Intergenerational mobility of earnings and income among sons and daughters in Viet Nam

#% with two regression stages. The primary sample consists of observations on son-father or daughter-father pairs in which information on children’s economic outcome and socio-economic characteristics, and fathers’ socio-economic characteristics, denoted by (" , are available. However, because information on fathers’ real economic outcome is not available in this sample, the regression of children’s economic status on that of fathers cannot be done. Therefore, in the first stage a secondary sample of ‘potential’ fathers, that are male workers from another sample that includes both observations’ economic outcome and same socio-economic characteristics classified and coded as in the primary sample, is employed to generate a regression of ‘potential’ fathers’ economic outcome on their socio-economic characteristics variables. To predict ‘true’ fathers’ economic status in the primary sample, ‘true’ fathers’ socio-economic characteristics, (" , are plugged into the regression presented as the following equation: & " = ) (" (2) where & " represents fathers’ predicted economic outcome, and ) is the corresponding coefficients of (" estimated in the first stage. 12 Empirically, the predictor set of fathers’ economic outcome is probably education (Lefranc et al. 2010), or occupation (Fortin and Lefebvre 1998), or education and occupation (Björklund and Jantti 1997; Núñez and Miranda 2010; Ueda and Sun 2013), or education, occupation, and industry (Gong et al. 2012; Kim 2013), or education, occupation, and geographical region (Lefranc et al. 2014). This study uses the set of education, occupation, industry, and geographical region to predict fathers’ individual earnings. In the second stage, children’s economic outcome is regressed on fathers’ imputed economic outcome. From this regression, #% that is IGE of children’s economic status with respect to their fathers’ economic success is obtained in this study. B. Transition Mobility Matrix Approach The transition matrix approach is a complementary method to the least squares regression approach, and it is also useful to examine the pattern of intergenerational mobility. A transition matrix of mobility indicates the possibility that an adult son or daughter changes his or her position from the economic outcome distribution relative to the position of their parents. The distribution is often presented in quartiles or deciles. This study uses the quartile matrices of mobility to express the mobility patterns of earnings and income across generations. To do this, a father’s and a child’s economic outcome are divided into four equal-sized groups and ranked orderly. The first quartile is indexed for the bottom quartile of those who are in the range from the 0th to 13 25th percentile while the fourth quartile is denoted for the top quartile of those who are in the range between the 75th and 100th percentile. IV. EMPIRICAL RESULTS A. First-Stage Results The analysis of the first-stage regression focuses on the estimates for these socio- economic characteristics because these are parameters of interest. The results are presented in Table 2. Accordingly, the model has a R2 of 0.19, which suggests that about 19% of the variation in the log of individual earnings of ‘potential’ fathers can be explained by these socio-economic characteristics. In Table 2, it can be seen that wage differentials occur among categories within each group as well as across groups. For example, tertiary generates the highest returns with 56.7% compared to non-diploma or primary (the omitted variable) from education group while two categories utilities and construction yield the highest and the lowest returns with 19.7% higher and 28.6% lower than mining (the omitted variable) respectively from industry group. Moreover, education and geographical region groups have larger variations on male workers’ individual earnings rather than occupation and industry. This can be explained by the accretion of wage differentials along with increasing returns to education (Imbert 2013, Liu 2006), and aggrandized earnings gaps among different geographical areas (van de Walle and Gunewardena 2001; World Bank 2014) in Vietnam over last two decades. It is important to note that age and age-squared are included in the group of independent variables in the first-stage model. However, its estimated coefficients are 14 not used to generate missing values of the log of ‘true’ fathers’ individual earnings in the primary samples because ‘true’ fathers’ individual earnings imputed must be a proxy for permanent rather than short-run outcome. B. Empirical Results for Sons Baseline Intergenerational Elasticity for Sons In Table 3, it can be seen that the baseline IGE estimates for sons are all statistically significant at the level of 1% for both individual earnings and individual income. In Column 1, an IGE estimate of 0.36 is found for individual earnings. Meanwhile, an IGE estimate of 0.39 is found for individual income in Column 2. These IGE estimates meaningfully point out that a 10% difference in fathers’ individual earnings likely leads to roughly 3.6% and 3.9% differences in sons’ individual earnings and individual income, respectively. These results also indicate that the baseline IGE estimate for individual income is higher than that for individual earnings. This is reasonable because a son’s individual income equals his individual earnings plus other adjunct incomes, the marginal effect of his father’s individual earnings on his individual income equals the sum of the marginal effect of his father’s individual earnings on his individual earnings and the marginal effect of his fathers’ individual earnings on his other additional income. Compared to other countries, these baseline IGE estimates for Vietnamese sons are ranked as the intermediate levels. These findings are relatively similar to the previous findings such as 0.42 in Spain (Cervini-Plá 2014), 0.40 in South Korea (Kim 2013), 0.35 in Japan (Lefranc et al. 2014), and 0.40 in French (Lefranc and Trannoy 2005). 