Economies of scale, resource dilution and education choice in developing countries: Evidence from Chinese households

•We document a puzzle: the quantity-quality trade-off theory holds true in developed countries but not in developing world.•The scale economies effect is introduced into the model of Beck and Lewis (1973) to explain the divergent empirical findings.•The interaction of scale economies effect and resource dilution effect determines an inverted U-shapedcorrelation.•Empirical analysis using China's 1990 and 2000 censuses gives supporting evidence for our hypothesis.

Recent empirical studies suggest that the negative correlation between the quantity of children within a family and their educational attainment, which is widely observed in developed countries, is inconsistent or even rejected in developing countries. This paper aims to integrate these divergent empirical results into a unified theoretical framework by introducing scale economies into the classical model of Becker and Lewis (1973). As a result, the “anomaly” of an observed upward or an inverted U-shaped relationship can be expected as the scale economies effect dominates when there are few children in a family. Using data from the China's 1990 and 2000 censuses, this study further tests some hypotheses induced from the model. Educational attainment increases with sibling size at first and then drops. Children with one or two siblings achieve the highest education during the period our sample covers. The inverted U-shaped correlation is more robust for rural subsamples, for older cohorts and for economically underdeveloped regions and groups, which is consistent with the prediction of the model.