Are migrants a threat? Migrant children and human capital investments among local households in urban China

The study addresses a significant research gap concerning the social spillovers of migrant children on the decision-making processes of native families regarding human capital investments. Existing research primarily focuses on the impact of migrant peers on native children's academic achievement, yielding mixed results. This study argues that the impact of migration on receiving areas may be underestimated if the behavioral responses of native families are not considered. China provides a compelling context due to its large internal migration, the Hukou system (household registration system) that restricts access to resources for migrant children, and the intense competition in the education market driven by parents' aspirations for their children's success. The study's primary research questions are: 1) Do migrant children have spillover effects on human capital investments in local urban households? and 2) If so, what are the underlying mechanisms driving these effects? The researchers aim to provide a comprehensive understanding of the behavioral consequences of migration on receiving areas by focusing on the causal effects of migrant children on the behavioral responses of local households, highlighting the role of shadow education investments.

The literature review examines existing studies on the educational spillovers of migrant children on native students, revealing mixed findings. Studies in developed countries show varied results, ranging from negative associations between migrant presence and native students' schooling access to no significant impact or even positive effects on educational attainment. There is a relative scarcity of research on the impact of internal migration, despite its larger scale. Studies in China utilizing random class assignment show ambiguous results regarding the impact of migrant peers on native students' academic performance, with some studies suggesting negative effects due to classroom environment changes and others finding no significant impact. The researchers acknowledge that the ambiguous findings might stem from various factors, including the strain on school resources due to the influx of migrant students and adjustments in teachers' teaching approaches. The literature also considers potential parental investment strategies, such as reinforcing the human capital of more advantaged children or adopting a compensatory approach for less-advantaged children. The possibility of 'native flight' from public schools due to concerns about migrant students is also considered. Finally, the study touches upon the potential 'competition effect,' where the presence of migrant children might intensify competition for education investment among local households.

The study utilizes data from the China Educational Panel Survey (CEPS), a nationally representative longitudinal survey of middle school students. The researchers restrict their sample to students in urban public middle schools employing random student-classroom assignment. This is done to address potential selection bias and to establish a causal link between the proportion of migrant children in a class and household human capital investment. The researchers further refine the sample by excluding schools where class assignments are based on test scores, as reported by school principals and teachers. The final sample includes 4127 students (799 migrant and 3518 local) from 152 classes across 41 schools. The main independent variable is the proportion of migrant children in a class. Dependent variables include total education expenditure, in-school expenditure, out-of-school expenditure, and parental time investment. The analysis employs ordinary least squares (OLS) regression models, with grade-by-school fixed effects to control for within-school variations and standard errors clustered at the class level to account for heteroskedasticity and intra-class correlation. Heterogeneity analysis is conducted based on household socioeconomic status (SES), student grade level, and gender. Mechanism analysis explores the underlying mechanisms through surveys investigating parents' beliefs about their children's academic performance, learning environment, and teachers' perspectives. The study also investigates the competition effect by analyzing the relationship between the proportion of migrant children and the ranking distance among peers.

The key findings indicate a significant positive relationship between the proportion of migrant children in a class and household educational spending for local students. A one-percentage-point increase in migrant student proportion is associated with a 2.3% increase in total education expenditure and a 4.8% increase in out-of-school expenditure. This positive effect is more pronounced among boys, ninth-graders, and students from high-SES families. The mechanism analysis reveals that this increased spending is not driven by actual negative impacts on the learning environment or academic achievement of local students. Instead, it stems from parents' overestimation of their children's academic performance and unfounded concerns about negative peer effects. The study finds no significant effects of migrant children on classroom learning environment, teaching effectiveness, students' emotional well-being, or actual academic test scores. Furthermore, the 'competition effect' hypothesis is not supported by the data. Robustness checks using alternative measures of out-of-school education investment and additional controls yield consistent results. Heterogeneity analysis shows that the positive effects are more prominent among boys, ninth graders, and students from families with high parental education and economic status. The presence of rural migrant children leads to a more significant increase in educational spending by local urban families compared to the presence of urban migrant children.

The findings challenge the assumption that migrant children pose a threat to the educational outcomes of native students. The increased educational spending by local families appears to be driven by biased beliefs and unfounded fears rather than actual negative peer effects. This highlights the importance of addressing biased perceptions and promoting accurate information about the integration of migrant children into the education system. The study's findings have implications for educational inclusion policies in China, suggesting the need for interventions that promote equal access to high-quality education for all students, regardless of their migration status. The limited effectiveness of compensatory investments in improving academic outcomes suggests the need for a more holistic approach to educational policy, one that addresses systemic inequalities in access to quality education.

This study reveals a surprising positive correlation between the presence of migrant children and increased educational investments by local families in urban China, driven primarily by misperceptions rather than actual negative peer effects. This underscores the need for policies that promote inclusive education and address biased beliefs. Future research could explore the long-term impacts of this increased investment, investigate the role of information dissemination in shaping parental perceptions, and examine the generalizability of these findings to other contexts with varying levels of migration and educational market competitiveness.

The study acknowledges several limitations. The inability to analyze heterogeneous effects based on the duration of migrant children's local residence and the lack of information on other forms of household time investment are noted. The focus on middle school students limits the generalizability of the findings to other age groups. The study's focus on internal migration in China also necessitates caution when extrapolating the results to other contexts with different migration patterns and levels of educational market competition.