LV Thạc sỹ_Sectoral compositions of growth and proverty reduction in Vietnam

National Economics University – Institute of Social Studies Vietnam –Netherlands Master Program in Development Economics SECTORAL COMPOSITIONS OF GROWTH AND POVERTY REDUCTION IN VIETNAM ABSTRACT This thesis examines the impact of sectoral economic growth on poverty reduction in Vietnam during the period from 1998 to 2008 It is found that an increase in the proportion of the agricultural sector will lead to higher poverty rate and the overall growth rate has a positive impact on poverty reduction in Vietnam during the review period These results support our hypothesis that the sectoral structure of economic growth affects poverty independently of the growth of the economy Moreover, these results also demonstrate that the process of restructuring the sectoral economy tending to reduce the proportion of agriculture and increase the share of industry will have positive impact on poverty reduction in the future ACKNOWLEDGMENTS: I sincerely thank the instructor for my thesis, Assoc Prof Dr …, for his enthusiasm, the careful, detailed and helpful instructions as well as encouragement for me to complete this thesis I would also like to thank Dr.The and members of the General Statistics Office Vietnam for the help of data Finally, I would like to thank all members and lecturers of VietnamNetherlands Center for Development Economics and Public Policy for their supports during the study and thesis process TABLE OF CONTENTS ABSTRACT CHAPTER INTRODUCTION 1.1 RESEARCH BACKGROUND 1.2 RESEARCH OBJECTIVES 1.3 THESIS STRUCTURE CHAPTER 10 LITERATURE REVIEW 10 2.1 THEORETICAL REVIEW 2.2 EMPIRICAL REVIEW 10 12 CHAPTER 22 OVERVIEW OF COMPOSITION OF ECONOMIC GROWTH 22 AND POVERTY REDUCTION IN VIETNAM 22 3.1 SECTORAL COMPOSITIONS OF ECONOMIC GROWTH IN VIETNAM 22 3.2 POVERTY REDUCTION IN VIETNAM 26 3.3 THE RELATIONSHIP BETWEEN SECTORAL COMPOSITION OF GROWTH AND POVERTY REDUCTION 34 3.3.1 Agricultural growth and poverty reduction 34 3.3.2 Industrial growth and poverty reduction 36 3.3.3 Service growth and poverty reduction 37 EMPIRICAL ANALYSIS 39 4.1 EMPIRICAL MODEL 39 4.2 EMPIRICAL RESULTS AND DISCUSSIONS 43 4.2.1 The relationship between sectoral compositions of growth and poverty reduction 43 4.2.2 The role of sectoral compositions of growth on poverty reduction in high- proportional industry provinces 51 CHAPTER 56 CONCLUSIONS AND POLICY RECOMMENDATIONS 56 5.2 POLICY RECOMMENDATIONS 58 5.2.1 Improving productivity and efficiency in the agricultural sector 58 5.2.2 Enhancing the economic transition process by increasing the proportion of industrial sector 60 REFERENCES 62 APPENDIX 65 LIST OF FIGURES AND TABLES ABSTRACT ABSTRACT CHAPTER INTRODUCTION 1.1 RESEARCH BACKGROUND 1.2 RESEARCH OBJECTIVES 1.3 THESIS STRUCTURE CHAPTER 10 LITERATURE REVIEW 10 2.1 THEORETICAL REVIEW 2.2 EMPIRICAL REVIEW 10 12 CHAPTER 22 OVERVIEW OF COMPOSITION OF ECONOMIC GROWTH 22 AND POVERTY REDUCTION IN VIETNAM 22 3.1 SECTORAL COMPOSITIONS OF ECONOMIC GROWTH IN VIETNAM 22 3.2 POVERTY REDUCTION IN VIETNAM 26 3.3 THE RELATIONSHIP BETWEEN SECTORAL COMPOSITION OF GROWTH AND POVERTY REDUCTION 34 3.3.1 Agricultural growth and poverty reduction 34 3.3.2 Industrial growth and poverty reduction 36 3.3.3 Service growth and poverty reduction 37 EMPIRICAL ANALYSIS 39 4.1 EMPIRICAL MODEL 39 4.2 EMPIRICAL RESULTS AND DISCUSSIONS 43 4.2.1 The relationship between sectoral compositions of growth and poverty reduction 43 4.2.2 The role of sectoral compositions of growth on poverty reduction in high- proportional industry provinces 51 CHAPTER 56 CONCLUSIONS AND POLICY RECOMMENDATIONS 