Abstract
Since the end of 1990s, approximately 160 million Chinese rural workers migrated to cities for work. Because of restrictions on migrant access to local health and education systems, many rural children are left behind in home villages to grow up without parental care. This article examines how exposure to cumulative parental migration affects children’s health and education outcomes. Using the Rural-Urban Migration Survey in China (RUMiC) data, we measure the share of children’s lifetime during which parents were away from home. We instrument this measure of parental absence with weather changes in their home villages when parents were aged 16–25, when they were most likely to initiate migration. Results show a sizable adverse effect of exposure to parental migration on the health and education outcomes of children: in particular, boys. We also find that the use of the contemporaneous measure for parental migration in previous studies is likely to underestimate the effect of exposure to parental migration on children’s outcomes.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The idea that human capital acquisition is a cumulative process goes back to Ben-Porath (1967), who provided the theory of individual human capital investment where one chooses the level of time and monetary investments over one’s life cycle. Leibowitz (1974) applied this idea to the investments in children, which include home investments and school inputs at various stages of child development.
In the study, we also made attempt to distinguish between one or both parents migrating, but due to lack of strong instruments, our result is not conclusive in this aspect.
Rural-urban migration in China occurs under a “guest worker” system, whereby migrants are unable to settle in cities (see Meng 2012).
The situation may have improved after significant investment in rural schools after 2009. The authors recently visited some rural schools in one county and found that most schools had new buildings and that the living conditions at these schools had improved significantly.
The first wave is not used because we as discuss later, two of the key outcome variables were not available in the first wave data, and one of the key questions on the duration of parental migration was asked only in the second wave.
The RUMiC Rural Household Survey uses National Bureau of Statistics annual household survey sample, which is designed to be representative sample nationwide as well as within each province. The nine provinces covered in the RUMiC survey include Hebei, Jiangsu, Zhejiang, Anhui, Henan, Hubei, Guangdong, Chongqing, and Sichuan. Of the nine provinces, five are considered to be the predominant migrant-sending provinces (Anhui, Henan, Hubei, Sichuan, and Chongqing), and the number of migrants from these provinces exceeds 50 % of total rural-to-urban outmigration.
The test was not conducted in the remaining four rural sample provinces because of the cost and complexity of conducting the test in rural areas.
The test instruments were designed specifically for the RUMiC project by the Research Institute for Education Statistics and Measurement (RIESM) at Beijing Normal University. The RIESM organized teachers from the nine sample provinces (including the sample province without rural areas) in order to design and test the instruments. They created two types of instruments: one for Years 1–3, and the other for Years 4–6. These tests are designed to take about 30 minutes. University students were sent to the sample households to conduct the test during the months of July through August.
As discussed earlier, in the RUMiC Rural Household Surveys, all individuals who were registered in the household were asked to record all the information. For those who were absent at the time of the survey, the household main respondent reported on their behalf, except for subjective questions.
For example, one-third of those who have ever migrated and reported the initial year of migration in the three waves of the RUMiC data spent the entire previous year in rural homes. Thus, parents with migration experience might have had years without migration between the initial year and 2006.
Rural individuals who were working in cities in 2008 had spent an average of 2.4 years there since the start of their current migration spell. On the other hand, they had spent an average of 7.1 years there since the start of their first migration spell, which occurred before the current migration spell for those with the experience of multiple migration spells. This difference indicates that, on average, these parents had more than one episode of migration, with their first migration spell starting in 2000 and the current spell starting in 2005.
The RUMiC defines children as individuals aged 0–15 years. After they turn age 16, they are classified as adults under the RUMiC framework, and their test scores and other school-related information are no longer collected.
For the small number of cases (10, 14, 78, and 89 for height, weight, the Chinese test score, and the math test score, respectively) with obvious reporting errors, we replaced the original data with the likely values. For instance, we use the adjusted score of 70 when the original reported score was 700 but the full score was 100; we use the average between 120 and 125 when the original height was reported to be 120cm in 2008, 600cm in 2009, and 125cm in 2010. Excluding these cases with obvious reporting errors does not change our regression results.
