The relationship between financial development and household welfare case study in five asian countries

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES HO CHI MINH CITY THE HAGUE VIETNAM THE NETHERLANDS VIETNAM-NETHERLANDS PROGRAMME FOR MASTER IN DEVELOPMENT ECONOMICS THE RELATIONSHIP BETWEEN FINANCIAL DEVELOPMENT AND HOUSEHOLD WELFARE: CASE STUDY IN FIVE ASIAN COUNTRIES By PHAN THI KHANH VAN This paper was submitted in partial fulfillment of the requirements for Master’s degree in Development Economics Ho Chi Minh City, July 2013 UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES HO CHI MINH CITY THE HAGUE VIETNAM THE NETHERLANDS VIETNAM-NETHERLANDS PROGRAMME FOR MASTER IN DEVELOPMENT ECONOMICS THE RELATIONSHIP BETWEEN FINANCIAL DEVELOPMENT AND HOUSEHOLD WELFARE: CASE STUDY IN FIVE ASIAN COUNTRIES By PHAN THI KHANH VAN Academic supervisor Dr DUONG NHU HUNG Thispaper was submitted in partial fulfillment of the requirements for Master’s degree in Development Economics Ho Chi Minh City, July 2013 CONTENTS ACKNOWLEDGEMENT .iii ABSTRACT .iv ABBREVIATION .v LIST OF FIGURES vi LIST OF TABLES .vi CHAPTER I: INTRODUCTION .1 Problem statement Research objectives 3 Research questions 4 Justification of the study Scope of the study Structure of the study .4 CHAPTER II: LITERATURE REVIEW .5 Definitions of key concepts 1.1 Financial development 1.2 Household welfare and Poverty Theoretical literature 2.1 Direct relationship 2.2 Indirect relationship Empirical studies CHAPTER III: ECONOMETRICS REVIEW 12 Stochastic Process, Stationarity and Random Walks 12 Unit Root Test 13 Cointegration 13 Granger Causality Test 14 Panel Unit Root Test 15 Panel Cointegration 16 [i] Instrumental Variables Regression (IV) 18 Generalized method of moments (GMM) 19 CHAPTER IV: DATA AND RESEARCH METHODOLOGY 22 Data 22 Research methodology 23 CHAPTER V: ANALYSIS RESULTS 26 Data descriptions 26 Empirical results 31 CHAPTER VI: CONCLUSIONS AND POLICY IMPLICATIONS 36 Conclusions 36 Policy implications 37 Limitations and directions for further studies 38 3.1 Limitations 38 3.2 Directions for further studies 39 REFERENCES 40 APPENDICES a Appendix 1: Description of FD and PR variables (1998-2011) a Appendix 2: Panel Unit Root Test of variables b Appendix 3: Pedroni Cointegration Test j Appendix 4: GMM m [ii] ACKNOWLEDGEMENT The study would be done successfully thank to the beautiful assistance and guidance of everybody who are always with me during the research period First, I would like to express my deepest appreciation to The Board of Management and Theoretical Supervision of this program In fact, I would like to send sincere thanks to Dr Nguyen Trong Hoai, Dr Pham Khanh Nam who give me a very first step guidance doing this study (theoretical review and techniques understanding) and give me a piece of advice when I got in stuck with my study Besides that, I also thank to Dr Duong Dang Thuy, who has given me theoretical advice and introduced me to Dr Le Van Chon for intensive support Moreover, I would like to express my gratitude to Dr Le Van Chon and Dr Phung Thanh Binh who supported my study and my mind when I lost inspiration during research period Not only the teaching staff but also the class MDE 17 classmates are those I really appreciate Thanks to their care, their understanding and their sharing, I know exactly what I should The more importantly, I would like to express my appreciation to my direct supervisor Dr Duong Nhu Hung He is a very kind teacher who always cares me and encourages me during the research period He is kind to me with his scientific guidance, soft but invaluable advice till the final stage of the study Last but not least, during the time doing this study, I encountered both mental and fiscal problems At that time, my family, especially my mother, my grandmother and my aunts who always