Saving behaviour of some developing countries in southeast asia

UNIVERSITY OF ECONOMICS HO CHI MINH CITY VIETNAM INSTITUTE OF SOCIAL STUDIES THE HAGUE THE NETHERLANDS VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS VIETNAM ECONOMIC GROWTH AND SAVING ANALYSIS IN THE PERIOD 1989 - 2012 BY LE DUC ANH MASTER OF ARTS IN DEVELOPMENT ECONOMICS HO CHI MINH CITY, DECEMBER 2014 UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES HO CHI MINH CITY VIETNAM THE HAGUE THE NETHERLANDS VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS VIETNAM ECONOMIC GROWTH AND SAVING ANALYSIS IN THE PERIOD 1989 - 2012 A thesis submitted in partial fulfilment of the requirements for the degree of MASTER OF ARTS IN DEVELOPMENT ECONOMICS By Le Duc Anh Academic Supervisor: Dr Dinh Cong Khai HO CHI MINH CITY, DECEMBER 2014 Acknowledgement Foremost, I would like to express my sincere gratitude to my supervisor Dr Dinh Cong Khai for the continuous support of my M.A study and research, for his patience, motivation, enthusiasm and immense knowledge His guidance helped me in all the time of research and writing of this thesis Besides my supervisor, I would like to thank VNP teaching staff for their encouragement, insightful comments, and hard questions Last but not the least, I would like to thank my family: my parents, my wife and my brothers for supporting me spiritually throughout my life OUTLINE Abstract Chapter 1: Introduction 1.1 Problem statement 1.2 Research objectives 1.3 Research questions 1.4 Methodology 1.5 Research scope 1.6 Structure of the study Chapter 2: Literature Review 2.1 Theoretical literature 2.2 Empirical literature 10 2.3 Conceptual framework 18 Chapter 3: Research Methodology 20 3.1 Data collection 20 3.2 Unit root test 21 3.3 Cointegration analysis 23 3.4 Granger causality analysis 24 Chapter 4: Empirical Analysis 26 4.1 Empirical evidence 26 4.1.1 Unit root test results 26 4.1.2 Cointegration test results 28 4.1.3 Granger causality test results 30 4.2 Discussion of findings 33 Chapter 5: Concluding Remarks 40 References Appendix LIST OF TABLES AND FIGURES Figures Figure 1: Relationship between saving and economic growth stated by Solow (1956) Figure 2: The grapth of LGDP, LGDS and LFDI series 26 Tables Table 1: The summary of the main empirical studies 13 Table 2: Unit root test of variables at level value 27 Table 3: Unit root test of variables at the first difference 28 Table 4: Results of ARDL bound test 29 Table 5: Estimated long run coefficients using the conditional ARDL(2, 0, 0) 30 Table 6: Results of VECM based on error terms taken from the conditional ARDL 31 Table 7: Results of diagnostic tests for equation 11 32 Table 8: Results of short-run Granger causality 33 Table 9: Average values of FDI, GDP and GDS in three stages 34 Table 10: Commercial bank network density 36 Table 11: Average percentage of rural population over total population in three stages 36 Table 12: Vietnam stock market in the period 2006 – 2012 37 LIST OF ABBREVIATIONS Abbreviation Meaning ADF Augmented Dickey-Fuller test AIC Akaike Information Criterion ARDL Autoregressive Distributed Lag ASEAN Association of Southeast Asian Nations ECT Error Correction Term FDI Foreign Direct Investment GDP Gross Domestic Product GDS Gross Domestic Saving LGDP Logarithm of Gross Domestic Product LGDS Logarithm of Gross Domestic Saving LFDI Logarithm of Foreign Direct Investment OLS Ordinary Least Squares PP Phillips-Perron test VECM Vector Error Correction Method WTO World Trade Organization VIETNAM ECONOMIC GROWTH AND SAVING ANALYSIS IN THE PERIOD 1989 – 2012 Abstract This paper employs Autoregressive-Distributed Lag model to detect the cointegrating vectors amongst three variables that are gross domestic saving per capita, gross domestic product per capita and foreign direct investment being stationary at the mixture of I(0) and I(1) in the period 1989 – 2012 The results not support for the hypothesis as Solow (1956) states in which domestic saving and foreign direct investment are sources for domestic investments, then push economic