15 These IGE results are apprently lower than those in some other countries such as 0.62 in South Africa (Piraino 2015), 0.60 in Brazil (Ferreira and Veloso 2006), 0.63 in urban China (Gong et al. 2012), 0.57 in Chile (Núñez and Miranda 2010), and 0.50 in Italy (Mocetti 2007, Piraino 2007). Transition Mobility Matrix for Sons Table 4 shows the father-to-son mobility of the quartiles from their individual earnings distributions. Focusing on the diagonal terms, it can be observed that the proportions for sons to be in the top and bottom as same as their fathers’ positions are nearly equal. For example, 39.76% of sons remain in the top quartile as their fathers, and 37.08% of sons have the same position as their fathers’ in the bottom quartile. The figures also indicate an almost symmetric pattern of mobility between the upward mobility from the bottom quartile to the top one, and the downward mobility from the top quartile to the bottom one. These figures evidently affirm the intermediate degree of mobility across generations for sons’ individual earnings as shown in the baseline IGE estimates. The pattern is the same for individual income and presented in Table A1 of Appendices. C. Empirical Results for Daughters Baseline Intergenerational Elasticity for Daughters Table 5 shows the baseline IGE estimates for daughters. The baseline IGE estimate of 0.28 is found for individual earnings in Column 1. This IGE degree manifests that a 16 10% difference in fathers’ individual earnings is likely to result in a 2.8% variation in daughters’ individual earnings. When the dependent variable is individual income, the IGE estimate is 0.33 as in Column 2. This figure implicates that a 10% variation in fathers’ individual earnings is likely to lead to a 3.3% difference in daughters’ individual income in Vietnam. The baseline IGE estimate for individual income is relatively 17.25% higher than that for individual earnings. These IGE estimates for Vietnamese daughters’ individual earnings and individual income explicitly demonstrate the average levels of intergenerational mobility compared to other countries. These average degrees of intergenerational mobility in Vietnam are nearly analogous to the estimates of around 0.39 in Spain (Cervini-Plá 2014), 0.35 in Japan (Lefranc et al. 2014), and 0.4 in South Korea (Ueda 2013). Meanwhile, some countries have lower IGE estimates for daughters than that of Vietnam such as 0.25 from Sweden (Hirvonen 2008). Also, it can be recognized that the patterns of intergenerational mobility of earnings and income are same for both Vietnamese sons and daughters. Particularly, the degree of persistence between children’s individual income and fathers’ individual earnings is higher than that between children’s individual earnings and fathers’ individual earnings. Importantly, daughters have the smaller degrees of economic outcome persistence from fathers’ background than sons for all two measures of economic outcome although these gaps are not considerable. Specifically, the baseline IGE estimates for sons and daughters are respectively 0.36 and 0.28 for individual earnings, and 0.39 and 0.33 for individual income. 17 This finding is similar to estimates from previous studies. For example, Chadwick and Solon (2002) find the estimates of 0.54 and 0.43 for American sons and daughters. Nilsen et al. (2012) conclude the IGE coefficients are between 0.16 and 0.34 for sons, and between 0.12 and 0.23 for daughter in Norway. On the contrary, sons is more mobile than daughters in some other countries. For example, Lefranc et al. (2014) find the baseline IGE estimates for sons are close to 0.34 while the corresponding figures for daughters are nearly 0.39 although the difference between these baseline estimates is small in Japan. Transition Mobility Matrix for Daughters Regarding the transition mobility matrix for daughters, Table 6 presents the changing mobility patterns of daughters’ position on individual earnings compared to their fathers’ individual earnings. In general, the transition matrix for individual earnings mobility for daughters is relatively symmetric, and it is analogously similar to that for sons. This transition matrix also provides evidence on the modest difference of degree of mobility across generations between sons and daughters as shown from the baseline IGE estimates. Nearly one third of daughters in the primary sample have the same top and bottom quartiles as their fathers with 37.13% and 31.01%, respectively. Moreover, the proportion of daughters whose fathers in the top quartile moves downwardly to the bottom quartile is 20.25%, and the rate of upwardly mobile daughters to the top quartile from their fathers’ bottom quartile is 15.57%. The result of the transition 18 mobility for individual income is the same as that for individual earnings and presented in Table A2 of Appendices. V. ROBUSTNESS CHECKS A. Robustness Check of IGE Estimates to Different First-Stage Model Specifications As noted from the literature, the TS2SLS estimator may endogenously biased because the socio-economic characteristics employed to predict fathers’ economic outcome probably have a direct impact on children’s economic outcome. Moreover, the magnitude of the bias depends on the set of socio-economic characteristics used to predict fathers’ economic outcome. Therefore, it is necessary to investigate the robustness of the baseline IGE estimates to the different sets of first-stage predictors. Analysis for Sons The full sample of sons is used to estimate the IGEs. Table 7 presents the results for fifteen cases in which different sets of fathers’ individual earnings predictors are used in the first stage model. Firstly, Column 1 reports the results of robustness checks for the IGE estimates of