Is There a “Neighborhood Effect” in the Household Education Expenditure?

Abstract

As a main source of educational input, the household education expenditure has a positive impact on the accumulation of children’s human capital. In recent years, the education expenditure in Chinese families has increased rapidly. Existing literature has explained this phenomenon from different aspects, including the status seeking motive, raising sons for old age, and parents’ pure altruism. In this paper, we suggest that an alternative explanation may be at work: the education expenditure in Chinese families may be affected by their neighbors, namely the neighborhood effect. We note that there is a shortage of literature giving an explanation for the rising education expenditure in Chinese families from this perspective. On the one hand, China is a country with strong cultural tightness and laying great emphasis on collectivism. For this reason, Chinese people’s thoughts and behaviors are easily affected by others and tend to be convergent in a group. On the other hand, Chinese families generally attach great importance to education. They often regard education, especially tertiary education as an important way to achieve intergenerational mobility and improve families’ social status. In a competence-based higher education selection system, whether a child can get the access to higher education depends on the relative position of his （her） academic performance, and the household education expenditure is a key factor in determining children’s academic performance. In order to guarantee children’s educational opportunities, the education expenditure in other families within a social group will be an important element that families need to take into account when making decisions on their own household educational expenditure. Therefore, the education expenditure in Chinese families is possibly affected by other families within a social group. 　　In this paper, we define families living in a village in rural areas or a community in cities as neighbors, and empirically analyze whether there is a neighborhood effect” in Chinese household education expenditure decisions by using data from the China Family Panel and Following Studies （CFPS） from 2010 to 2014. As illustrated by some relevant literature, identifying the neighborhood effect is usually complicated by the problem of the residential sorting effect which refers to a phenomenon that households may self-select into the community with whom they associate. We alleviate this problem by limiting our research samples to rural Chinese households due to the special hukou system, controlling more variables which may affect residents’ selecting of communities, and matching a virtual neighbor to city households based on the Mahalanobis distance.　　The empirical findings indicate that the average community education expenditure has a positive impact on the household education expenditure. With 1% increase in the community education expenditure, the household education expenditure will increase by 0.307%. The results also show that the estimated effect cannot be driven by the residential sorting effect, households in a community facing identical educational policies and sharing the same access to information. We further analyze the relevant influencing mechanisms and find that in rural samples, the widening of income disparities between the middle and the high income groups and between the middle and the low income groups will make the household education expenditure more sensitive to the average community education expenditure. This indicates that the status seeking motive serves as an important mechanism for the neighborhood effect to work in rural households. In city samples, we find that both the status seeking motive and the convergence of the importance that residents attach to education are not the influencing mechanisms for the neighborhood effect to work in city households. This paper provides a new perspective to understand the increasing household education expenditure in Chinese households in recent years.