growth The long run relationship amongst these indicators is consistent with some recent empirical studies for the case of developing countries That is, the long run causal direction running from economic growth to domestic saving Additionally, foreign direct investment does not cause economic growth in both short and long run, but tends to reduce domestic saving in long run Furthermore, combining the estimated results with some statistical measurements of demographic change in total population and financial market development gives to us evidences in which financial market in Vietnam is still weak, thus does not strengthen the channel of domestic private saving accumulation, especially from huge amount of rural population FDI has increased annually, but it only takes a small percentage in GDP As a result, there is no evidence to support for the nexus from saving to growth GDS increasing annually is explained by a rise of GDP, and perhaps by some positive demographic change in total population Last but not the least, in long run FDI tends to reduce domestic saving due to ineffective channel of domestic capital accumulation, especially for Vietnam stock market Then, it casts doubts on an assumption of Solow about the positive nexus running from saving to growth, especially for developing countries Chapter 1: Introduction In this chapter, the study presents the reason of choosing this kind of research for the specific case of Vietnam to analyze the interaction amongst domestic saving, foreign direct investment and economic growth in the period 1989 – 2012 1.1 Problem statement The central assumption of the Solow’s (1956) growth model is a positive nexus between saving rate and economic growth in which saving rate plays a role of conditional factor to push growth This model implies some policy implications for a country to concentrate on increasing saving, thus economic growth will increase in response However, inversely if there is a possibility of negative impact, a country shall focus on removing barriers to growth instead of accumulating saving There are some robust empirical findings about the positive association between saving and growth Nevertheless, this correlation does not imply a causal direction amongst them, and then this controversy is still unsolved Moreover, the debate concerning with the priority of policy implication for these indicators is an important issue raised at the current macroeconomics, as stated by Schmidt-Hebbel et al (1996) Hence, determining the causal direction of saving-growth linkage is crucial, and has many implications for policy makers in developing countries At the notion of growth to saving, it has been supported by some empirical evidences of several recent studies analyzing in developing countries For instance, in the research by Ramesh (2006), he finds that in the group of low-middle income countries there is mostly the causal direction from growth to saving Furthermore, the study by Pradeep, Pravakar, and Ranjan (2008) also shows the empirical evidences from countries – Bangladesh, India, Pakistan, Srilanka and Nepal – there is a causal direction from growth to saving, however, only Bangladesh with bi-directional causality In an open economy, two capital sources contributing to domestic investment are domestic and foreign saving With foreign saving, it is represented by some forms of international capital inflows Theoretically, these inflows are assumed to be additive, or supplement domestic saving to increase an amount of investment, hence to push economic growth This hypothesis is seemingly relevant in some certain capital inflows, especially to foreign direct investment (noted as FDI) because this inflow is more stable than the other components of foreign saving in term of predictability and volatility (Taylor, 1997 and Lipsey, 1999 cited in Maite, Ana and Vicente, 2004) In the research composed by Matie, Ana and Vicente (2004), they analyze the role of FDI to the nexus between economic growth and domestic saving in Mexico and find that FDI causes both GDP and