56 5.2 POLICY RECOMMENDATIONS 58 5.2.1 Improving productivity and efficiency in the agricultural sector 58 5.2.2 Enhancing the economic transition process by increasing the proportion of industrial sector 60 REFERENCES 62 APPENDIX 65 ABSTRACT ABSTRACT CHAPTER INTRODUCTION 1.1 RESEARCH BACKGROUND 1.2 RESEARCH OBJECTIVES 1.3 THESIS STRUCTURE CHAPTER 10 LITERATURE REVIEW 10 2.1 THEORETICAL REVIEW 2.2 EMPIRICAL REVIEW 10 12 CHAPTER 22 OVERVIEW OF COMPOSITION OF ECONOMIC GROWTH 22 AND POVERTY REDUCTION IN VIETNAM 22 3.1 SECTORAL COMPOSITIONS OF ECONOMIC GROWTH IN VIETNAM 22 3.2 POVERTY REDUCTION IN VIETNAM 26 3.3 THE RELATIONSHIP BETWEEN SECTORAL COMPOSITION OF GROWTH AND POVERTY REDUCTION 34 3.3.1 Agricultural growth and poverty reduction 34 3.3.2 Industrial growth and poverty reduction 36 3.3.3 Service growth and poverty reduction 37 EMPIRICAL ANALYSIS 39 4.1 EMPIRICAL MODEL 39 4.2 EMPIRICAL RESULTS AND DISCUSSIONS 43 4.2.1 The relationship between sectoral compositions of growth and poverty reduction 43 4.2.2 The role of sectoral compositions of growth on poverty reduction in high- proportional industry provinces 51 CHAPTER 56 CONCLUSIONS AND POLICY RECOMMENDATIONS 56 5.2 POLICY RECOMMENDATIONS 58 5.2.1 Improving productivity and efficiency in the agricultural sector 58 5.2.2 Enhancing the economic transition process by increasing the proportion of industrial sector 60 REFERENCES 62 APPENDIX 65 CHAPTER INTRODUCTION 1.1 Research background One of Millennium Development Goals to 2015 proposed by UNDP is that poverty reduction has been the most prominent target for all countries over the world, especially for developing countries It was clearly found that the relationship between poverty reduction and economic growth was virtually admitted since early researches Lewis (1954) and Kuznets (1955) with U-shape curve indicated through graphs that with the high growth rate of the economy, increasing inequality must be accepted In contrast, in recent years, many evidences to the contrary for this relationship have been provided Chen, Datt and Ravallion (1993) found that with the cumulative distribution of consumption changing only negligibly, the numbers of the poor had been growing as the same rate as the population of the developing world Lopez revealed through his paper (2004) “It is difficult to argue that poverty reduction can be achieved through income redistribution policies in the absence of economic growth” This exposed a positive relation from economic growth to poverty reduction However, it is uneasy to give a conclusion that the sectoral composition of growth affects poverty reduction through economic development The answer to this problem was found differently from one country to another country Montalvo and Ravallion (2009) discovered the fact that China had great success in reducing poverty through economic growth and the primary sector for poverty reduction was agriculture rather than manufacturing or service sectors Another study by Suryahadi, Suryadarma and Sumarto (2009) investigating how poverty was reduced by differentiating growth in Indonesia indicated that the growth of services sector had the most powerful effect on the improvement of poverty situation in both rural and urban areas, followed by growth in rural agriculture Additionally, one noticeable conclusion was that industrial growth had a relatively small impact on poverty reduction in Indonesia These outcomes from investigating in Indonesia were mimic the results inquired from India Meanwhile, doing the research on Taiwan Warr and Wang (1999) provide the evidence that the growth of the industrial sector had the