advise me to try my best and give me spiritual assistance I also send my sincere thanks to my husband who is always with me I am proud of his patience and his sympathy He has given me a chance to concentrate on my studying instead of housework [3] ABSTRACT The study presents the empirical result of the relationship between financial development and household welfare, which has been the hotly debated issue recently To detect the nexus of financial development and household welfare, the Pedroni cointegration test is run to find out the long-run relationship between financial development and household welfare In empirical study, it is affirmed that there exists the long-run relationship between financial development and household welfare However, the impact of financial development on household welfare cannot be shown through Pedroni cointegration test Thus, 2SLS GMM is deployed to identify the impact of financial development on household welfare Keywords: financial development, household welfare, cointegration, two stage least squares, ABBREVIATION 2SLS: Two Stage Least Squares ADF: Augmented Dickey Fuller ADRL: Autoregressive Distributed Lag Model AIC: Akaike information criterion AR: Auto-regressive DCBS: Domestic credit provided by banking sector as a percentage of GDP DCP/GDP: Domestic credit to the private sector as a ratio of gross domestic product DCPS: Domestic credit to the private sector as a percentage of GDP DF: Dickey Fuller DMBA: Domestic money bank assets EG: Economic growth FD: Financial development GDP: Gross Domestic Product GMM: Generalized method of moments HLSS: Household Living Standards Survey IMF: International Monetary Fund IV: Instrumental Variable M2/GDP: money and quasi money as percentage of GDP M3: the broadest definition of money OECD: Organization for Economic Co-operation and Development OLS: Ordinary Least Square PP: Phillips-Perron PR: Poverty reduction SBC: Schwarz’s Bayesian criterion SME: Small, medium-sized enterprise VAR: Vector Auto-regressive VECM: Vector Error Correction Model WB: World Bank WDI: World Development Indicator WEF: World Economic Forum LIST OF FIGURES Figure 1: Money and quasi money (M2) as percentage of GDP Figure 2: Household per capita consumption (constant 2000 US$) Figure 3: Financial Sector Development and Poverty Reduction .6 Figure 4: Line graph of proxies of FD and household welfare from 1960 to 2011in five Asian countries 28 Figure 5: Relationship between FD and per capita consumption in five Asian countries (19602011) 30 LIST OF TABLES Table 1: Empirical studies about the causal nexus of FD and PR .10 Table 2: Proxy variables 23 Table 3: Description of FD and household welfare variables (1960-2011) 27 Table 4: t-statistics panel unit root tests 32 Table 5: t-statistics panel unit root test: Variables at the first difference 32 Table 6: Pedroni cointegration tests: Variables from 1960 to 2011 33 Table 7: Pedroni cointegration tests: Variables from 1998 to 2011 33 Table 8: Two stage least squares estimator between FD and PR 34 CHAPTER I: INTRODUCTION Problem statement It is undeniable that the relationship between financial development (FD) and economic growth (EG) has been one of the most attractive areas of research in the field of economic development over recent decades Some related studies have been employed, yet this relation remains controversial issues In fact, there have been some conflicts on the relationship between finance and growth in earlier literature In fact, Robinson (1952) and Lucas (1988) dismiss the role of finance in understanding EG while McKinnon (1973) and Miller (1988) insist on this relation between FD and EG In recent researches, the harmony of vital roles of finance in enhancing growth has been reached For example, Kirkpatrik (2000) states that good financial system that mobilizes savings and allocates