GDS positively, hence to reinforce the relationship between economic growth and domestic saving However, this hypothesis does not hold in the study by Ahmad, Marwan and Salim (2002) for the case of Malaysia, Thailand and Philippines They find foreign savings affecting domestic saving negatively in both short and long run Moreover, in long run the causal direction runs from foreign to domestic saving, but the inverse direction does not exist In the history, Latin America crisis happened in Mexico after the peso was devaluated by the authorities in 1994, then quickly spreading to several other countries in this region Consequently, a huge amount of capital outflow left these countries into shambles However, on the contrary to that of Latin America countries, East Asian economies were not seriously affected by the crisis Then, based on this fact, some economists conclude that Latin America countries were more sensitive to the shift of investor sentiments and the fluctuation of international capital outflows than Asian countries This difference relates to domestic saving rate in which the higher the domestic saving in a country is, therefore can reduce the international capital outflows’ influence through a crisis Furthermore, in this period, an increase of foreign investment in East Asian countries correlated with a huge decrease of domestic saving rate in Latin America countries In the mid of year 1997, the Asian crisis started from Thailand, then influenced to several other countries Based on this fact, it casts doubts on the economic notion held earlier through Latin America crisis by some economists The Asian economies’ currencies were devaluated sharply, then leading to a massive outflow of foreign capital at the end of year 1997 On the other hand, the experience was similar with the case of Latin America crisis The retreat of foreign capital emphasizes the necessity of more domestic financing seen as a complement to that decline in capital As dramatized by Latin America and Asian crisis, therefore, it is important to analyze the nexus between domestic and foreign saving because foreign capital could be withdrawn as easily as it enters, consequently a domestic economy could be left into shambles through a crisis In recent years, Vietnam’s economy has succeeded in the new stage of development after opening the economy in 1988 Especially, Vietnam has been a member of ASEAN since 1995, and WTO since 2007 Then, its economic growth has increased rapidly in the period 1989 – 2012 Tariffs and trade barriers have been reduced step by step amongst members in ASEAN and WTO, therefore, Vietnam’s trade balance has been improved as well From a poor country with per capita income only equals to 290 U.S dollars a year in 1989, Vietnam has reached to nearly 1000 U.S dollars a year in 2012 Moreover, domestic saving rate and foreign direct investment in Vietnam have also increased annually through this period In detail, its gross domestic saving has changed from 0.837 to 26.9 billion U.S dollars, and FDI has also moved from 0.01 to 4.7 billion U.S dollars, respectively in 1989 and 2012 (World Bank Indicators, 2012) Optimum lag-length selection based on AIC for equation Dependent Variable: DLFDI Method: Least Squares Date: 09/03/14 Time: 20:12 Sample (adjusted): 1992 2012 Included observations: 21 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDP(-1) LFDI(-1) DLGDS(-1) DLGDS(-2) DLGDP(-1) DLGDP(-2) DLFDI(-1) DLFDI(-2) 5.257945 0.581454 -0.135269 -0.369457 -0.078175 0.771442 4.418955 8.281550 0.128082 0.312346 3.312308 0.551122 0.951735 0.164284 0.395039 0.250704 6.964296 6.500569 0.206230 0.169907 1.587396 1.055037 -0.142129 -2.248890 -0.197892 3.077108 0.634516 1.273973 0.621062 1.838330 0.1407 0.3140 0.8895 0.0460 0.8467 0.0105 0.5387 0.2289 0.5472 0.0931 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.676193 0.411261 0.230009 0.581943 7.854293 2.552321 0.072698 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.082999 0.299766 0.204353 0.701745 0.312300 2.867991 Dependent Variable: DLFDI Method: Least