largest effect on reducing poverty From different findings discovered in various countries, many incomprehensible questions to point out which pattern of the economic growth has the biggest impact on poverty reduction, has arisen in developing countries Vietnam, one of the developing countries in the world, has experienced a high economic growth with a huge reduction in the incidence of extreme poverty since the economic renovation started in the mid1980s Reforms began primarily in the agriculture sector which accounted for nearly 40% of GDP and 70% of total employment Further, the reform included the stabilization of inflation and liberalization of foreign trade and investment As the result of the process, the average annual rate of economic growth from 1986 to 2009 was about 7% This high rate of economic growth led to a sharp reduction in poverty The poverty rate has been decreased from 58.1% in 1993 to 14.5% in 2008 Such a sharp drop in poverty is one of the greatest success stories in Vietnam’s economic development It appears that economic growth has been the key determinant of poverty reduction in Vietnam This result was proved by the research on the linkages between growth, poverty and inequality in Vietnam from 1996 to 2004 by Hoi Le (2008) The most robust result had been found in this research was the negative association between poverty rate and subsequence GDP growth rate Nevertheless, the growth process has been highly uneven across sectors and regions It is also evident that there is a sizeable variance in the impacts of a given rate of growth on poverty It is more important to go beyond cross-country average index to find out what is the robust element of growth-poverty relationship A question raised is that whether the pattern of Vietnam’s growth matters to poverty reduction Debate on how Vietnam deals with this question will affect the willingness of policy makers to pursue more rapid economic and poverty elimination in the future This thesis attempts to answer this question 1.2 Research objectives The primary objectives of this thesis are presented as three following points Firstly, I analyze and identify the situation and drivers of economic growth and poverty reduction in Vietnam Second, I conduct an empirical study on the impacts of sectoral composition of growth on poverty reduction in Vietnam These results will be very helpful and reliable for us to study deeply the growth structure and its effects on poverty Third, having relied on empirical outcomes, challenges of Vietnam poverty alleviation will be introduced and policies recommendations will be raised My thesis is attempted to find the answer to the central question “Does the sectoral composition of growth affect the poverty reduction independently of the aggregate rate of growth?” The sub-research questions are “Which is the sector having the most impact on poverty reduction?” and “What other factors affect poverty reduction?” According to the questions above, the hypothesis of the research is as follows: “Sectoral compositions of growth have effects on poverty reduction independently of the aggregate rate of growth.” 