resources to more productivity contributes to growth by supporting capital accumulation, promoting investment efficiency, and improving technology Furthermore, many people believe that EG reduces absolute poverty because the more growth the economy reaches, the more jobs would be generated for the poor or the fewer differentials in wage between the skilled and unskilled labor at a later stage of development (Galor and Tsiddon, 1996) benefits the poor Then, a consensus emerged recently is that EG overall leads to poverty reduction (PR) through the improvement of household welfare However, these close relationships between FD and EG or between EG and household welfare not mean that FD contributes to PR (Beck et al, 2007) through the improvement of household’s welfare The explanation follows that the goal of EG in most developing countries is linked with both PR and income distribution In other words, if FD stimulates EG by increasing income of the rich, which results in worsening income equality, FD will not benefit the poor This debate appeals many researchers to conduct studies on relationship between FD and household welfare In addition, this paper aims to examine the relationship between FD and household welfare in five Asian countries including Indonesia, Malaysia, Philippines, Thailand, and Vietnam The reason why the research focuses on a set of these five Asian countries is that there is little research on FD and PR in Asia Due to the limit of short time series data, the research will identify the relationship between FD and PR in panel data, especially panel five Asian countries with the assumption that these countries are nearly at the same foundation of development [1] Panel unit root test: Summary Series: D(DCPS) Date: 07/23/13 Time: 20:49 Sample: 1960 2011 Exogenous variables: Individual effects Automatic selection of maximum lags Automatic selection of lags based on SIC: to Newey-West bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* -5.67304 0.0000 Crosssections Obs 204 5 204 204 206 Null: Unit root (assumes individual unit root process) Im, Pesaran and Shin W-stat -6.45018 0.0000 ADF - Fisher Chi-square 66.8768 0.0000 PP - Fisher Chi-square 75.6148 0.0000 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality Panel unit root test: Summary Series: D(DCPS) Date: 07/23/13 Time: 20:49 Sample: 1960 2011 Exogenous variables: Individual effects, individual linear trends Automatic selection of maximum lags Automatic selection of lags based on SIC: to NeweyWest bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* -4.40321 0.0000 Breitung t-stat -5.83748 0.0000 Im, Pesaran and Shin W-stat ADF - Fisher Chi-square PP - Fisher Chi-square -5.65782 56.1273 60.3519 0.0000 0.0000 0.0000 Crosssections Obs 5 203 198 5 203 203 206 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [e] Null: Unit root (assumes individual unit root process) Im, Pesaran and Shin W-stat ADF - Fisher Chi-square PP - Fisher Chi-square -5.65782 56.1273 60.3519 0.0000 0.0000 0.0000 5 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [e] 203 203 206 Panel unit root test: Summary Series: M2GDP Date: 07/23/13 Time: 20:49 Sample: 1960 2011 Exogenous variables: Individual effects Automatic selection of maximum lags Automatic selection of lags based on SIC: to Newey-West bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* 0.40374 0.6568 Crosssections Obs 211 5 211 211 212 Null: Unit root (assumes individual unit root process) Im, Pesaran and Shin W-stat 2.35189 0.9907 ADF - Fisher Chi-square 2.74098 0.9869 PP - Fisher Chi-square 2.55015 0.9901 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality Panel unit root test: Summary Series: M2GDP Date: 07/23/13 Time: 20:49 Sample: 1960 2011 Exogenous variables: Individual effects, individual linear trends Automatic selection of maximum lags Automatic