Squares Date: 09/03/14 Time: 20:12 Sample (adjusted): 1992 2012 Included observations: 21 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDP(-1) LFDI(-1) DLGDS(-1) DLGDS(-2) DLGDP(-1) DLGDP(-2) DLFDI(-1) 2.686276 -0.103946 0.884778 -0.388055 0.433882 0.565684 1.144748 8.375941 0.276764 3.286620 0.444286 0.846446 0.179494 0.306640 0.245568 7.370036 7.115714 0.207667 0.817337 -0.233963 1.045287 -2.161932 1.414959 2.303575 0.155325 1.177105 1.332728 0.4297 0.8190 0.3165 0.0515 0.1825 0.0399 0.8791 0.2620 0.2074 R-squared Adjusted R-squared S.E of regression Sum squared resid 0.576712 0.294520 0.251782 0.760730 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion 0.082999 0.299766 0.377020 0.824673 Log likelihood F-statistic Prob(F-statistic) 5.041285 2.043689 0.127614 Hannan-Quinn criter Durbin-Watson stat 0.474172 2.724471 Dependent Variable: DLFDI Method: Least Squares Date: 09/03/14 Time: 20:12 Sample (adjusted): 1992 2012 Included observations: 21 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDP(-1) LFDI(-1) DLGDS(-1) DLGDS(-2) DLGDP(-1) DLFDI(-1) 2.734238 0.088998 0.440747 -0.294228 0.400482 0.470989 6.257795 0.261587 3.334748 0.419023 0.768897 0.163192 0.309819 0.235432 6.041581 0.210318 0.819923 0.212394 0.573220 -1.802952 1.292633 2.000528 1.035788 1.243770 0.4270 0.8351 0.5763 0.0946 0.2186 0.0668 0.3192 0.2355 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.527838 0.273596 0.255489 0.848567 3.893939 2.076128 0.121168 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.082999 0.299766 0.391053 0.788967 0.477411 2.963486 Dependent Variable: DLFDI Method: Least Squares Date: 09/03/14 Time: 20:13 Sample (adjusted): 1991 2012 Included observations: 22 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDP(-1) LFDI(-1) DLGDS(-1) DLGDP(-1) DLFDI(-1) 4.842817 0.267582 0.037154 -0.317840 0.321122 9.543777 0.247111 3.340500 0.388485 0.719168 0.155571 0.278419 5.957044 0.174060 1.449728 0.688782 0.051662 -2.043061 1.153375 1.602100 1.419689 0.1677 0.5015 0.9595 0.0590 0.2668 0.1300 0.1761 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.404979 0.166971 0.273892 1.125248 1.486777 1.701534 0.188730 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.097257 0.300088 0.501202 0.848352 0.582980 2.363185 A.3 Granger Causality Analysis A.3.1 Long run coefficient equation estimation in form of equation Dependent Variable: LGDS Method: Least Squares Date: 08/05/14 Time: 17:52 Sample (adjusted): 1991 2012 Included observations: 22 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDS(-2) LGDP LFDI -0.951677 0.333711 0.041683 1.056886 -0.123968 1.294507 0.103050 0.095930 0.191612 0.059244 -0.735165 3.238332 0.434518 5.515771 -2.092517 0.4723 0.0048 0.6694 0.0000 0.0517 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.974970 0.969081 0.108414 0.199812 20.49898 165.5475 0.000000 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 4.828157 0.616556 -1.408998 -1.161034 -1.350585 2.063561 Optimum lag-length selection based on AIC for equation Dependent Variable: LGDS Method: Least Squares Date: 09/03/14 Time: 21:18 Sample (adjusted): 1991 2012 Included observations: 22 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDS(-2) LGDP LGDP(-1) LGDP(-2) LFDI LFDI(-1) LFDI(-2) -1.338778 0.213865 0.028654 -2.369643 6.308413 -2.699361 -0.142467 -0.045145 0.059690 1.507274 0.223974 0.124162 3.219322 5.547556 3.119132 0.110007 0.148712 0.077590 -0.888212 0.954863 0.230776 -0.736069 1.137152 -0.865421 -1.295074 -0.303571 0.769303 0.3906 0.3571 0.8211 0.4748 0.2760 0.4025 0.2178 0.7663 0.4555 R-squared Adjusted R-squared S.E of regression Sum squared resid 0.978760 0.965690 0.114205 0.169557 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion 4.828157 0.616556 -1.209552 -0.763216 Log likelihood F-statistic Prob(F-statistic) 22.30507 74.88221 0.000000 Hannan-Quinn criter Durbin-Watson stat -1.104408 2.082214 Dependent Variable: LGDS Method: Least Squares Date: 09/03/14 Time: 21:19 Sample (adjusted): 1991 2012 Included observations: 22 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDS(-2) LGDP