Stacey Chen

Home  | Curriculum Vitae|  Research  | Course 

 

RESEARCH STATEMENT & PUBLICATIONS

Research Statement

Within-group wage inequality can be driven by unobserved heterogeneity or by unforeseeable economic shocks, which give distinct interpretations for rising inequality in the eyes of policymakers. If the inequality is driven primarily by unobserved heterogeneity, such as unmeasured earning ability, rising wage inequality would have been caused mostly by rising skill price and it should be embraced and celebrated by social planners. But if the inequality is driven primarily by randomness in the markets, policymakers should seek effective actions to reduce the degree of uncertainty and make the markets transparent. In my paper published in Review of Economics and Statistics, I took a parametric approach to disentangle these two components, while tackling selection issues in measuring wage inequality by education levels. A follow-up paper with Professor Shakeeb Khan (Duke University, USA) relaxed the parametric assumptions and generalized the methods to allow outliers in wage distributions.  In a most recent working paper, I further relax the required conditions in latent index models. I will apply these methods to a newly available data link from 1990 to 2000 US census. I will explore several plausible sources of exogenous variation in education to evaluate the causal impact of education on within-group wage inequality. 

Another recent interest of mine is female inequality in developing countries. Using birth registry and health insurance data from the entire Taiwan, where preference for boys is extremely prevalent even to date, several Taiwanese researchers (Dr. Yenchien Chen, Chi-Nan University; Professor Jin-Tan Liu, National Taiwan University) and I are investigating the causal impact of having a brother on girls’ health and education outcomes. We exploit the randomness of twin sex composition as a plausible source of variation in sibling sex. We find the impact is either nearly zero or statistically insignificant, suggesting that the rivalry of male siblings might have been offset by the spillover effects.  Furthermore, because girls tend to have more siblings due to pro male bias, we turn our attention to the population of singletons and further distinguish the direct effect of having a brother (relative to a sister), from the indirect effect via decreasing family size. But family size is another outcome given a sibling sex shock, so observed family size cannot serve as a controlled variable, though the results in previous work were built upon the presumption that family size is exogenous. To address this in our recent article, we clarify the concept of the indirect effect using the change in potential family size. We show that even if family size is exogenous, we still need to instrument for family size, as long as it depends on the treatment (sibling sex composition in this case).  We find that if there is any effect of having a brother (relative to a sister) on girls, it is mostly by indirect effects via changes in family size, not the direct effect per se.

 

Publications

Articles in Refereed Journals (3)

 

“Estimating the Variance of Wages in the Presence of Selection and Unobservable Heterogeneity” (2008) Review of Economics and Statistics 90(2), pp.275-289. (Corrected version of Table 4).

 

"The Impact of Unexpected Maternal Death on Education – First Evidence from Three National Administrative Data Links” (2009) with equal contributions from Yen-Chien Chen and Jin-Tan Liu. The 2009 American Economic Review Papers and Proceedings, 99:2, PP.149-153.  

 

“Estimating Effective Subsidy Rates of Student Aid Programs” (2010) Annales d Economie et de Statistique, 91, PP. 401-417.

 

 

Work in Progress

Articles under Revision for Resubmission

“Did Vietnam Veterans Get Sicker in the 1990s? The Complicated Effects of Military Service on Self-Reported Health” with equal contributions from Joshua Angrist and Brigham Frandsen (current version: January 2010) Revised and Resubmitted to the Journal of Public Economics. National Bureau of Economic Research Working Paper W14781.

“Long-Term Effect of Vietnam-Era Conscription: Schooling, Experience, and Earnings” with equal contributions from Joshua Angrist (current version: November 2008) Revising for American Economic Journal: Applied Economics. National Bureau of Economic Research Working Paper W13411.

“The Impact of Sibling Sex Composition on Women's Educational Achievement” with Yen-Chien Chen and Jin-Tan Liu (current version: October 2009). Submission planned: May 2010.

“Separate Effects of Family Size and Sibling Sex Composition on Education: Methods and First Evidence from Population Birth Registry” with Yen-Chien Chen and Jin-Tan Liu (current version: May 2009). Previously known as “We Prefer Sons, But Does It Matter?”  Submission planned: December 2010.

“Estimation of Causal Effects of Education on Wage Inequality Using IV Methods and Sample Selection Models” with Shakeeb Khan (current version: February 2007).

“Does College Teach Young Men to Smoke Pot?” (current version: 2006)

Other Drafts

“Existence of Rational Continuous Social Choice on Infinite Discrete Alternatives” with Wu-Hsiung Huang (current version: October 2008).

“Estimate the Educational Composition Effect on Wage Inequality in the Roy Model” (current version: November 2009).