Thus, if a child has two data points on health, for example, his/her health measure is constructed as the mean value of the two data points. However, if a child has only one data point, be it in 2009 or 2010, his/her health measure is the nonmissing data point.
Similarly, exposure up to 2010 requires information from all three waves.
The 2008 wave contains 4,548 children, 142 (3.1 %) of whom attrited by 2009.
The survey team within NBS changed between 2009 and 2010, resulting in 11 villages from the 2009 wave not being included in the 2010 wave; instead, 8 new villages were included in the 2010 wave. In total, 19 villages were not matched between the two waves.
The rate of divorce has been rising in China recently, but the share of children with divorced parents in the RUMIC survey is very low, at approximately 1 % in all three survey waves.
Approximately 1 % and 5 % of fathers and mothers, respectively, have missing values in the years of schooling variable. We coded them as 0 years of schooling and used a dummy variable to identify this group.
The error term, ν ijt , is assumed to be independent across households in the basic OLS estimation. In the IV estimation, we assume that it is independent across groups defined by county and cohort for each parent. As discussed later, our IVs vary across counties and parental cohorts.
We focus on spring months because March through May (or in Chinese calendar terms, Chun Fen to Xiao Man) is the period believed to be crucial to the year’s harvest.
To allocate the nearest weather station for our sample counties, we marked all the counties and the stations in Google Earth based on their latitude and longitude and then measured their straight line distance.
Unfortunately, we do not have good measures for the inputs into children’s health production function.
References
Antman, F. M. (2011). The intergenerational effects of paternal migration on schooling and work: What can we learn from children’s time allocations? Journal of Development Economics, 96, 200–208.
Antman, F. M. (2012). Gender, educational attainment, and the impact of parental migration on children left behind. Journal of Population Economics, 25, 1187–1214.
Arguillas, M. J. B., & Williams, L. (2010). The impact of parents’ overseas employment on educational outcomes of Filipino children. International Migration Review, 44, 300–319.
Astone, N. M., & McLanahan, S. S. (1991). Family structure, parental practices and high school completion. American Sociological Review, 56, 309–320.
Ben-Porath, Y. (1967). The production of human capital and the life cycle of earnings. Journal of Political Economy, 75, 352–365.
Bertrand, M., & Pan, J. (2013). The trouble with boys: Social influences and the gender gap in disruptive behaviour. American Economic Journal: Applied Economics, 5(1), 32–64.
Biblarz, T. J., & Raftery, A. E. (1999). Family structure, educational attainment, and socioeconomic success: Rethinking the “Pathology of Matriarchy.” American Journal of Sociology, 105, 321–365.
Centers for Disease Control and Prevention (CDC). (n.d.). Modified z-scores in the CDC growth charts. Retrieved from http://www.cdc.gov/nccdphp/dnpa/growthcharts/resources/BIV-cutoffs.pdf
Chang, H., Dong, X.-y., & MacPhail, F. (2011). Labor migration and time use patterns of the left-behind children and elderly in rural China. World Development, 39, 2199–2210.
Chen, J. J. (2013). Identifying non-cooperative behavior among spouses: Child outcomes in migrant-sending households. Journal of Development Economics, 100, 1–18.
Chen, X., Huang, Q., Rozelle, S., Shi, Y., & Zhang, L. (2009). Effect of migration on children’s educational performance in rural China. Comparative Economic Studies, 51, 323–343.
Chen, X., Huang, Q., Rozelle, S., & Zhang, L. (2007, July). Migration, money and mother: The effect of migration on children’s educational performance in rural China. Paper presented at the annual meeting of the American Agricultural Economics Association, Portland, OR. Retrieved from http://ageconsearch.umn.edu/bitstream/9791/1/sp07ch01.pdf
Chen, Y., & Feng, S. Z. (2013). Access to public schools and the education of migrant children in China. China Economic Review, 26, 75–88.