LGDP(-1) LGDP(-2) LFDI LFDI(-1) -0.973239 0.361077 -0.015383 -2.156144 6.159165 -2.902911 -0.172365 0.043904 1.409414 0.114681 0.108561 3.160237 5.462744 3.062249 0.101402 0.091992 -0.690527 3.148536 -0.141696 -0.682273 1.127486 -0.947967 -1.699818 0.477257 0.5012 0.0071 0.8893 0.5062 0.2785 0.3592 0.1113 0.6405 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.977793 0.966690 0.112528 0.177276 21.81536 88.06261 0.000000 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 4.828157 0.616556 -1.255942 -0.859199 -1.162481 2.258088 Dependent Variable: LGDS Method: Least Squares Date: 09/03/14 Time: 21:19 Sample (adjusted): 1991 2012 Included observations: 22 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDS(-2) LGDP LGDP(-1) LFDI LFDI(-1) -1.054291 0.358857 0.017258 -0.603240 1.611369 -0.152952 0.051972 1.402057 0.114269 0.102608 2.693369 2.603883 0.098976 0.091287 -0.751960 3.140463 0.168192 -0.223972 0.618833 -1.545337 0.569329 0.4637 0.0067 0.8687 0.8258 0.5453 0.1431 0.5776 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.976368 0.966915 0.112147 0.188655 21.13102 103.2879 0.000000 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 4.828157 0.616556 -1.284638 -0.937488 -1.202860 2.143463 Dependent Variable: LGDS Method: Least Squares Date: 09/03/14 Time: 21:19 Sample (adjusted): 1991 2012 Included observations: 22 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDS(-2) LGDP LGDP(-1) LFDI -0.834951 0.363877 0.024841 -0.915932 1.912897 -0.109992 1.319307 0.111496 0.099568 2.580483 2.495026 0.062684 -0.632871 3.263588 0.249490 -0.354946 0.766684 -1.754703 0.5358 0.0049 0.8062 0.7273 0.4544 0.0984 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.975857 0.968312 0.109753 0.192731 20.89585 129.3443 0.000000 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 4.828157 0.616556 -1.354168 -1.056611 -1.284073 2.121864 Dependent Variable: LGDS Method: Least Squares Date: 09/03/14 Time: 21:20 Sample (adjusted): 1991 2012 Included observations: 22 after adjustments Variable Coefficient Std Error t-Statistic Prob C LGDS(-1) LGDS(-2) LGDP LFDI -0.951677 0.333711 0.041683 1.056886 -0.123968 1.294507 0.103050 0.095930 0.191612 0.059244 -0.735165 3.238332 0.434518 5.515771 -2.092517 0.4723 0.0048 0.6694 0.0000 0.0517 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.974970 0.969081 0.108414 0.199812 20.49898 165.5475 0.000000 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 4.828157 0.616556 -1.408998 -1.161034 -1.350585 2.063561 A.3.2 VECM based on the error terms of the conditional ARDL(2, 0, 0) Dependent Variable: DLGDS Method: Least Squares Date: 08/05/14 Time: 17:54 Sample (adjusted): 1992 2012 Included observations: 21 after adjustments Variable Coefficient Std Error t-Statistic Prob C DLGDP DLFDI ECT(-1) 0.001472 2.068093 -0.104839 -0.706134 0.165790 3.052517 0.114398 0.323348 0.008879 0.677504 -0.916444 -2.183822 0.9930 0.5072 0.3723 0.0433 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.277102 0.149532 0.140489 0.335531 13.63621 2.172156 0.128800 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.108279 0.152340 -0.917734 -0.718777 -0.874555 1.499069 Wald Test: Equation: Untitled Test Statistic F-statistic Chi-square Value 4.769077 4.769077 df Probability (1, 17) 0.0433 0.0290 Value Std Err Null Hypothesis Summary: Normalized Restriction (= 0) C(4) Restrictions are linear in coefficients -0.706134 0.323348 Wald Test: Equation: Untitled Test Statistic F-statistic Chi-square Value 0.459012 0.459012 df Probability (1, 17) 0.5072 0.4981 Value Std Err 2.068093 3.052517 Null Hypothesis Summary: Normalized Restriction (= 0) C(2) Restrictions are linear in coefficients Wald Test: Equation: Untitled Test Statistic F-statistic Chi-square Value 0.839870 0.839870 df Probability (1, 17) 0.3723 0.3594 Value Std Err Null Hypothesis Summary: Normalized Restriction (= 0) C(3) Restrictions are linear in coefficients -0.104839 0.114398 A.3.3 Diagnostics tests of VECM Ramsey RESET Test: F-statistic Log likelihood ratio 0.394801 0.511886 Prob F(1,16) Prob Chi-Square(1) 0.5387 0.4743 