selection of lags based on SIC: to NeweyWest bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* 0.29729 0.6169 Breitung t-stat -0.23661 0.4065 Crosssections Obs 5 210 205 5 210 212 Null: Unit root (assumes individual unit root process) ADF - Fisher Chi-square PP - Fisher Chi-square 9.00226 7.97541 0.5319 0.6312 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [f] Im, Pesaran and Shin W-stat 0.13789 0.5548 210 ADF - Fisher Chi-square PP - Fisher Chi-square 9.00226 7.97541 0.5319 0.6312 5 210 212 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [f] Panel unit root test: Summary Series: D(M2GDP) Date: 07/23/13 Time: 20:50 Sample: 1960 2011 Exogenous variables: Individual effects Automatic selection of maximum lags Automatic selection of lags based on SIC: to Newey-West bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* -8.22166 0.0000 Crosssections Obs 205 5 205 205 206 Null: Unit root (assumes individual unit root process) Im, Pesaran and Shin W-stat -9.33656 0.0000 ADF - Fisher Chi-square 99.2533 0.0000 PP - Fisher Chi-square 103.895 0.0000 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality Panel unit root test: Summary Series: D(M2GDP) Date: 07/23/13 Time: 20:50 Sample: 1960 2011 Exogenous variables: Individual effects, individual linear trends Automatic selection of maximum lags Automatic selection of lags based on SIC: Newey-West bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* -8.46849 0.0000 Breitung t-stat -4.85869 0.0000 ADF - Fisher Chi-square PP - Fisher Chi-square 84.7537 86.2576 0.0000 0.0000 Crosssections Obs 5 206 201 5 206 206 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [g] Null: Unit root (assumes individual unit root process) Im, Pesaran and Shin W-stat -8.88168 0.0000 ADF - Fisher Chi-square PP - Fisher Chi-square 84.7537 86.2576 0.0000 0.0000 206 5 206 206 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [g] Panel unit root test: Summary Series: POV Date: 07/23/13 Time: 20:50 Sample: 1960 2011 Exogenous variables: Individual effects Automatic selection of maximum lags Automatic selection of lags based on SIC: to Newey-West bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* 6.09644 1.0000 Crosssections Obs 219 5 219 219 221 Null: Unit root (assumes individual unit root process) Im, Pesaran and Shin W-stat 8.85880 1.0000 ADF - Fisher Chi-square 0.05213 1.0000 PP - Fisher Chi-square 0.01772 1.0000 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality Panel unit root test: Summary Series: POV Date: 07/23/13 Time: 20:50 Sample: 1960 2011 Exogenous variables: Individual effects, individual linear trends Automatic selection of maximum lags Automatic selection of lags based on SIC: to NeweyWest bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* 1.64128 0.9496 Breitung t-stat -1.67915 0.0466 Crosssections Obs 5 218 213 5 218 221 Null: Unit root (assumes individual unit root process) ADF - Fisher Chi-square PP - Fisher Chi-square 2.55141 1.92876 0.9901 0.9969 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [h] Im, Pesaran and Shin W-stat 3.14508 0.9992 218 ADF - Fisher Chi-square PP - Fisher Chi-square 2.55141 1.92876 0.9901 0.9969 5 218 221 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [h] Panel unit root test: Summary Series: D(POV) Date: 07/23/13 Time: 20:50 Sample: 1960 2011 Exogenous variables: Individual effects Automatic selection of maximum lags Automatic selection of lags based on SIC: Newey-West bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* -7.26526 0.0000 Null: Unit root (assumes individual unit root process) Im, Pesaran and Shin W-stat -6.09008 0.0000 ADF - Fisher Chi-square 60.9841 0.0000 PP - Fisher Chi-square 61.4249 0.0000 Crosssections Obs 216 5 216 216 216 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality Panel unit root test: Summary Series: D(POV) Date: 07/23/13 Time: 20:50 Sample: 1960 2011 Exogenous variables: Individual effects, individual linear trends Automatic selection of maximum lags Automatic selection of lags based on SIC: Newey-West bandwidth selection using Bartlett kernel Method Statistic Prob.