Chetty, R., Friedman, J. N., Hilger, N., Saez, E., Schanzenbach, D. W., & Yagan, D. (2011). How does your kindergarten classroom affect your earnings? Evidence from Project STAR. Quarterly Journal of Economics, 126, 1593–1660.
de Brauw, A., & Mu, R. (2011). Migration and the overweight and underweight status of children in rural China. Food Policy, 36, 88–100.
de Onis, M., & Blossner, M. (1997). WHO Global Database on Child Growth and Malnutrition, World Health Organization. Retrieved from http://www.who.int/nutgrowthdb/publications/methodology/en/
de Onis, M., Onyango, A. W., Borghi, E., Siyam, A., Nishida, C., & Siekmann, J. (2007). Development of a WHO growth reference for school-aged children and adolescents. Bulletin of the World Health Organization, 85, 660–667.
Edwards, A., & Ureta, M. (2003). International migration, remittances and schooling: Evidence from El Salvador. Journal of Development Economics, 72, 429–461.
Ginther, D. K., & Pollak, R. A. (2003). Does family structure affect children’s educational outcomes? (National Bureau of Economic Research Working Paper No. 9628). Cambridge, MA: National Bureau of Economic Research. Retrieved from http://www.nber.org/papers/w9628.pdf
Grossman, M. (2000). The human capital model. In A. J. Culyer & J. P. Newhouse (Eds.), Handbook of health economics (Vol. 1A, pp. 347–408). Amsterdam, The Netherlands: Elsevier.
Hanson, G., & Woodruff, C. (2003). Emigration and educational attainment in Mexico (Working paper). San Diego: University of California.
Hanushek, E., & Woessmann, L. (2012). Do better schools lead to more growth? Cognitive skills, economic outcomes, and causation. Journal of Economic Growth, 17, 267–321.
Hanushek, E., & Zhang, L. (2009). Quality-consistent estimates of international schooling and skill gradients. Journal of Human Capital, 3, 107–143.
Hanushek, E. A. (2003). The failure of input-based schooling policies. Economic Journal, 113(485), F64–F98.
Jampaklay, A. (2006). Parental absence and children’s school enrolment: Evidence from a longitudinal study in Kanchanaburi, Thailand. Asian Population Studies, 2, 93–110.
Krein, S. F., & Beller, A. H. (1988). Educational attainment of children from single-parent families: Differences by exposure, gender, and race. Demography, 25, 221–234.
Kuczmarski, R. J., Ogden, C. L., Guo, S. S., Grummer-Strawn, L. M., Flegal, K. M., Mei, Z., . . . Johnson, C. L. (2002). 2000 CDC growth charts for the United States: Methods and development (Series Report 11, No. 246). Atlanta, GA: Centers for Disease Control and Prevention, National Center for Health Statistics. Retrieved from https://www.cdc.gov/nchs/data/series/sr_11/sr11_246.pdf
Lamb, M. (1998). Non-parental childcare: Context, quality, correlates and consequences. In W. Damon, I. E. Sigel, & K. A. Renninger (Eds.), Handbook of child psychology, Vol. 4: Child psychology in practice (5th ed., pp. 73–133). New York, NY: Wiley.
Lee, L., & Park, A. (2010). Parental migration and child development in China (Gansu Survey of Children and Families Working Paper). Philadelphia: University of Pennsylvania, Penn Libraries, Population Studies Center. Retrieved from http://repository.upenn.edu/gansu_papers/24/
Leibowitz, A. (1974). Home investments in children. Journal of Political Economy, 82(Part 2), S111–S131.
Love, J. M., Schochet, P. Z., & Meckstroh, A. L. (1996). Are they in danger? What research does—and doesn’t—tell us about childcare quality and children well-being (Child Care Research and Policy Papers). Princeton, NJ: Mathematica Policy Research. Retrieved from https://www.mathematica-mpr.com/-/media/publications/pdfs/realdanger.pdf
Lu, Y. (2012). Education of children left behind in rural China. Journal of Marriage and Family, 74, 328–341.
McKenzie, D., & Rapoport, H. (2011). Can migration reduce educational attainment? Evidence from Mexico. Journal of Population Economics, 24, 1331–1358.
McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent: What helps, what hurts. Cambridge, MA: Harvard University Press.
Meng, X. (2012). Labor market outcomes and reforms in China. Journal of Economic Perspectives, 26(4), 75–102.