Test Equation: Dependent Variable: DLGDS Method: Least Squares Date: 08/05/14 Time: 17:56 Sample: 1992 2012 Included observations: 21 Variable Coefficient Std Error t-Statistic Prob C DLGDP DLFDI ECT(-1) FITTED^2 0.018412 0.951269 -0.070442 -0.320143 2.386702 0.170961 3.580660 0.128713 0.696988 3.798474 0.107698 0.265669 -0.547277 -0.459324 0.628332 0.9156 0.7939 0.5917 0.6522 0.5387 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.294510 0.118138 0.143058 0.327451 13.89215 1.669820 0.205948 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.108279 0.152340 -0.846872 -0.598176 -0.792898 1.357751 Heteroskedasticity Test: White F-statistic Obs*R-squared Scaled explained SS 0.639758 2.130354 1.154204 Prob F(3,17) Prob Chi-Square(3) Prob Chi-Square(3) 0.5998 0.5458 0.7640 Test Equation: Dependent Variable: RESID^2 Method: Least Squares Date: 08/05/14 Time: 17:56 Sample: 1992 2012 Included observations: 21 Variable Coefficient Std Error t-Statistic Prob C DLGDP^2 DLFDI^2 ECT(-1)^2 0.005824 2.834862 -0.017715 0.326384 0.013438 4.022488 0.024542 0.339890 0.433395 0.704753 -0.721830 0.960264 0.6702 0.4905 0.4802 0.3504 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.101445 -0.057123 0.021646 0.007965 52.91296 0.639758 0.599782 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.015978 0.021053 -4.658377 -4.459420 -4.615198 1.318549 Breusch-Godfrey Serial Correlation LM Test: F-statistic Obs*R-squared 1.762745 3.996401 Prob F(2,15) Prob Chi-Square(2) 0.2053 0.1356 Test Equation: Dependent Variable: RESID Method: Least Squares Date: 08/05/14 Time: 17:56 Sample: 1992 2012 Included observations: 21 Presample missing value lagged residuals set to zero Variable Coefficient Std Error t-Statistic Prob C DLGDP DLFDI ECT(-1) RESID(-1) RESID(-2) 0.033104 -0.576611 -0.011823 -0.639362 0.613442 -0.381206 0.160102 2.941687 0.128708 0.549759 0.448616 0.262876 0.206766 -0.196014 -0.091856 -1.162984 1.367409 -1.450134 0.8390 0.8472 0.9280 0.2630 0.1916 0.1676 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.190305 -0.079594 0.134580 0.271678 15.85273 0.705098 0.628489 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 1.98E-18 0.129524 -0.938355 -0.639920 -0.873587 1.900160 Series: Residuals Sample 1992 2012 Observations 21 Mean Median Maximum Minimum Std Dev Skewness Kurtosis Jarque-Bera Probability 1.98e-18 0.012558 0.258575 -0.268860 0.129524 -0.147083 2.653490 0.180777 0.913576 -0.3 -0.2 -0.1 -0.0 0.1 0.2 0.3 12 -4 -8 -12 1996 1998 2000 2002 CUSUM 2004 2006 2008 5% Significance 2010 2012 1.6 1.2 0.8 0.4 0.0 -0.4 1996 1998 2000 2002 2004 CUSUM of Squares 2006 2008 2010 2012 5% Significance A.3.4 Short-run coefficient estimation Lag length criteria table VAR Lag Order Selection Criteria Endogenous variables: DLGDS DLGDP DLFDI Exogenous variables: C Date: 08/05/14 Time: 18:05 Sample: 1989 2012 Included observations: 20 Lag LogL LR FPE AIC SC HQ 71.24097 78.41364 82.74040 87.35824 NA* 11.47627 5.624783 4.617837 2.18e-07* 2.66e-07 4.59e-07 8.71e-07 -6.824097* -6.641364 -6.174040 -5.735824 -6.674738* -6.043925 -5.128521 -4.242225 -6.794941* -6.524738 -5.969944 -5.444258 * indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion In form of equation 12 with lag 0, but without ECT Dependent Variable: DLGDP Method: Least Squares Date: 08/05/14 Time: 18:06 Sample (adjusted): 1990 2012 Included observations: 23 after adjustments Variable Coefficient Std Error t-Statistic Prob C DLGDS DLFDI 0.054680 -0.004729 -0.004449 0.003103 0.009904 0.003257 17.62169 -0.477531 -1.366006 0.0000 0.6382 0.1871 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.085932 -0.005475 0.012195 0.002974 70.32701 0.940100 0.407181 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.052880 0.012161 -5.854523 -5.706415 -5.817274 0.773277 Wald Test: Equation: Untitled Test Statistic F-statistic Chi-square Value 0.228035 0.228035 df Probability (1, 20) 0.6382 0.6330 Value Std Err Null Hypothesis Summary: Normalized Restriction (= 0) C(2) Restrictions are linear in coefficients -0.004729 0.009904 Wald Test: Equation: Untitled Test Statistic