1.3 Thesis structure The thesis includes four chapters The first chapter provides an introduction on the importance of studying the impact of the sectoral structure of growth to poverty reduction in Vietnam Chapter focuses on the review of previous research about the structure of growth and poverty reduction In this chapter, empirical studies on both developed and developing countries will be reviewed to show that there are different results between nations This chapter also clarifies the theoretical relationship between composition growth and poverty reduction This theoretical model will be severed as the theoretical background for the empirical study in Chapter Chapter provides an overview of economic growth in Vietnam in the period 1998 to 2008 in three sectors: agriculture, service and industry In addition, this chapter analyzes the specific situation of poverty as well as Vietnam's achievements in the implementation of poverty reduction policies Chapter provides an empirical analysis on the influence of growth structure on poverty reduction in Vietnam Based on data from Vietnam General Statistics Office (GSO) the impact of the share of agriculture, industry and services on poverty reduction in Vietnam during the period 1998-2008 will be clarified The last chapter offers general conclusions and provides some policy recommendations to accelerate poverty reduction in Vietnam in the future CHAPTER LITERATURE REVIEW The relationship between economic growth and poverty reduction is virtually accepted, however, whether which composition of economic growth has the closest relationship to poverty alleviation seems to generate more controversy On discussion of these matters, there were many different points of view and this chapter reviews theoretical and empirical studies to give the answer to this issue 2.1 Theoretical review The correlation between the distribution of income and economic growth was early proved by many theoretical studies Lewis (1954) was the first who proposed dual-sector model based on assumption that developing countries had dual economies with traditional agriculture sector and modern industry sector In his model, the agricultural sector was labor-intensive and characterized by low wages and low productivity As the manufacturing sector was determined by higher wage than the agricultural sector, there is a demand for more workers initially and higher marginal productivity He confirmed that because all the wealthy of the economy were produced by industry sector agriculture should not be invested due to its low productivity Lewis also proved that in the end of this process, the agricultural and manufacturing wages rates are equals and the agricultural and manufacturing marginal product of labor are the same Hence, there would be no motivation for enlargement in industry sector when labor had no monetary incentive to turn from agriculture sector to industry sector This process would remain until there was no labor with low or zero- marginal product Then the rule was opposite The industry sector must raise wages to absorb agriculture labors, which improve the quality of the poor in the society Together with the rising income for the poor, the amount of benefit taken by the rich would be reduced relatively in comparison with the previous period According to Lewis’ theory, the living standard of the poor and whether the gap between classes in the society would change depended on transferring labors between two sectors The evident trend was found is that more 10 Table 11: Results of random effect of equation (3) Random-effects GLS regression obs = Number of 305 Group variable (i): id groups = Number of 61 R-sq: within = 0.2946 group: = Obs per between = 0.6780 avg = 5.0 overall = 0.5197 max = Random effects u_i ~ Gaussian chi2(8) = 215.59 corr(u_i, X) chi2 = Wald = (assumed) Prob > 0.0000 -logpov Coef Std Err z P>|z| [95% Conf Interval] -logS1 -.0727677 logS2 1.013759 logS3 -1.165999 loggdppc 8432915 4673857 1.80 0.071 5948133 3.66 0.000 6788378 0.24 0.809 1113015 -7.46 0.000 -1.48 0.139 1.759351 2.179572 3.345385 1644985 1.494996 -.8298875 -1.048034 -.6117407 loggini -12.82606 -5.51821 3.728565 1.789643 72 logS1GINI -1.420945 3842364 9210276 0.42 0.677 1.035977 3.88 0.000 1.371838 -0.07 0.947 2.189417 logS2GINI 1.989061 4.019538 6.050015 logS3GINI -.0908331 -2.779587 2.597921 _cons -1.073417 -5.09835 2.053575 -0.52 0.601 2.951515 -sigma_u 1585754 sigma_e 26069089 rho 27008126 (fraction of variance due to u_i) -Table 12: Results of fixed effect of equation (3) Fixed-effects (within) regression obs = Number of 305 Group variable (i): id groups = Number of 61 R-sq: within = 0.3003 group: = Obs per between = 0.6703 avg = 5.0 overall = 0.5096 max = F(8,236) = 12.66 corr(u_i, Xb) = -0.5142 = Prob > F 0.0000 73 -logpov Coef Std Err t P>|t| [95% Conf Interval] -logS1 0250542 1.208261 2.01 0.045 7446612 3.14 0.002 7368732 0.69 0.494 1529484 -5.62 0.000 4.087118 -1.56 0.121 2.391467 logS2 8725661 600592 2.339598 3.806631 logS3 -.9467109 loggdppc 5049785 1.956668 -.858951 -1.16027 -.5576325 loggini -14.41332 logS1GINI -1.506311 logS2GINI 2.040517 logS3GINI -2.618309 _cons -7.118169 -6.361424 1.690472 3939483 9645657 0.41 0.683 1.184218 3.69 0.000 1.481224 0.20 0.840 2.294207 4.373506 6.706496 299801 3.217911 -2.155765 2.518901 -0.86 0.393 2.80664 -sigma_u 22999979 sigma_e 26069089 rho 437697 (fraction of variance due to u_i) -F test that all u_i=0: F(60, 236) = 2.71 Prob > F = 74 0.0000 75 Table 13: Hausman Test between random and fixed effect of equation (3) Coefficients -(b) (B) fe re (b-B) sqrt(diag(V_b-V_B)) Difference S.E logS1 1.208261 8432915 364969 logS2 2.339598 2.179572 1600265 logS3 3771754 448015 5049785 1644985 34048 2866381 loggdppc -.858951 -.8298875 -.0290635 1049056 loggini -6.361424 logS1GINI 3939483 logS2GINI 4.373506 logS3GINI 299801 -5.51821 3842364 -.8432147 0097119 1.674019 2865225 4.019538 3539688 5736944 -.0908331 390634 558645 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(8) = (b-B)'[(V_b-V_B)^(-1)](b-B) = Prob>chi2 = 7.32 0.5026 76 Table 14: Results of random effect of equation (1) with highproportional industry provinces: Random-effects GLS regression Number of obs = 184 Group variable (i): id Number of groups = R-sq: within = 0.2581 37 Obs per group: = between = 0.7022 avg = overall = 0.5363 5.0 max = Random effects u_i ~ Gaussian Wald chi2(5) = Prob > chi2 = 127.53 corr(u_i, X) = (assumed) 0.0000 -logpov Coef Std Err z P>|z| [95% Conf Interval] -logS1 5189643 1477107 3.51 0.000 2294567 -.5746079 2601091 -2.21 0.027 -1.084412 -.1902265 2485553 -0.77 0.444 -.677386 -.5937984 1373851 -4.32 0.000 -.8630682 6761254 302182 2.24 0.025 0838595 8084718 logS2 -.0648035 logS3 296933 loggdppc -.3245287 loggini 77 1.268391 _cons 3.384299 9826389 3.44 0.001 1.458362 5.310236 -sigma_u 18017531 sigma_e 28578543 rho 28442396 (fraction of variance due to u_i) Table 15: Results of fixed effect of equation (1) with highproportional industry provinces: Fixed-effects (within) regression obs = 184 Group variable (i): id groups = Number of 37 R-sq: within = 0.2585 = Number of Obs per group: between = 0.6976 avg = 5.0 overall = 0.5340 max = 78 F(5,142) = 9.90 corr(u_i, Xb) = -0.1176 = Prob > F 0.0000 -logpov Coef Std Err t P>|t| [95% Conf Interval] -logS1 -.1866584 logS2 -1.344546 logS3 -.8223922 loggdppc 4874282 340997 1.43 0.155 3443949 -1.93 0.056 1.161515 -.6637428 0170608 -.1826707 3236129 -0.56 0.573 4570508 -.665501 1755882 -3.79 0.000 7539752 3919972 1.92 0.056 2.86 0.005 -1.012606 -.3183964 loggini -.0209292 _cons 1.166351 1.52888 3.777465 1.320872 6.38858 -sigma_u 21257518 sigma_e 28578543 rho 35620106 (fraction of variance due to u_i) -F test that all u_i=0: F(36, 142) = 2.68 Prob > F = 0.0000 79 80 Table 16: Hausman test between fixed and random