sons’ individual earnings with respect to their fathers’ individual earnings. The estimated coefficients of IGE are all statistically significant at 1%. The IGE estimates using the different sets of fathers’ economic outcome predictors modestly vary around the baseline IGE estimate of 0.36 (education, occupation, industry, and geographical region). In particular, the IGE estimates are between 0.26 (occupation and industry) 19 and 0.40 (occupation and geographical region). These extreme IGE estimates are smaller with a maximum proportion of 26.87% or higher with a maximum proportion of 9.70% than the baseline IGE estimate. When using an individual predictor in the first-stage model, the results from cases 1–4 in Column 1 indicate that the estimator with education generates the largest IGE with an estimate of 0.37 while that with industry produces the smallest IGE with an estimate of 0.27. Secondly, the robustness check for sons’ individual income is shown in Column 2. The coefficients of the IGE estimates in all cases are statistically significant at 1%. The results demonstrate that when changing the set of socio-economic characteristics for predicting fathers’ individual earnings, the IGE estimates insignificantly alter around the baseline value of 0.39 (education, occupation, industry, and geographical region). Specifically, the minimum IGE estimate is 0.32 (geographical region), and the maximum IGE estimate is 0.43 (occupation and region). When using an individual predictor in the first stage model as shown in cases 1–4, the estimator with education produces the largest IGE of 0.40 while that with geographical region creates the smallest IGE estimate of 0.32. However, the gap between these two extreme IGE estimates is relatively small with a degree of 0.08. The above analysis shows that the baseline IGE estimates for sons are highly robust. The degrees of the IGE estimates when changing the set of fathers’ individual earnings predictors is varied insignificantly for both sons’ individual earnings and individual income. 20 Analysis for Daughters The full sample of daughters is used to check the robustness for the IGE estimates to the first-stage model specifications. The results are presented in Table 8. Firstly, Column 1 shows that the IGE estimates for individual earnings in different cases vary around the baseline IGE estimate of 0.28 (education, occupation, industry, and geographical region). Specifically, the estimates span from 0.24 (education) to 0.41 (occupation, and geographical region). All estimated coefficients are statistically significant at 1%. Compared to the baseline estimate, the IGE estimates can be smaller with a maximum proportion of 16.55%, or higher with a maximum proportion of 42.96%. When using only one sole socio-economic characteristic in the first stage model, the results from cases 1–4 indicate that the estimator with occupation produces the largest IGE estimate of 0.38 while that with education yields the smallest IGE of 0.24. The result is different with the finding for in which education produces the largest IGE estimate. Secondly, the robustness check for daughters’ individual income is provided in Column 2. Accordingly, all IGE estimates are statistically significant at 1%. The IGE estimates from the various first-stage specifications fluctuate around the baseline estimate of 0.33 (education, occupation, industry, and geographical region). In particular, the IGE estimates vary from 0.27 (education) to 0.48 (occupation, and geographical region). Hence, these IGE estimates are higher or smaller than the baseline estimate with a maximum proportion of 43.24% or 18.02%, respectively. 21 When using the sole predictor, the specification with occupation produces the largest IGE estimate of 0.43 while the estimator with education yields the smallest IGE estimate of 0.27. This result is different for sons’ individual income where the estimator with education produces the largest IGE and the estimate with geographical region is the smallest one. B. Robustness Check of IGE Estimates to Different Age Ranges From the existing literature, changes in children’s age ranges in the primary sample may lead to the variation of the IGE estimates (Grawe 2006; Haider and Solon 2006). In this section, the sensitivity of the IGE estimates to different sub-samples of various age intervals is analyzed for both sons and daughters. Analysis for Sons Table 9 presents the IGE estimates for sons in various sub-samples of different age ranges. The IGE estimates are reported for two measures of sons’ economic outcome including individual earnings in Column 1, and individual income in Column 2. There are three age intervals considered including 25–29 in Panel A, 30–34 in Panel B, and 35–54 in Panel C. The IGE coefficients are all statistically significant at 1%. The results explicitly provide evidence on the variation of IGE estimates across sub-samples. In Column 1, the IGE estimates span from 0.34 in the 25–29 sub-sample in Panel A to 0.48 in the 35–54 sub-sample in Panel C for individual earnings. The result in Column 2 gives an analogous pattern with a range of the IGE estimates between 0.36 in the 25–29 sub-sample and 0.49 in the 35–54 sub-sample for 22 individual income. The IGE estimates are generally larger in the older sub-samples than the younger sub-samples. In addition, using a rule of age selection from Haider and Solon (2006), a sub- sample of 450 sons aged 30–50 is formed to achieve the IGE estimates with the minimized lifecycle bias as shown in Panel D. In particular, the IGE estimates for individual earnings and individual income are respectively 0.41 and 0.47. These estimates are all statistically significant at 1%. These estimates are 14.13% and 18.78% higher than the baseline IGE estimates, respectively for individual earnings and individual income. Therefore, a sub-sample of sons aged around 40 is less intergenerationally mobile than the full sample of sons aged 25–54 for both individual earnings and individual income. Analysis for Daughters Table 10 reports the IGE estimates using sub-samples of daughters with different age ranges, including 25–29 in Panel A, and 30–47 in Panel B. The IGE coefficients are all statistically significant at 1%. The results show that changes in the IGE estimates of the different age intervals for daughters are same as the results for sons. The IGE estimates rise from 0.24 to 0.44 for individual earnings, and from 0.29 to 0.48 for individual income. There are differences among the IGE estimates from these two sub-samples. Specifically, the increased percentages of the IGE estimates in the 30–34 sub-sample compared to the 25–29 sub-sample are 82.08% and 66.21% for individual earnings and individual income. 