F-statistic Chi-square Value df 1.865973 1.865973 Probability (1, 20) 0.1871 0.1719 Value Std Err Null Hypothesis Summary: Normalized Restriction (= 0) C(3) -0.004449 0.003257 Restriction are linear in coeffiecients In form of equation 13 with lag 0, but without ECT Dependent Variable: DLFDI Method: Least Squares Date: 08/05/14 Time: 18:11 Sample (adjusted): 1990 2012 Included observations: 23 after adjustments Variable Coefficient Std Error t-Statistic Prob C DLGDS DLGDP 1.386815 -0.828929 -19.18281 0.768105 0.627217 14.04299 1.805502 -1.321598 -1.366006 0.0861 0.2012 0.1871 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.149762 0.064738 0.800783 12.82508 -25.91853 1.761409 0.197427 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat 0.258729 0.828034 2.514655 2.662763 2.551904 0.613484 Wald Test: Equation: Untitled Test Statistic F-statistic Chi-square Value 1.746622 1.746622 df Probability (1, 20) 0.2012 0.1863 Value Std Err Null Hypothesis Summary: Normalized Restriction (= 0) C(2) -0.828929 0.627217 Restrictions are linear in coefficients Wald Test: Equation: Untitled Test Statistic F-statistic Chi-square Value 1.865973 1.865973 df Probability (1, 20) 0.1871 0.1719 Value Std Err Null Hypothesis Summary: Normalized Restriction (= 0) C(3) Restrictions are linear in coefficients -19.18281 14.04299 C Vietnam Macroeconomic Data Year 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 PCGDS PCGDP FDI 12.93 292.19 0.01 10.03 301.31 0.55 31.20 313.45 0.82 45.35 334.55 1.10 55.04 355.35 1.74 60.96 380.32 3.22 73.97 409.83 2.53 75.56 440.99 3.13 94.67 469.56 2.89 106.35 489.06 2.27 125.06 504.78 1.90 128.32 531.86 1.59 151.54 557.68 1.62 140.81 586.09 1.72 140.28 619.29 1.69 132.52 658.03 1.75 212.93 699.49 1.95 234.27 740.04 2.23 203.88 784.25 5.72 187.33 819.88 6.74 229.12 855.07 5.27 247.47 900.49 5.40 242.55 946.80 4.56 303.16 985.98 4.70 GDP 18.93 19.89 21.08 22.90 24.75 26.94 29.51 32.26 34.89 36.90 38.66 41.29 43.85 46.62 49.83 53.59 57.63 61.65 66.05 69.79 73.56 78.28 83.17 87.53 GDS 0.84 0.66 2.10 3.10 3.83 4.32 5.33 5.53 7.03 8.02 9.58 9.96 11.91 11.20 11.29 10.79 17.54 19.52 17.17 15.95 19.71 21.51 21.31 26.91 MVA 2.40 2.25 2.37 2.70 2.96 3.23 3.67 4.17 4.70 5.18 5.60 6.25 6.96 7.77 8.67 9.61 10.85 12.30 13.82 15.17 15.59 16.90 18.76 19.84 AVA 5.98 6.04 6.17 6.60 6.82 7.05 7.38 7.71 8.04 8.33 8.76 9.17 9.46 9.85 10.22 10.68 11.12 11.55 12.00 12.57 12.81 13.23 13.76 14.13 IVA 4.85 4.96 5.35 6.03 6.79 7.70 8.75 10.01 11.28 12.22 13.15 14.48 15.72 16.86 18.44 20.27 21.98 23.58 25.31 26.36 27.94 29.94 31.94 33.77 RP 79.88 79.75 79.36 78.97 78.59 78.20 77.81 77.38 76.94 76.50 76.06 75.63 75.04 74.46 73.88 73.30 72.72 72.10 71.47 70.85 70.23 69.61 68.96 68.32 DR 43.39 43.12 42.86 42.61 42.33 41.97 41.52 40.97 40.32 39.59 38.81 38.00 37.16 36.30 35.43 34.56 33.69 32.83 31.99 31.22 30.56 30.03 29.66 29.42 PCGDS, PCGDP are per capita GDS and GDP respectively MVA, AVA and IVA are Manufacturing value added, Agricultural value added and Industrial value added series respectively (unit billion U.S dollars) FDI, GDP and GDS are measured in billion U.S dollars RP and DR are rural population and dependency ratio compared to total population Year Market capitalization of listed companies (% of GDP) Stocks traded, total value (% of GDP) Commercial bank branches (per 100,000) 2006 2007 2008 2009 2010 2011 2012 13.70 25.24 9.67 20.00 17.58 13.51 21.14 1.61 6.58 20.47 14.83 4.22 2.17 3.31 3.32 3.25 3.57 3.18 16.25 ... analyzing in developing countries For instance, in the research by Ramesh (2006), he finds that in the group of low-middle income countries there is mostly the causal direction from growth to saving. .. period, an increase of foreign investment in East Asian countries correlated with a huge decrease of domestic saving rate in Latin America countries In the mid of year 1997, the Asian crisis started... of saving- growth linkage is crucial, and has many implications for policy makers in developing countries At the notion of growth to saving, it has been supported by some empirical evidences of