** Null: Unit root (assumes common unit root process) Levin, Lin & Chu t* -8.13128 0.0000 Breitung t-stat -3.19801 0.0007 Null: Unit root (assumes individual unit root process) Im, Pesaran and Shin W-stat -6.52658 0.0000 ADF - Fisher Chi-square 60.0781 0.0000 PP - Fisher Chi-square 58.9557 0.0000 Crosssections Obs 5 216 211 5 216 216 216 ** Probabilities for Fisher tests are computed using an asymptotic Chi -square distribution All other tests assume asymptotic normality [i] Appendix 3: Pedroni Cointegration Test Pedroni Residual Cointegration Test Series: DCBS POV Date: 07/23/13 Time: 20:52 Sample: 1960 2011 Included observations: 260 Cross-sections included: Null Hypothesis: No cointegration Trend assumption: Deterministic intercept and trend Lag selection: Automatic SIC with max lag of to 10 Newey-West bandwidth selection with Bartlett kernel Alternative hypothesis: common AR coefs (within-dimension) Weighted Statistic Prob Statistic Panel v-Statistic 0.727063 0.3063 -0.133148 Panel rho-Statistic 2.271678 0.0302 2.286869 Panel PP-Statistic -0.905831 0.2647 -0.796727 Panel ADF-Statistic -0.853273 0.2772 -0.244467 Prob 0.3954 0.0292 0.2904 0.3872 Alternative hypothesis: individual AR coefs (between-dimension) Statistic 2.398442 -2.294637 -1.395455 Group rho-Statistic Group PP-Statistic Group ADF-Statistic Prob 0.0225 0.0287 0.1507 Cross section specific results Phillips-Peron results (non-parametric) Cross ID _IDN _MYS _PHL _THA _VNM AR(1) 0.862 0.710 0.850 0.875 -0.280 Variance 42.79343 137.8900 31.19693 65.39810 35.28130 HAC Bandwidth 54.39705 1.00 149.4048 2.00 53.32089 4.00 99.87640 2.00 8.410977 15.00 Obs 31 51 51 51 16 Augmented Dickey-Fuller results (parametric) Cross ID _IDN _MYS _PHL _THA _VNM AR(1) 0.862 0.710 0.808 0.825 -0.872 Variance 42.79343 137.8900 28.74644 55.73794 29.56857 Lag 0 1 [j] Max lag 10 10 10 Obs 31 51 50 50 15 Pedroni Residual Cointegration Test Series: DCPS POV Date: 07/23/13 Time: 20:53 Sample: 1960 2011 Included observations: 260 Cross-sections included: Null Hypothesis: No cointegration Trend assumption: Deterministic intercept and trend Lag selection: Automatic SIC with max lag of to 10 Newey-West bandwidth selection with Bartlett kernel Alternative hypothesis: common AR coefs (within-dimension) Weighted Statistic Prob Statistic Panel v-Statistic 0.150332 0.3945 -0.249718 Panel rho-Statistic 2.629713 0.0126 2.613810 Panel PP-Statistic 0.135096 0.3953 0.216440 Panel ADF-Statistic -1.571483 0.1161 -0.649761 Prob 0.3867 0.0131 0.3897 0.3230 Alternative hypothesis: individual AR coefs (between-dimension) Group rho-Statistic Group PP-Statistic Group ADF-Statistic Statistic 2.498062 -0.978685 -1.875898 Prob 0.0176 0.2471 0.0687 Cross section specific results Phillips-Peron results (non-parametric) Cross ID _IDN _MYS _PHL _THA _VNM AR(1) 0.961 0.852 0.849 0.908 -0.320 Variance 14.68723 71.96520 14.56459 59.13588 27.87482 Cross ID _IDN _MYS _PHL _THA _VNM AR(1) 0.933 0.738 0.792 0.858 -0.320 Variance 13.80759 52.49283 11.96756 44.42363 27.87482 HAC Bandwidth 27.92197 4.00 114.4923 4.00 22.73068 2.00 99.32617 2.00 11.00250 9.00 Lag 1 [k] Max lag 10 10 10 Obs 42 51 51 51 16 Obs 41 48 50 50 16 Augmented Dickey-Fuller results (parametric) Cross ID _IDN _MYS _PHL _THA _VNM AR(1) 0.933 0.738 0.792 0.858 -0.320 Variance 13.80759 52.49283 11.96756 44.42363 27.87482 Lag 1 [k] Max lag 10 10 10 Obs 41 48 50 50 16 Pedroni Residual Cointegration Test Series: M2GDP POV Date: 07/23/13 Time: 20:54 Sample: 1960 2011 Included observations: 260 Cross-sections included: Null Hypothesis: No cointegration Trend assumption: Deterministic intercept and