23 When applying Haider and Solon’s (2006) rule of age selection, there is a sample limited to 182 daughters aged 30–50. The corresponding IGE estimates are found to be 0.40 and 0.45 for individual earnings and individual income as shown in Panel C. In comparison with the baseline results, these lifecycle-minimized IGE estimates are higher. In particular, the IGE estimates increase from 0.28 to 0.40 for individual earnings, and from 0.33 to 0.45 for individual income, equivalent to the increased proportions of 41.90% and 43.23%, respectively. VI. CONCLUDING REMARKS This paper uses household survey data to investigate intergenerational mobility of earnings and income for sons and daughters in Vietnam. The baseline IGE estimates explicitly reveal that Vietnam has the intermediate degrees of individual earnings and individual income mobility across generations for both sons and daughters by the conventional international scale of intergenerational mobility as shown in Black and Devereux (2011), and Blanden (2013). These results indicate that Vietnam has comparetively the same mobile position as Japan (Lefranc et al. 2014), Taiwan (Kan et al. 2015), and South Korea (Kim 2013) in Asia. Meanwhile, the results indicate that Vietnam is more mobile than other developing countries such as Brazil (Dunn 2007), and South Africa (Hertz 2001, Piraino 2015). The baseline results is highly robust when using various specifications of the first-stage model. The paper also finds the existence of age effects on the IGE estimates and this result is consistent with the literature. Apparently, this paper 24 provides more empirical evidence for the literature of intergenerational mobility in developing countries and Vietnam as well. Last three decades have witnessed the impressive transition of Vietnam’s economy from the planning system to the market-oriented one with the increasing integration into international economy (Irvin 1995). During this period, Vietnamese labor markets also have reformed and more actively functioned in the context of the emergence of other economic sectors including the private and the foreign investment sectors in addition to the state sector. The transition has created more jobs and economic opportunities for many Vietnamese workers (Nghiep and Quy 2000) to improve their earnings and income and escape poverty (Sakellariou and Fang 2014) relatively compared to their previous generations who had lived in an isolated economy. Therefore, many Vietnamse laborers have upwardly moved in the ladder of income compared to their parents’ economic status, and then the relative degree inequality of opportunity in Vietnam is not low compared to other developing countries which have the similar context of development like Vietnam. This is likely an appropriate explanation for the intermediate positions of intergenerational mobility for Vietnam found from this paper. REFERENCES Aaronson, Daniel, and Bhashkar Mazumder. 2008. “Intergenerational Economic Mobility in the US: 1940–2000.” Journal of Human Resources 43: 139–72. Adger, W. Neil. 1999. “Exploring income inequality in rural, coastal Viet Nam.” Journal of Development Studies 35: 96–119. 25 Angrist, Joshua D., and Alan B. Krueger. 1992. “The Effect of Age at School Entry on Educational Attainment: An Application of Instrumental Variables with Moments from Two Samples.” Journal of the American Statistical Association 87: 328–36. Asadullah, M. Niaz. 2012. “Intergenerational Wealth Mobility in Rural Bangladesh.” Journal of Development Studies 48: 1193–1208. Atkinson, A. B. 1981. “On Intergenerational Income Mobility in Britain.” Journal of Post Keynesian Economics 3: 194–218. Atkinson, A. B.; A.K. Maynard; and C. G. Trinder. 1983. Parents and Children: Incomes in Two Generations. London: Heinemann. Aughinbaugh, Alison. 2000. “Reapplication and Extension: Intergenerational Mobility in the United States.” Labour Economics 7: 785–96. Aydemir, Abdurrahman; Chen, Wen-Hao; and Miles Corak. 2009. “Intergenerational Earnings Mobility among the Children of Canadian Immigrants.” Review of Economics and Statistics 91: 377–97. Bauer, Philipp, and Regina Therese Riphahn. 2009. “Age at school entry and intergenerational educational mobility.” Economics Letters 103: 87–90. Bhattacharya, Debopam, and Bhashkar Mazumder. 2011. “A nonparametric analysis of black– white differences in intergenerational income mobility in the United States.” Quantitative Economics 2: 335–79. Binder, Melissa, and Christopher Woodruff. 2002. “Inequality and Intergenerational Mobility in Schooling: The Case of Mexico.” Economic Development and Cultural Change 50: 249–67. Björklund, Anders, and Laura Chadwick. 2003. “Intergenerational income mobility in permanent and separated families.” Economics Letters 80: 239–46. Björklund, Anders, and Markus Jäntti. 1997. “Intergenerational Income Mobility in Sweden Compared to the United States.” American Economic Review 87: 1009–18. Björklund, Anders, and Markus Jäntti. 2009. “Intergenerational income mobility and the role of family background.” In Oxford Handbook of Economic Inequality, edited by Brian Nolan, Timothy Smeeding, and Wiemer Salverda. New York: Oxford University Press. 