Meng, X., & Manning, C. (2010). The great migration in China and Indonesia—Trend and institutions. In X. Meng, C. Manning, S. Li, & T. N. Effendi (Eds.), The Great Migration: Rural-urban migration in China and Indonesia (pp. 1–19). Camberley, UK: Edward Elgar.
Meng, X., & Zhang, J. (2001). The two-tier labor market in urban China: Occupational segregation and wage differentials between urban residents and rural migrants in Shanghai. Journal of Comparative Economics, 29, 485–504.
Meyerhoefer, C. D., & Chen, C. J. (2011). The effect of parental labor migration on children’s educational progress in rural China. Review of Economics of the Household, 9, 379–396.
Mo, D., Yi, H. M., Zhang, L. X., Shi, Y. J., Rozelle, S., & Medina, A. (2012). Transfer paths and academic performance: The primary school merger program in China. International Journal of Educational Development, 32, 423–431.
Mulligan, C. B. (1999). Galton versus the human capital approach to inheritance. Journal of Political Economy, 107(Suppl. 6), S184–S224.
National Bureau of Statistics of China. (2012). China statistical yearbook. Beijing: China Statistics Press.
National Bureau of Statistics of China. (2016). China Household Survey yearbook, 2016. Beijing: China Statistics Press.
Ruan, M., & Sun, X. Y. (2011). What means to love you, my children: Survey of adulterants’ psychological crisis. Changsha, China: Hunan Literature and Art Publishing House. (in Chinese)
Rural Education Action Program (REAP). (2009). Educational challenges—Boarding schools. Stanford, CA: REAP. Retrieved from http://reap.fsi.stanford.edu/docs/boarding_schools
Shi, Y., & Zhang, L. (2010). Boarding school management (and PTAs) in China’s poor rural areas. Retrieved from https://reap.fsi.stanford.edu/sites/default/files/res/LXZhang%26YJShi_Boarding_school_FF_field_trip_Apr2010.pdf
Shu, B., & Tong, Y. (2015, April). Boarding school and students’ well-being: The case of rural China. Paper presented at the annual meeting of the Population Association of America, San Diego, CA.
Sigle-Rushton, W., & McLanahan, S. (2006). Father absence and child well-being: A critical review. In D. P. Moynihan, T. Smeeding, & L. Rainwater (Eds.), The future of the family (pp. 116–155). New York, NY: Russell Sage Foundation.
Todd, P. E., & Wolpin, K. I. (2003). On the specification and estimation of the production function for cognitive achievement. Economic Journal, 113(485), F3–F33.
Wen, M., & Lin, D. (2012). Child development in rural China: Children left behind by their migrant parents and children of nonmigrant families. Child Development, 83, 120–136.
Whitebook, M., Howes, C., & Phillips, D. (1989). Who cares? Child care teachers and the quality of care in America: Final report, National Child Care Staffing Study. Berkeley, CA: Child Care Employee Project.
World Health Organization (WHO). (2006). WHO child growth standards: Length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: Methods and development. Geneva, Switzerland: WHO.
Xu, H., & Xie, Y. (2015). The causal effects of rural-to-urban migration on children’s well-being in China. European Sociological Review, 31, 502–519.
Yang, D. (2008). International migration, remittances and household investment: Evidence from Philippine migrants’ exchange rate shocks. Economic Journal, 118, 591–630.
Zhang, H., Behrman, J. R., Fan, C. S., Wei, X., & Zhanga, J. (2014). Does parental absence reduce cognitive achievements? Evidence from rural China. Journal of Development Economics, 111, 181–195.
Zhou, C., Sylvia, S., Zhang, L., Luo, R., Yi, H., Liu, C., . . . Rozelle, S. (2015). China’s left-behind children: Impact of parental migration on health, nutrition and educational outcomes. Health Affairs, 34, 1964–1971.
Zhou, M., Murphy, R., & Tao, R. (2014). Effects of parents’ migration on the education of children left behind in rural China. Population and Development Review, 40, 273–292.
Acknowledgments
This research has benefited from the Australian Research Council Grant (LP066972 and LP140100514) and the Grant-in-Aid for Scientific Research (No. 24730239) of the Japan Society for the Promotion of Science. The authors acknowledge financial support from Australian Research Council (ARC) Linkage Grants LP0669728 and LP140100514 for funding the RUMiC survey, as well as JSPS KAKENHI Grant No. JP24730239 for research support.
