Abstract: we investigate the causal effect of education on health using an instrumental variable approach. The instruments we employ consist of two institutional changes in China that generated discontinuities in educational attainment among individuals. To ensure the validity of the instruments and obtain prudent conclusions, we adopt more restrictive identification tests than previous studies. The results indicate no causal impact of education on either perceived health or anthropometric health. With regard to the impact of education on male health behavior, namely smoking, we cannot provide conclusive results due to a violation of the exogeneity of our instruments. Nevertheless, we can confirm that education has no causal effect on female health behavior. To overcome the widely documented shortage of quasi-experimental identification, we also employ spouse`s education as an alternative instrument to examine the causal effect of education. Identical results are obtained, with the exception that the impact of education on the reduction of overweight among women becomes significant. We conclude that this provides some evidence of a causal impact of education on health.
Key Words: Education; Health; Instrumental variable;
Compulsory education law; Spouse`s education
该文发表在《China Economic Review》(SSCI)2014第29期,第1-18页（全文见网址：http://dx.doi.org/10.1016/j.chieco.2013.12.003）。