26 Black, Sandra E., and Paul J. Devereux. 2011. “Recent Developments in Intergenerational Mobility.” In Handbook of Labor Economics, edited by Orley Ashenfelter, and David Card. Amsterdam: North-Holland. Blanden, Jo. 2013. “Cross-Country Rankings in Intergenerational Mobility: A Comparison of Approaches from Economics and Sociology.” Journal of Economic Surveys 27: 38–73. Bratberg, Espen; Øivind Anti Nilsen; and Kjell Vaage. 2005. “Intergenerational earnings mobility in Norway: Levels and trends.” Scandinavian Journal of Economics 107: 419–35. Cervini-Plá, María. 2014. “Intergenerational Earnings and Income Mobility in Spain.” Review of Income and Wealth: 1–17. doi:10.1111/roiw.12130 Chadwick, Laura, and Gary Solon. 2002. “Intergenerational Income Mobility among Daughters.” American Economic Review 92: 335–44. Charles, Kerwin Kofi, and Erik Hurst. 2003. “The Correlation of Wealth across Generations.” Journal of Political Economy 111: 155–82. Charles, Kerwin Kofi; Sheldon Danziger; Geng Li; and Robert Schoeni. 2014. “The Intergenerational Correlation of Consumption Expenditures.” American Economic Review 104, no. 5: 136–40. Chetty, Raj; Nathaniel Hendren; Patrick Kline; and Emmanuel Saez. 2014a. “Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States.” Quarterly Journal of Economics: 1–71. doi: 10.1093/qje/qju022 Chetty, Raj; Nathaniel Hendren; Patrick Kline; Emmanuel Saez; and Nicholas Turner. 2014b. “Is the United States Still a Land of Opportunity? Recent Trends in Intergenerational Mobility.” American Economic Review 104, no. 5: 141–47. Corak, Miles, and Andrew Heisz. 1999. “The Intergenerational Earnings and Income Mobility of Canadian Men: Evidence from Longitudinal Income Tax Data.” Journal of Human Resources 34: 504–33. Corak, Miles. 2006. “Do Poor Children Become Poor Adults? Lessons from a Cross Country Comparison of Generational Earnings Mobility.” IZA Discussion Papers No. 1993. Bonn: Institute for the Study of Labor (IZA). 27 Corak, Miles. 2013. “Income Inequality, Equality of Opportunity, and Intergenerational Mobility.” Journal of Economic Perspectives 27: 79–102. Dearden, Lorraine; Stephen Machin; and Howard Reed. 1997. “Intergenerational Mobility in Britain.” Economic Journal 107: 47–66. Dunn, Christopher E. 2007. “The intergenerational transmission of lifetime earnings: evidence from Brazil.” The B.E. Journal of Economic Analysis and Policy. doi:10.2202/1935-1682.1782. Emran, M. Shahe, and Forhad Shilpi. 2011. “Intergenerational Occupational Mobility in Rural Economy: Evidence from Nepal and Vietnam.” Journal of Human Resources 20: 427–58. Ferreira, Sergio Guimaraes, and Fernando A. Veloso. 2006. “Intergenerational mobility of wages in Brazil.” Brazilian Review of Econometrics 26: 181–211. Fortin, Nicole M., and Sophie Lefebvre. 1998. “Intergenerational Income Mobility in Canada.” In Labour Markets, Social Institutions, and the Future of Canada’s Children, edited by Miles Corak: Ottawa: Statistics Canada. Gong, Honge; Andrew Leigh; and Xin Meng. 2012. “Intergenerational Income Mobility in Urban China.” Review of Income and Wealth 58: 481–503. Grawe, Nathan D. 2006. “Lifecycle bias in estimates of intergenerational earnings persistence.” Labour Economics 13: 551–70. Gustafsson, Björn. 1994. “The degree and pattern of income immobility in Sweden.” Review of Income and Wealth 40: 67–85. Haider, Steven, and Gary Solon. 2006. “Life-Cycle Variation in the Association between Current and Lifetime Earnings.” American Economic Review 96: 1308–20. Haughton, Jonathan. 2001. “Introduction: Extraordinary changes.” In Living Standards During an Economic Boom: The Case of Vietnam, edited by Jonathan Haughton, Dominique Haughton, and Phong Nguyen. Hanoi: UNDP and Statistical Publishing House. Hertz, Thomas Nathaniel. 2001. Education, Inequality and Economic Mobility in South Africa (Unpublished doctoral dissertation). Boston: University of Massachusetts. 28 Hertz, Tom; Tamara Jayasundera; Patrizio Piraino; Sibel Selcuk; Nicole Smith; and Alina Verashchagina. 2007. “The inheritance of educational inequality: international comparisons and fifty-year trends.” The B.E. Journal of Economic Analysis and Policy. doi:10.2202/1935-1682.1775. Hirvonen, Lalaina H. 2008. “Intergenerational Earnings Mobility among Daughters and Sons: Evidence from Sweden and a Comparison with the United States.” American Journal of Economics and Sociology 67: 777–826. Hout, Michael. 1988. “More universalism, less structural mobility: The American occupational structure in the 1980s.” American Journal of Sociology 93, no. 6: 1358–1400. Imbert, Clement. 2013. “Decomposing the Labor Market Earnings Inequality: The Public and Private Sectors in Vietnam, 1993–2006.” World Bank Economic Revew 27: 55–79. Inoue, Atsushi, and Gary Solon. 2010. “Two-Sample Instrumental Variables Estimators.” Review of Economics and Statistics 92: 557–61. Irvin, Geogre. 1995. “Vietnam: Assessing the achievements of Doi Moi.” Journal of Development Studies 31: 725–50. Kan, Kamhon; I-Hsin Li; and Ruei-Hua Wang. 2015. “Intergenerational Income Mobility in Taiwan: Evidence from TS2SLS and Structural Quantile Regression.” The B.E. Journal of Economic Analysis and Policy 15: 257–84. Kim, Soobin. 2013. “Intergenerational Earnings Mobility in Korea.” Department of Economics Working Paper. Lansing: Michigan State University. Krueger, Alan B. 2012. “The Rise and Consequences of Inequality in the United States.” White House. https://www.whitehouse.gov/sites/default/files/krueger_cap_speech_final_re marks.pdf (accessed 3 March, 2015). Lee, Yong Suk. 2014. “Exams, districts, and intergenerational mobility: Evidence from South Korea.” Labour Economics 29: 62–71. Lefranc, Arnaud, and Alain Trannoy. 2005. “Intergenerational earnings mobility in France: is France more mobile than the US?.” Annales d’Economie et de Statistique 78: 57–77. 29 Lefranc, Arnaud; Fumiaki Ojima; and Takashi Yoshida. 