effect of equation (1) with high-propotional industry province: (b) (B) (b-B) fe re Difference sqrt(diag(V_b-V_B)) S.E -logS1 4874282 5189643 -.0315361 -.6637428 -.5746079 -.0891349 -.1826707 -.1902265 0075558 3073443 logS2 2257235 logS3 2072331 loggdppc -.665501 -.5937984 -.0717026 109346 loggini 7539752 6761254 0778498 2496955 -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(5) = (b-B)'[(V_b-V_B)^(-1)](b-B) = Prob>chi2 = 1.35 0.9295 81 Table 17: Results of random of equation (3) with high-proportional industry province: Random-effects GLS regression obs = Number of 184 Group variable (i): id groups = Number of 37 R-sq: within = 0.3043 group: = Obs per between = 0.7131 avg = 5.0 overall = 0.5586 max = Random effects u_i ~ Gaussian chi2(8) = 140.08 corr(u_i, X) chi2 = Wald = (assumed) Prob > 0.0000 -logpov Coef Std Err z P>|z| [95% Conf Interval] -82 logS1 1.119937 -.0357792 1.90 0.058 9756337 2.10 0.036 9917088 0.09 0.927 1409072 -4.59 0.000 2.275653 logS2 133216 5896618 2.045423 3.95763 logS3 -1.853452 loggdppc 0902619 2.033975 -.646339 -.922512 -.3701661 loggini -20.05954 logS1GINI -1.338076 logS2GINI 1.147123 logS3GINI -3.850133 _cons -7.891853 -7.980341 6.16297 -1.29 0.195 1.197527 0.84 0.399 1.710437 2.63 0.009 2.128153 0.15 0.880 4.098859 1.009033 3.356143 4.499518 7.851913 3209699 4.492073 -1.708093 3.155037 -0.54 0.588 4.475667 -sigma_u 18120614 sigma_e 27869372 rho 29713959 (fraction of variance due to u_i) 83 Table 18: Results of fixed effect of equation (3) with highproportional industry province: Fixed-effects (within) regression obs = Number of 184 Group variable (i): id groups = Number of 37 R-sq: within = 0.3097 group: = Obs per between = 0.7022 avg = 5.0 overall = 0.5480 max = F(8,139) = 7.80 corr(u_i, Xb) = -0.4987 = Prob > F 0.0000 -logpov Coef Std Err t P>|t| [95% Conf Interval] -logS1 1.440526 -.0215594 1.95 0.053 1.211797 2.26 0.026 1.074761 0.49 0.623 1793879 -4.03 0.000 2.902612 logS2 2.733672 3377341 5.12961 logS3 -1.59597 7394811 5290236 2.654017 loggdppc -.7230516 -1.077733 -.3683697 loggini -24.31108 -10.6153 6.926933 -1.53 0.128 3.08048 84 logS1GINI -1.598612 logS2GINI 1.496184 logS3GINI -3.293299 _cons -11.16675 9113369 1.26946 0.72 0.474 1.954385 2.74 0.007 2.367622 0.59 0.559 3.421286 5.360351 9.224519 1.38791 6.069118 -3.590263 3.83197 -0.94 0.350 3.986222 -sigma_u 246988 sigma_e 27869372 rho 43990508 (fraction of variance due to u_i) -F test that all u_i=0: F(36, 139) = 2.76 Prob > F = 0.0000 85 Table 19: Hausman Test between random and fixed effect of equation (3) with high-proportional industry provinces: Coefficients -(b) (B) fe re (b-B) sqrt(diag(V_b- V_B)) Difference S.E logS1 1.440526 1.119937 3205895 4462413 logS2 2.733672 2.045423 6882493 7187426 logS3 5290236 0902619 4387617 4142757 loggdppc -.7230516 -.646339 -.0767125 1110189 loggini -10.6153 -7.980341 -2.634958 3.16231 logS1GINI 9113369 1.009033 -.0976964 4212589 logS2GINI 5.360351 4.499518 8608329 9455303 logS3GINI 1.38791 3209699 1.06694 1.037592 -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(8) = (b-B)'[(V_b-V_B)^(-1)](b-B) = Prob>chi2 = 4.89 0.7688 86 ... OVERVIEW OF COMPOSITION OF ECONOMIC GROWTH 22 AND POVERTY REDUCTION IN VIETNAM 22 3.1 SECTORAL COMPOSITIONS OF ECONOMIC GROWTH IN VIETNAM 22 3.2 POVERTY REDUCTION IN VIETNAM. .. OVERVIEW OF COMPOSITION OF ECONOMIC GROWTH 22 AND POVERTY REDUCTION IN VIETNAM 22 3.1 SECTORAL COMPOSITIONS OF ECONOMIC GROWTH IN VIETNAM 22 3.2 POVERTY REDUCTION IN VIETNAM. .. OVERVIEW OF COMPOSITION OF ECONOMIC GROWTH 22 AND POVERTY REDUCTION IN VIETNAM 22 3.1 SECTORAL COMPOSITIONS OF ECONOMIC GROWTH IN VIETNAM 22 3.2 POVERTY REDUCTION IN VIETNAM