712222 -2.113059 -.9192629 3342861 -.4361131 8086782 2.188864 2.989104 3.240965 4.341027 5.25256 3.339857 -.1229632 38.34295 0976368 0222776 0016475 2.877987 0483589 1.671793 1200745 5429731 007697 0315651 1.21862 1.006511 9837842 1.046351 1.041934 9480375 1.039433 1.004106 1.048949 1.113439 1.298892 1.511326 1.45962 1.70855 1.782405 16.85323 countr·y absorbed PRIVATECRE~T t [95% Conf Interval] P>ltl -2.32 -2.37 2.86 -2.57 2.98 2.43 -1.87 1.01 -5.35 0.09 2.04 -0.13 -0.38 -2.02 -0.88 0.35 -0.42 0.81 2.09 2.68 2.50 2.87 3.60 1.95 -0.07 2.28 0.021 0.018 0.005 0.010 0.003 0.016 0.062 0.312 0.000 0.930 0.042 0.896 0.706 0.044 0.378 0.725 0.675 0.421 0.038 0.008 0.013 0.004 0.000 0.052 0.945 0.024 -.4186846 -.0966682 0014646 -13.0736 04915 775946 -.461039 -.5180265 -.0562939 -.0593306 0885127 -2.111859 -2.306937 -4.171881 -2.969393 -1.531093 -2.481323 -1.167021 124929 7982777 6852368 1.367311 2.380581 -.0219223 -3.630061 5.182178 test POP ( 1) 007948 -1.748022 2394541 7.354856 0114832 1.618704 -.0260044 0648855 4.884077 1.849006 1.564492 -.0542372 1.130868 2.199665 1.609097 2.784378 4.252798 5.17993 5.796693 7.314743 8.124538 6.701636 3.384135 71.50371 (24 categories) Figure D-7: The Wold Test on joint significance of POP in the Two-way FEM Model2b- using PRIVATECREDIT • -.03446 -.0090004 POP = F( 1, 311) Prob > F • D-14 = 1.03 0.3116 Figure D-8: The Two-way FEM Regression Model 2b with Robust Standard Errors-using PRIVATECREDIT after eliminating POP Number of obs = 360 F( 24, 312) = 6.89 Prob > F 0.0000 = R-squared 0.5746 = Adj R-squared = 0.5106 Root MSE = 3.2275 Linear regression, absorbing indicators • GROWTH Coef Robust std Err FDI iGDP INVEST OPENNESS GOVEXP INFLATION EDUC GOVEFF _ryear_l996 _Iyear_l997 _Iyear_l998 _Iyear_l999 _Iyear_2000 _Iyear_2001 _Iyear_2002 _Iyear_2003 _Iyear_2004 _Iyear_2005 _Iyear'_2006 _ryear_2007 _Iyear·_2008 _Iyear_2009 _cons -.2213065 -.0513155 0047374 -6.968327 1467062 3.840885 -.2278498 -.0413748 0018828 2.73695 -.1690432 -.4311495 -2.172115 -.9990131 2405171 -.5347169 6953581 2.072405 2.875725 3.090106 4.111373 4.943612 2.998962 -.5105703 37.22432 0969968 0220347 0016243 2.847722 0479926 1.627826 1200525 0074481 0325724 1.21093 9844843 970456 1.0369 1.020748 9156043 1.017564 9825771 1.026967 1.09063 1.27609 1.475846 1.413198 1.654077 72131 16.74387 country absorbed PRIVATECRE~T FDixPRIVAT~T t -2.28 -2.33 2.92 -2.45 3.06 2.36 -1.90 -5.56 0.06 2.26 -0.17 -0.44 -2.09 -0.98 0.26 -0.53 0.71 2.02 2.64 2.42 2.79 3.50 1.81 -0.30 2.22 [95% Conf Interval] P>ltl 0.023 0.021 0.004 0.015 0.002 0.019 0.059 0.000 0.954 0.024 0.864 0.657 0.037 0.328 0.793 0.600 0.480 0.044 0.009 0.016 0.006 0.001 0.071 0.767 0.027 -.4121571 -.0946709 0015415 -12.57149 0522761 6379807 -.4640646 -.0560296 -.0622066 3543276 -2.106111 -2.340615 -4.212315 -3.007433 -1.561023 -2.536873 -1.237957 0517487 7298054 5792758 1.207503 2.163009 -.2555937 -3.897414 4.279148 (24 categories) - Figure D-9: The Wald Test on joint significance of EDUC in the Two-way FEM Mode/2b- using PRIVATECREDIT fl test: EDUC ( 1) EDUC F( -.0304559 -.0079601 0079334 -1.36516 2411364 7.043789 008365 -.0267201 0659721 5.119572 1.768025 1.478316 -.131915 1.009407 2.042057 1.467439 2.628673 4.093061 5.021644 5.600935 7.015243 7.724216 6.253518 2.876273 70.1695 =0 1, 312) = Prob > F = D-15 0.00 0.9539 I Final regression models: Figure D-10: The final Two-way FEM Regression Modell a -using LL Y inear regression, absorbing indicators Number of obs F( 23, 313) Prob > F R-squared Adj R-squared Root MSE ~ • "' GROWTH Coef Robust Std Err FDI LLY FDixLLY iGDP INVEST OPENNESS GOVEXP INFLATION GOVEFF _Iyear_1996 _Iyear_1997 _Iyear·_l998 _Iyear_l999 _Iyear·_2000 _Iyear_2001 _Iyear_2002 _Iyear·_2003 _Iyear_2004 _Iyear·_2005 _Iyear_2006 _Iyear·_2007 _Iyear_2008 _Iyear·_2009 _cons -.2466713 -.0602514 0036776 -6.867126 1329443 4.118697 -.2067933 -.0409251 2.806065 -.1222713 -.4359639 -1.99895 -.6200958 6571932 0166183 1.350196 2.68078 3.468395 3.548695 4.608533 5.469803 3.468786 0995689 36.37885 0902736 0231229 0010099 2.870681 043507 1.70669 119554 0070175 1.24679 932096 9618854 1.044489 9963094 908384 1.067383 1.005276 1.014971 1.090953 1.269214 1.431801 1.410299 1.640443 1.624664 17.51961 countr·y absorbed t -2.73 -2.61 3.64 -2.39 3.06 2.41 -1.73 -5.83 2.25 -0.13 -0.45 -1.91 -0.62 0.72 0.02 1.34 2.64 3.18 2.80 3.22 3.88 2.11 0.06 2.08 P>ltl 0.007 0.010 0.000 0.017 0.002 0.016 0.085 0.000 0.025 0.896 0.651 0.057 0.534 0.470 0.988 0.180 0.009 0.002 0.005 0.001 0.000 0.035 0.951 0.039 = = 360 7.48 = 0.0000 = 0.5802 = 0.5186 = 3.201 [95% Conf Interval] -.424291 -.1057473 0016905 -12.5154 047341 7606607 -.4420243 -.0547325 3529145 -1.956237 -2.328543 -4.054057 -2.580406 -1.130118 -2.083534 -.6277568 6837505 1.321866 1.051426 1.791361 2.694938 2410967 -3.097075 1.907752 -.0690515 -.0147555 0056647 -1.218855 2185475 7.476733 0284377 -.0271178 5.259215 1.711695 1.456615 0561577 1.340215 2.444504 2.11677 3.328149 4.677809 5.614924 6.045964 7.425705 8.244668 6.696476 3.296213 70.84994 (24 categories) ~ D-16 Figure D-11: The final Two-way FEM Regression Mode/2b- using PRIVATECREDIT Linear regression, absorbing indicators GROWTH Coef FDI INVEST OPENNESS GOVEXP INFLATION GOVEFF _Iyear_l996 _Iyear_l997 _Iyear_l998 _Iyear_l999 _Iyear_2000 _Iyear·_2001 _Iyear_2002 _Iyear·_2003 _Iyear_2004 _IyeaP_2005 _IyeaP_2006 _Iyear_2007 _Iyear_2008 _Iyear_2009 _cons -.2215232 -.0510652 0047363 -6.981062 1465622 3.843331 -.2281068 -.0413784 2.742244 -.1664142 -.4276732 -2.165701 -.98839 2532932 -.5179762 7152988 2.097502 2.900049 3.118716 4.142181 4.976495 3.034315 -.4728692 37.40453 country absorbed PRIVATECRE~T FDixPRIVAT~T iGDP Robust Std Er·r .0963241 0217374 0016236 2.870017 0477933 1.6278 1200888 0074262 1.214143 9851724 9763974 1.040513 1.022455 9245867 1.060198 9908455 1.014508 1.095467 1.266431 1.436829 1.415628 1.662898 1.690445 16.85996 Number of obs F( 23, 313) Prob > F R-squared Adj R-squared Root MSE t -2.30 -2.35 2.92 -2.43 3.07 2.36 -1.90 -5.57 2.26 -0.17 -0.44 -2.08 -0.97 0.27 -0.49 0.72 2.07 2.65 2.46 2.88 3.52 1.82 -0.28 2.22 P>ltl 0.022 0.019 0.004 0.016 0.002 0.019 0.058 0.000 0.025 0.866 0.662 0.038 0.334 0.784 0.625 0.471 0.040 0.009 0.014 0.004 0.001 0.069 0.780 0.027 = 360 7.08 0.0000; = = 5746: = 0.5121 = 3.2223 = [95% Conf Interval] -.4110477 -.0938352 0015416 -12.62803 0525254 6405171 -.4643903 -.05599 3533301 -2.104812 -2.348805 -4.212985 -3.000145 -1.565898 -2.603993 -1.234261 1013832 7446402 6269206 1.315117 2.191145 -.2375557 -3.798942 4.231337 -.0319986 -.0082952 0079309 -1.334097 2405989 7.046145 0081766 -.0267668 5.131159 1.771983 1.493459 -.1184173 1.023365 2.072484 56804 2.664859 4.09362 5.055459 5.610511 6.969244 7.761846 6.306186 2.853204 70.57773 (24 categories) D-17 I ... Examine the impact of FDI on economic growth in Asian developing countries (2) Investigate the role of financial development in FDI -Growth nexus m Asian developing countries The research findings... few studies examining the role of financial development in enhancing the contributions of FDI on economic growth in Asian developing countries Vietnam is one of Asian developing countries where... one of developing economy in Asia Finally, there are still few studies examine the role of financial development in enhancing the contributions of FDI on economic growth in Asian developing countries