2010. “The Intergenerational Transmission of Income and Education: A Comparison of Japan and France.” In Quality and Inequality of Education: Cross-National Perspectives, edited by Jaap Dronkers. The Netherlands: Springer. Lefranc, Arnaud; Fumiaki Ojima; and Takashi Yoshida. 2014. “Intergenerational earnings mobility in Japan among sons and daughters: Levels and trends.” Journal of Population Economics 27: 91–134. Liu, Amy Y. C. 2006. “Changing wage structure and education in Vietnam, 1992-98.” Economics of Transition 14: 681–706. Mazumder, Bhashkar, and Miguel Acosta. 2015. “Using Occupation to Measure Intergenerational Mobility.” ANNALS 657: 174–93. Mazumder, Bhashkar. 2005. “Fortunate Sons: New estimates of intergenerational mobility in the United States using social security earnings data.” Review of Economics and Statistics 87: 235–55. Mocetti, Sauro. 2007. “Intergenerational earnings mobility in Italy.” The B.E. Journal of Economic Analysis and Policy. doi:10.2202/1935-1682.1794. Ng, Irene YH; Xiaoyi Shen; and Kong Weng Ho. 2009. “Intergenerational earnings mobility in Singapore and the United States.” Journal of Asian Economics 20: 110–19. Ng, Irene YH. 2007. “Intergenerational income mobility in Singapore.” The BE Journal of Economic Analysis and Policy. doi:10.2202/1935-1682.1713. Ng, Irene YH. 2013. “The political economy of intergenerational income mobility in Singapore.” International Journal of Social Welfare 22: 207–18. Nguyen, Binh T.; James W. Albrecht; Susan B. Vroman; and M. Daniel Westbrook. 2007. “A quantile regression decomposition of urban–rural inequality in Vietnam.” Journal of Development Economics 83: 466–90. Nicoletti, Cheti, and John F Ermisch. 2008. “Intergenerational earnings mobility: changes across cohorts in Britain.” The B.E. Journal of Economic Analysis and Policy. doi:10.2202/1935-1682.1755. Nilsen, Øivind Anti; Kjell Vaage; Arild Aakvik; and Karl Jacobsen. 2012. “Intergenerational Earnings Mobility Revisited: Estimates Based on Lifetime Earnings.” Scandinavian Journal of Economics 114: 1–23. 30 Núñez, Javier I, and Leslie Miranda. 2010. “Intergenerational income mobility in a less-developed, high-inequality context: The case of Chile.” The B.E. Journal of Economic Analysis and Policy. doi:10.2202/1935-1682.2339. Organization for Economic Co-operation and Development. 2003. “Literacy Skills for the World of Tomorrow: Further Results from PISA 2000.” (accessed 3 January, 2015). Österberg, Torun. 2000. “Intergenerational Income Mobility in Sweden: What do Tax-Data Show?” Review of Income and Wealth 46: 421–36. Peters, H Elizabeth. 1992. “Patterns of Intergenerational Mobility in Income and Earnings.” Review of Economics and Statistics 74: 456–66. Piketty, Thomas. 2014. Capital in the Twenty-First Century. Cambridge: Harvard University Press. Piketty, Thomas. 2015. “About Capital in the Twenty-First Century.” American Economic Review: Papers and Proceedings 105: 1–6. Piraino, Patrizio. 2007. “Comparable estimates of intergenerational income mobility in Italy.” The B.E. Journal of Economic Analysis and Policy. doi:10.2202/1935-1682.1711. Piraino, Patrizio. 2015. “Intergenerational Earnings Mobility and Equality of Opportunity in South Africa.” World Development 67: 396–405. Sakellariou, Chris, and Zheng Fang. 2014. “The Vietnam reforms, change in wage inequality and the role of the minimum wage.” Economics of Transition 22: 313–40. Solon, Gary. 1992. “Intergenerational Income Mobility in the United States.” American Economic Review 82: 393–408. Solon, Gary. 1999. “Intergenerational Mobility in the Labor Market.” In Handbook of Labor Economics, edited by Orley Ashenfelter, and David Card. Amsterdam: North-Holland. Solon, Gary. 2002. “Cross-country differences in intergenerational earnings mobility.” Journal of Economic Perspectives 16: 59–66. Sun, Fengye, and Atsuko Ueda. 2015. “Intergenerational earnings mobility in Taiwan.” Economics Bulletin 35: 187–97. 31 Torche, Florencia. 2015. “Analyses of Intergenerational Mobility: An Interdisciplinary Review.” ANNALS 657: 37–62. Ueda, Atsuko. 2009. “Intergenerational Mobility of Earnings and Income in Japan.” The B.E. Journal of Economic Analysis and Policy 9: 1–25. Ueda, Atsuko. 2013. Intergenerational mobility of earnings in South Korea. Journal of Asian Economics 27: 33–41. van de Walle, Dominique, and Dileni Gunewardena. 2001. “Sources of ethnic inequality in Viet Nam.” Journal of Development Economics 65: 177–207. Waldkirch, Andreas; Serena Ng; and Donald Cox. 2004. “Intergenerational Linkages in Consumption Behavior.” Journal of Human Resources 39, no. 2: 355–81. World Bank. 2001. “Vietnam Living Standards Survey (VLSS), 1997–98: Basic Information.” World Bank. 1181743055198/3877319-1181827303756/vn98bifBasicInformation.pdf (accessed 12 December, 2014). World Bank. 2014. Taking Stock: An Update on Vietnam’s Recent Economic Developments. Hanoi: The World Bank. Zimmerman, David J. 1992. “Regression towards mediocrity in Economic Stature.” American Economic Review 82: 409–29. 32 APPENDICES Fig. A1. The distribution of sons’ age in the primary sample Fig. A2. The distribution of daughters’ age in the primary sample 33 TABLE A1 Transition matrix – Probability of sons’ individual income quartile given fathers’ individual earnings quartile Fathers’ individual earnings quartile (%) Sons’ individual income quartile (%) Bottom Second Third Top Bottom 38.20 26.69 19.38 15.73 Second 28.44 29.05 22.32 20.18 Third 22.46 25.75 27.84 23.95 Top 14.37 21.41 23.24 40.98 Notes: 1. Father’s individual earnings is predicted based on the set of socio- economic characteristics including education, occupation, industry, and geographical region. TABLE A2 Transition matrix – Probability of daughter’s income quartile given father’s individual earnings quartile Father’s individual earnings quartile (%)* Daughter’s individual income quartile (%) Bottom Second Third Top Bottom 38.92 23.95 22.16 14.97 Second 22.00 26.00 31.33 20.67 Third 21.02 24.84 26.11 28.03 Top 17.72 24.68 21.52 36.08 Notes: 1. Father’s individual earnings is predicted based on the set of socio- economic characteristics including education, occupation, industry, and geographical region. 