回不去的家乡?——教育公共品供给与人口回流的实证研究

When can I Go Home?School Provisioning and the Decision to Immigrate

随着我国流动人口回流趋势显现以及户籍制度改革深化,如何促进合理、公正、畅通、有序的社会性流动,正成为众多城市思考的重要问题。本文基于2001年起在全国范围内实施的"撤点并校"政策外生冲击,结合2005年全国1%人口抽样调查数据和295个地级市层面特征数据,实证研究了户籍地教育公共品供给状况对流动人口回流决策的影响。研究发现,户籍地教育公共品减少显著阻碍了流动人口回流,减少幅度越大,阻碍越严重。异质性分析结果表明,这种阻碍倾向不分户口类型,但对家中有男孩、家庭规模较小的流动人口更加明显。引入代际视角后的研究结果显示,教育公共品减少增加了流动人口家庭中适龄入学儿童迁出户籍地的概率,教育公共品供给影响两代人的迁移。上述结果意味着,改进公共资源配置效率,提高供给体系质量、优化公共品供给布局,有助于破除劳动力流动障碍,应成为未来城市工作的重要方向。

In recent years,the existing population of China has begun to relocate among different cities.This is exemplified by a new trend among China’s massive migrant population of returning to their hometowns.With the relaxation of the Hukou system,it is worth considering how to promote the reasonable and orderly flow of labor and improve the efficiency of human resource allocation.Many factors influence migration.Tiebout proposed that residents flow into the region that best matches their preferences,after considering the public goods and tax levels of each location.Studies of developed countries suggest that this mechanism exists(Bayoh et al.,2006;Kleven et al.,2013;Dustmann and Okatenko,2014;Akcigit et al.,2016;Moretti and Wilson,2017).However,due to the uniform tax system and Hukou system in China,the effectiveness of the Tiebout model is debated,especially in the context of populations returning to their hometowns.We empirically investigate the causal effect of public goods supply on a migrants’decisions to return to their hometowns.We obtain data on residential address and"usual place of residence five years ago"from the 2005 National Sample Survey to construct a dynamic variable denoting whether migrants return to their hometowns.We also construct a proxy variable for change in the number of schools at the city level according to the extended implementation of the"School Consolidation"policy in China.This is a public policy promoted in China since 2001 that aims to improve the quality of education by eliminating many smaller schools and merging them into larger institutions.Using a quasi-natural experimental approach,we divide the sample into treat and control groups based on the presence of children aged 6-10 years in the family(a child was 10 years old in 2005 when the data was collected,given that he/she started primary school at 6 years of age in 2001)and build a difference-in-differences model.Our findings are as follows.First,a decrease in the number of schools significantly hinders the return of the migrant population.Furthermore,the greater the decrease in the number of schools,the greater the negative impact on the return of the migrant population.This result still holds when considering endogeneity and other factors.Second,the results of our heterogeneity analysis show that the impact above is valid for both rural and urban areas and is more obvious among smaller families and families with boys.Third,the impact is not only on parents but also on their school-age children,meaning that the availability of educational public goods affects the migration decisions of two generations.Beyond confirming the Tiebout model,this paper contributes to the literature in at least in two additional ways.First,we demonstrate the impact of public goods on migrants returning home and enrich the literature in this field.Most existing studies exploring the impact of public services on migration are based on an outflow perspective,and little attention has been paid to returning inflows.Further,due to limited data,the available studies are more theoretical in nature.Second,our identification setup is new.We attempt to explore the dynamics of migration based on cross-sectional census data at the household level,which may inform future studies.The conclusions of this paper are informative in the context of policy decisions.In developing countries,improving the efficiency of public resource distribution and optimizing the provision of public goods such as education could help break down barriers to labor mobility and thus further promote economic growth and quality of life.For cities trying to attract talent,enhancing the provisioning of public goods could help to stimulate further population inflows.In this paper,we focus on the impact of changes in the number of schools.Future studies could assess the impacts of public education quality or other public goods(e.g.,health care)to explore the effect of public goods supply on population mobility to a broader and deeper extent.