Author information
Authors and Affiliations
Corresponding author
Data Appendix
Data Appendix
The share of a child’s lifetime during which parents were away is calculated using two types of information from the RUMiC study. One is the number of months in the previous year during which parents were away from home. This information is available from all the waves. Thus, we know the number of months parents were away in 2007, 2008, and 2009. The other piece of information is when parents started migration, which was asked only in the 2009 wave for individuals who reported having been away in 2008. To assess the number of months parents were away since the child’s birth, we combine the information from the two sources. First, if a child was born after 2007, we aggregated the number of months in which parents were away between 2007 and 2009. Second, if a child was born before 2007, we compare the year of birth and the year in which the 2008 parental migration started. If the child was born after the starting year, we assume that parents have been away for work since the birth of the child until the end of 2006. If the child was born before the starting year, we assume that parents have been away since the starting year until the end of 2006. For individuals who answered that their migration started before 2007, we added the number of months between the beginning of the migration until the end of 2006. In all cases, the duration of migration since the birth of children is expressed in months and is divided by the number of months since the child was born. In a robustness test, we use an alternative indicator for the year in which parental migration started: the year in which parents migrated for the first time.
The measures for migration of one parent and both parents are created in a similar manner. We compare the timing of the three events: the birth of a child, start of paternal migration, and start of maternal migration. For example, if the father was away for work before the child’s birth, and the mother joined the father for work in cities some years after the child’s birth, the child is assumed to have been exposed to migration of one parent since birth, and started to be exposed to migration of both parents since the mother started migration. We count the number of months that fall in each period and divide by the total number of months in the child’s lifetime. Because the months of absence are unknown between 2007 and 2009, parents are assumed to have been away if they were away for six months or more. Thus, if one parent was away for six months or more and the other was away for five months or less, we assume that the child was exposed to the migration of one parent in that year.
Individual test scores were asked for the previous semester, together with the full score for each subject. The ratio of the individual score over the full score is used as our outcome variables.
Distance to public facilities is coded using five categories: (1) <2km, (2) 2–5km, (3) 5–10km, (4) 10–20km, and (5) >20km.
The height-for-age z score is created using parameters from the Centers for Disease Control (CDC) 2000 Growth Charts and 2006 World Health Organization (WHO) Growth Charts (de Onis et al. 2007; Kuczmarski et al. 2002; WHO 2006). The results do not change substantively depending on the choice over the two parameter sources. In this article, we report the estimates based on the CDC parameters because they provide a more suitable reference group for our analysis. In the WHO growth charts, the comparison group for children aged 0–5 comprises children following optimal health practices. Thus, the charts depict the standard that is likely to realize under optimal conditions, rather than just a reference. However, the same standard was not used for the comparison group for 5- to 19-year-olds, and it is based on the U.S. reference children used in the 1977 National Center for Health Statistics (NCHS) Growth Charts. On the other hand, the CDC 2000 charts provide a more consistent reference for children in our sample, based on a group of children in the United States. We apply the CDC parameters for infants (based on length) to observations aged 0–24 months, and the parameters for children (based on statue) observations aged 25–180 months. The transition between these two charts have been made smooth in the 2000 CDC charts (Kuczmarski et al. 2002).
Because our anthropometric data are based on reports by parents, there are likely to be measurement errors. Several methods have been suggested to deal with them. The WHO recommends the use of different formula for those children whose z scores are larger than 3 in the absolute terms. The CDC training modules contain a note that recommends distinguishing “biologically implausible values” (CDC n.d.), which are observations whose z scores are larger than certain values, or are away from the mean z score in terms of standard deviations. We report the results that are commonly found regardless of the choice over these alternative adjustments to the measures for outcomes.
Rights and permissions
About this article
Cite this article
Meng, X., Yamauchi, C. Children of Migrants: The Cumulative Impact of Parental Migration on Children’s Education and Health Outcomes in China. Demography 54, 1677–1714 (2017). https://doi.org/10.1007/s13524-017-0613-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13524-017-0613-z