34 TABLES TABLE 1 Descriptive statistics of samples Variables Secondary sample (VLSS 1997–1998) Primary sample of son-father pairs (VHLSS 2012) Primary sample of daughter-father pairs (VHLSS 2012) Potential fathers Fathers Sons Fathers Daughters Mean SD Mean SD Mean SD Mean SD Mean SD Age (years) 39.97 5.90 57.59 7.29 29.06 4.04 57.60 6.81 28.46 3.52 Education (1) non-diploma or primary (= 1 if yes, = 0 if no) 0.13 0.34 0.40 0.50 0.20 0.42 0.34 0.49 0.17 0.37 (2) secondary (= 1 if yes, = 0 if no) 0.34 0.47 0.32 0.47 0.20 0.41 0.29 0.46 0.16 0.37 (3) vocational (= 1 if yes, = 0 if no) 0.14 0.34 0.06 0.22 0.07 0.25 0.08 0.27 0.04 0.20 (4) high school (= 1 if yes, = 0 if no) 0.26 0.44 0.15 0.37 0.33 0.48 0.20 0.40 0.33 0.48 (5) tertiary (= 1 if yes, = 0 if no) 0.13 0.34 0.07 0.26 0.20 0.41 0.09 0.29 0.30 0.47 Occupation (1) very highly skilled (= 1 if yes, = 0 if no) 0.14 0.34 0.07 0.26 0.16 0.36 0.09 0.28 0.21 0.42 (2) lower highly skilled (= 1 if yes, = 0 if no) 0.09 0.29 0.04 0.17 0.09 0.28 0.05 0.18 0.18 0.39 (3) typical non-manual (= 1 if yes, = 0 if no) 0.21 0.40 0.14 0.34 0.12 0.32 0.17 0.37 0.18 0.38 (4) lower-grade (= 1 if yes, = 0 if 0.10 0.30 0.04 0.20 0.15 0.36 0.05 0.22 0.14 0.35 35 no) (5) skilled manual (= 1 if yes, = 0 if no) 0.21 0.41 0.16 0.36 0.01 0.09 0.16 0.36 0.01 0.07 (6) semi- and un-skilled manual (= 1 if yes, = 0 if no) 0.17 0.38 0.11 0.32 0.27 0.45 0.10 0.29 0.16 0.36 (7) farmers and farm workers (= 1 if yes, = 0 if no) 0.08 0.29 0.44 0.50 0.20 0.41 0.40 0.49 0.12 0.32 Industry (1) agriculture (= 1 if yes, = 0 if no) 0.12 0.32 0.53 0.50 0.10 0.30 0.51 0.50 0.09 0.20 (2) manufacturing (= 1 if yes, = 0 if no) 0.17 0.37 0.10 0.30 0.20 0.40 0.09 0.29 0.38 0.49 (3) public management (= 1 if yes, = 0 if no) 0.16 0.37 0.07 0.25 0.09 0.29 0.09 0.29 0.08 0.27 (4) health and education (= 1 if yes, = 0 if no) 0.20 0.40 0.03 0.16 0.07 0.25 0.03 0.18 0.23 0.42 (5) trade and finance (= 1 if yes, = 0 if no) 0.10 0.30 0.07 0.26 0.10 0.30 0.09 0.28 0.10 0.31 (6) utilities (= 1 if yes, = 0 if no) 0.01 0.11 0.02 0.05 0.03 0.10 0.01 0.04 0.01 0.09 (7) transportation and communication (= 1 if yes, = 0 if no) 0.06 0.23 0.05 0.21 0.09 0.29 0.05 0.22 0.03 0.16 (8) construction (= 1 if yes, = 0 if no) 0.11 0.31 0.08 0.28 0.23 0.42 0.07 0.26 0.03 0.18 (9) mining (= 1 if yes, = 0 if no) 0.01 0.11 0.02 0.11 0.04 0.15 0.03 0.10 0.01 0.10 (10) community, and social services (= 1 if yes, = 0 if no) 0.06 0.23 0.03 0.17 0.05 0.18 0.03 0.17 0.04 0.20 Geographical Region 36 (1) Red River Delta (RRD) (= 1 if yes, = 0 if no) 0.27 0.44 0.24 0.43 0.24 0.43 0.22 0.41 0.22 0.41 (2) Northern Midland and Mountain Areas (NMMA) (= 1 if yes, = 0 if no) 0.07 0.25 0.14 0.35 0.14 0.35 0.10 0.31 0.10 0.31 (3) North Central and Central Coastal Areas (NCCCA) (= 1 if yes, = 0 if no) 0.26 0.44 0.25 0.43 0.25 0.43 0.24 0.43 0.24 0.43 (4) Central Highlands (CH) (= 1 if yes, = 0 if no) 0.02 0.13 0.03 0.16 0.03 0.16 0.02 0.15 0.02 0.15 (5) South East (SE) (= 1 if yes, = 0 if no) 0.21 0.42 0.11 0.32 0.11 0.32 0.15 0.36 0.15 0.36 (6) Mekong River Delta (MRD) (= 1 if yes, = 0 if no) 0.17 0.37 0.23 0.42 0.23 0.42 0.27 0.44 0.27 0.44 Log of monthly individual earnings (VND 1000) 5.64 0.89 5.04 0.42 7.84 0.60 5.07 0.43 7.71 0.63 Log of monthly individual income (VND 1000) 7.93 0.63 7.82 0.66 Observations 1041 1344 632 Notes: 1. Potential fathers’ age are 31–54 in the secondary sample. 2. Sons’ age are 25–54 in the primary father-son sample. 3. Daughters’ age are 25–47 in the primary father-daughter sample. 37 TABLE 2 Preferred first-stage regressions. Dependent variable: Individual earnings (monthly, VND 1,000, in log) Preferred variable Coefficient Education (2) secondary 0.27** (0.12) (3) vocational 0.30** (0.13) (4) high school 0.45*** (0.11) (5) tertiary 0.57*** (0.12) Occupation (1) very highly skilled 0.25 (0.19) (2) lower highly skilled 0.38** (0.18) (3) typical non-manual 0.22 (0.19) (4) lower-grade 0.29 (0.21) (5) skilled manual 0.12 (0.21) (6) semi- and un-skilled manual 0.06 (0.18) Industry (1) agriculture – 0.07 (0.27) (2) manufacturing 0.11 (0.23) (3) public management – 0.18 (0.25) (4) health and education 0.14 (0.26) (5) trade, and finance 0.08 (0.26) (6) utilities 0.20 (0.31) (7) transportation and communication 0.19 (0.27) (8) construction – 0.29 (0.27) 38 (10) community and social services – 0.27 (0.27) Geographical Region (1) Red River Delta (RRD) 0.50** (0.21) (2) Northern Midland and Mountain Areas (NMMA) 0.48** (0.22) (3) North Central and Central Coastal Areas (NCCCA) 0.31 (0.21) (5) South East (SE) 0.29 (0.24) (6) Mekong River Delta (MRD) – 0.04 (0.23) R2 0.19 Observations 104 Notes: 1. ***, **, and * represent statistical significance at the 1%, 5%, and 10% level, respectively. 2. Omitted variables: (1) non-diploma or primary in the education group; (7) farmers, and farm workers in the occupation group; (9) mining in the industry group; and (4) Central Highlands (CH) in the geographical region group. 39 TABLE 3 Baseline IGE estimates for sons (full sample) Dependent variable (monthly, VND 1000, in log): Sons’ individual earnings (1) individual income (2) "# 0.36*** (0.04) 0.39*** (0.04) R2 0.08 0.08 Observations 1344 1344 Notes: 1. ***, **, and * represent statistical significance at the 1%, 5%, and 10% level, respectively. 2. Bootstrapping standard errors (with 1000 replications) are in parentheses. 3. Father’s individual earnings is predicted using education, occupation, industry, and geographical region. TABLE 4 Transition matrix – Probability of sons’ individual earnings quartile given fathers’ individual earnings quartile Fathers’ individual earnings quartile (%) Sons’ individual earnings quartile (%) Bottom Second Third Top Bottom 37.08 26.12 20.51 16.29 Second 26.61 26.91 26.61 19.88 Third 21.86 26.05 28.14 23.95 Top 13.76 20.49 25.99 39.76 Notes: 1. Father’s individual earnings is predicted using education, occupation, industry, and geographical region. 