Time: 15:00 -- 16:30, April 24th

Location: 1610, Rear Main Building

Speaker: Xue Sen

Abstract：The role of locus of control in shaping socioeconomic behaviour has been investigated extensively over the past decade. Most studies estimate the effect of locus of control using OLS or other cross-sectional methods. This approach may suffer from omitted variable problems, such as shared family background and genetic factors, and thus lead to biased estimates of the role of locus of control. In this paper, we examine the correlation between locus of control and a number of socioeconomic variables. We adopt a twin design in an attempt to mitigate the omitted variable problem. As in the existing literature, we find that the OLS estimates of the impact of locus of control for both monozygodic and dizygodic twins are statistically significant and economically sizable. In addition our analysis finds sizeable and significant locus of control effect for dizygodic twins using a within-twins fixed effect estimator. However, the within-twins fixed effect estimates for monozygodic twins are almost without exception statistically insignificant and economically small in size. This finding suggests that OLS estimates of locus of control may be substantially biased due to omitted variable problems, and thus an improved identification strategy is called for in future research.

Speaker introduction: Xue Sen, PhD, Australian National University, Assistant professor, Institute of Economic and Social Research, Jinan University, deputy Director of data Survey Center, Jinan University. His research interests include: Labor economics, Development economics, microeconometrics, Chinese economy. He has published several papers on SSCI journals such as Journal of Comparative Economics and China Economic Review, and is an anonymous reviewer of journals such as China Economic Review.