trend Lag selection: Automatic SIC with max lag of to 10 Newey-West bandwidth selection with Bartlett kernel Alternative hypothesis: common AR coefs (within-dimension) Weighted Statistic Prob Statistic Panel v-Statistic 2.974354 0.0048 0.200065 Panel rho-Statistic 1.908384 0.0646 2.749520 Panel PP-Statistic -1.470154 0.1354 0.972745 Panel ADF-Statistic -1.781571 0.0816 0.620047 Prob 0.3910 0.0091 0.2486 0.3292 Alternative hypothesis: individual AR coefs (between-dimension) Group rho-Statistic Group PP-Statistic Group ADF-Statistic Statistic 2.248401 0.003741 -0.537990 Prob 0.0319 0.3989 0.3452 Cross section specific results Phillips-Peron results (non-parametric) Cross ID _IDN _MYS _PHL _THA _VNM AR(1) 0.992 0.603 0.898 0.816 -0.095 Variance 6.017865 112.4427 7.786163 17.55899 41.20204 HAC Bandwidth 14.92323 4.00 113.6337 2.00 9.081815 3.00 22.97473 1.00 40.11209 1.00 Obs 42 51 51 51 16 Augmented Dickey-Fuller results (parametric) Cross ID _IDN _MYS _PHL _THA _VNM AR(1) 0.945 0.603 0.898 0.747 -0.095 Variance 4.232040 112.4427 7.786163 15.03292 41.20204 Lag 0 [l] Max lag 10 10 10 Obs 41 51 51 50 16 Appendix 4: GMM Dependent Variable: POV Method: Panel Generalized Method of Moments Date: 07/23/13 Time: 20:57 Sample (adjusted): 1961 2011 Periods included: 51 Cross-sections included: Total panel (unbalanced) observations: 202 2SLS instrument weighting matrix Instrument list: C POV(-1) C DCBS Coefficient Std Error t-Statistic Prob -52.38577 13.30543 56.77871 0.814982 -0.922630 16.32604 0.3573 0.0000 Effects Specification Cross-section fixed (dummy variables) R-squared Adjusted R-squared S.E of regression Durbin-Watson stat Second-Stage SSR 0.589519 0.579048 334.8660 0.178489 316839.6 Mean dependent var S.D dependent var Sum squared resid J-statistic Instrument rank 791.0080 516.1248 21978501 9.41E-31 6.000000 Dependent Variable: POV Method: Panel Generalized Method of Moments Date: 07/23/13 Time: 20:59 Sample (adjusted): 1961 2011 Periods included: 51 Cross-sections included: Total panel (unbalanced) observations: 213 2SLS instrument weighting matrix Instrument list: C POV(-1) C DCPS Coefficient Std Error t-Statistic Prob 42.65682 13.76771 44.03985 0.749614 0.968596 18.36641 0.3339 0.0000 Effects Specification Cross-section fixed (dummy variables) R-squared Adjusted R-squared S.E of regression Durbin-Watson stat Second-Stage SSR 0.676839 0.669033 300.7580 0.131910 317483.5 Mean dependent var S.D dependent var Sum squared resid J-statistic Instrument rank [m] 757.4938 522.7870 18724265 5.55E-27 6.000000 Dependent Variable: POV Method: Panel Generalized Method of Moments Date: 07/23/13 Time: 20:59 Sample (adjusted): 1961 2011 Periods included: 51 Cross-sections included: Total panel (unbalanced) observations: 213 2SLS instrument weighting matrix Instrument list: C POV(-1) C M2GDP Coefficie nt Std Error t-Statistic Prob 146.5662 15.45568 47.90715 0.755479 -3.059381 20.45813 0.0025 0.0000 Effects Specification Cross-section fixed (dummy variables) R-squared Adjusted R-squared S.E of regression Durbin-Watson stat Second-Stage SSR 0.739543 0.733252 270.0074 0.238032 317483.5 Mean dependent var S.D dependent var Sum squared resid J-statistic Instrument rank [n] 757.4938 522.7870 15091129 7.40E-29 6.000000 ... improve household welfare This study uses the panel cointegration to examine the relationship between FD and welfare in five Asian countries including Indonesia, Malaysia, Philippines, Thailand and. .. five Asian countries are increasing years by years while the household welfare values in these countries are also getting increasing Thus, there exist the positive relationship between FD and welfare. .. finance and household welfare in five Asian countries including Malaysia, Indonesia, Philippines, Thailand and Vietnam with the data time series spanning from 1960 to 2011 Structure of the study