40 TABLE 5 Baseline IGE estimates for daughters (full sample) Dependent variable (monthly, VND 1000, in log): Daughters’ individual earnings (1) individual income (2) "# 0.28*** (0.06) 0.33*** (0.06) R2 0.06 0.07 Observations 632 632 Notes: 1. ***, **, and * represent statistical significance at the 1%, 5%, and 10% level, respectively. 2. Bootstrapping standard errors (with 1000 replications) are in parentheses. 3. Father’s individual earnings is predicted using education, occupation, industry, and geographical region. TABLE 6 Transition matrix – Probability of daughter’s individual earnings quartile given father’s individual earnings quartile Father’s individual earnings quartile (%) Daughter’s individual earnings quartile (%) Bottom Second Third Top Bottom 37.13 27.54 19.76 15.57 Second 26.00 26.00 28.00 20.00 Third 20.38 30.57 23.57 25.48 Top 20.25 27.85 20.89 31.01 Notes: 1. Father’s individual earnings is predicted using education, occupation, industry, and geographical region. 41 TABLE 7 Robustness check for sons to different first-stage model specifications The set of fathers’ earnings predictors in the first stage Dependent variable (monthly, VND 1000, in log): Sons’ individual earnings (1) individual income (2) "# R2 "# R2 (1) education 0.37*** (0.05) 0.06 0.40*** (0.05) 0.07 (2) occupation 0.30*** (0.06) 0.03 0.36*** (0.06) 0.04 (3) industry 0.27*** (0.07) 0.02 0.34*** (0.08) 0.03 (4) geographical region 0.32*** (0.07) 0.03 0.32*** (0.07) 0.03 (5) education and occupation 0.38*** (0.04) 0.07 0.42*** (0.05) 0.07 (6) education and industry 0.35*** (0.04) 0.06 0.39*** (0.05) 0.07 (7) education and geographical region 0.35*** (0.04) 0.07 0.36*** (0.04) 0.07 (8) occupation and industry 0.26*** (0.06) 0.03 0.32*** (0.06) 0.04 (9) occupation and geographical region 0.40*** (0.05) 0.06 0.43*** (0.05) 0.07 (10) industry and geographical region 0.33*** (0.05) 0.05 0.36*** (0.05) 0.05 (11) education, occupation and industry 0.35*** (0.04) 0.06 0.39*** (0.05) 0.07 (12) education, occupation and geographical region 0.39*** (0.04) 0.08 0.41*** (0.04) 0.08 (13) education, industry and geographical region 0.34*** (0.04) 0.07 0.37*** (0.04) 0.08 (14) occupation, industry and geographical region 0.37*** (0.05) 0.06 0.41*** (0.05) 0.06 (15) education, occupation, industry and geographical region 0.36*** (0.04) 0.08 0.39*** (0.04) 0.08 Notes: 1. ***, **, and * represent statistical significance at the 1%, 5%, and 10% level, respectively. 2. Bootstrapping standard errors (with 1000 replications) are in parentheses. 3. Sample size is 1344 observations. 42 TABLE 8 Robustness check for daughters to different first-stage specifications The set of fathers’ earnings predictors in the first stage Dependent variable (monthly, VND 1000, in log): Daughters’ individual earnings (1) individual income (2) "# R2 "# R2 (1) education 0.24*** (0.06) 0.04 0.27*** (0.07) 0.05 (2) occupation 0.38*** (0.08) 0.05 0.43*** (0.08) 0.06 (3) industry 0.32*** (0.10) 0.04 0.39*** (0.10) 0.04 (4) geographical region 0.31*** (0.10) 0.04 0.37*** (0.11) 0.04 (5) education and occupation 0.30*** (0.07) 0.06 0.35*** (0.07) 0.06 (6) education and industry 0.25*** (0.07) 0.05 0.29*** (0.07) 0.05 (7) education and geographical region 0.27*** (0.06) 0.06 0.31*** (0.06) 0.06 (8) occupation and industry 0.29*** (0.08) 0.04 0.34*** (0.08) 0.05 (9) occupation and geographical region 0.41*** (0.07) 0.08 0.48*** (0.07) 0.09 (10) industry and geographical region 0.31*** (0.07) 0.05 0.37*** (0.08) 0.06 (11) education, occupation and industry 0.26*** (0.07) 0.05 0.31*** (0.07) 0.05 (12) education, occupation and geographical region 0.33*** (0.06) 0.07 0.38*** (0.06) 0.08 (13) education, industry and geographical region 0.26*** (0.06) 0.06 0.31*** (0.059) 0.06 (14) occupation, industry and geographical region 0.33*** (0.07) 0.06 0.39*** (0.07) 0.07 (15) education, occupation, industry and geographical region 0.28*** (0.06) 0.06 0.33*** (0.06) 0.07 Notes: 1. ***, **, and * represent statistical significance at the 1%, 5%, and 10% level, respectively. 2. Bootstrapping standard errors (with 1000 replications) are in parentheses. 3. Sample size is 632 observations. 43 TABLE 9 IGE estimates by the different age ranges for sons Dependent variable (monthly, VND 1000, in log): Sons’ individual earnings (1) individual income (2) Panel A. Sons aged 25–29 "# 0.34*** (0.05) 0.36*** (0.05) R2 0.07 0.07 Observations 892 892 Panel B. Sons aged 30–34 "# 0.39*** (0.07) 0.46*** (0.07) R2 0.10 0.13 Observations 317 317 Panel C. Sons aged 35–54 "# 0.48*** (0.15) 0.49*** (0.17) R2 0.10 0.10 Observations 135 135 Panel D. Sons aged 30–50 "# 0.41*** (0.07) 0.47*** (0.07) R2 0.09 0.11 Observations 450 450 Notes: 1. ***, **, and * represent statistical significance at the 1%, 5%, and 10% level, respectively. 2. Bootstrapping standard errors (with 1000 replications) are in parentheses. 3. Father’s individual earnings is predicted using education, occupation, industry, and geographical region. 44 TABLE 10 IGE estimates by different age ranges for daughters Dependent variable (monthly, VND 1000, in log): Daughters’ individual earnings (1) individual income (2) Panel A. Daughters aged 25–29 "# 0.24*** (0.07) 0.29*** (0.07) R2 0.04 0.05 Observations 450 450 Panel B. Daughters aged 30–34 "# 0.44*** (0.14) 0.48*** (0.14) R2 0.10 0.10 Observations 149 149 Panel C. Daughters aged 30–47 "# 0.40*** (0.11) 0.45*** (0.12) R2 0.10 0.10 Observations 182 182 Notes: 1. ***, **, and * represent statistical significance at the 1%, 5%, and 10% level, respectively. 2. Bootstrapping standard errors (with 1000 replications) are in parentheses. 3. Fathers’ individual earnings is predicted using education, occupation, industry, and geographical region.