Abstract: Two concerns in US education are shortages of qualified math and science teachers and the feminization of teaching. I draw on the National Education Longitudinal Study of 1988 to identify a common reason for the two phenomena. Controlling for self-selection by a propensity score matching estimator and the Heckit, I find that the rigid pay structure for teachers, rather than the pay level, appears to generate both problems. Specifically, the returns to math skills in non-teaching are twice as high as those in teaching. The higher returns to math skills in non-teaching attract individuals with high math skills to the non-teaching sector (shortages of qualified math and science teachers). An earnings gap, potentially from discrimination, occurs against women in non-teaching. Along with this earnings gap, the lower average math skills of women than those of men provide women with an incentive to teach (the feminization of teaching). These findings imply that if teachers are paid commensurate with their math skills, individuals with high math skills would be attracted to teaching. The majority of them are expected to be men considering their current advantage over women in math.

“Teacher turnover: An issue of workgroup diversity” (2009, Education Policy Analysis Archives, 17(11), 1-23.)

Abstract: Effective responses to continual quantitative and qualitative shortages of teachers require understanding of the supply of teachers. One neglected aspect of the supply is a recent increase in the proportion of minority teachers. Using the Schools and Staffing Survey and the Teacher Follow-up Survey, I estimate the relationship between workgroup racial diversity and turnover of white teachers. I find that young white teachers are more likely to stay in their original schools when the proportion of minority teachers is smaller. However, the opposite pattern emerges for older teachers. This poses a policy dilemma for catching two rabbits at one time, i.e. recruiting and retaining teachers on the one hand and diversifying teaching staff on the other hand.

“A new insight into the gender gap in math” (presented at an annual conference held by the European Asssociation of Labour Economists in 2008)

Abstract: The US gender gap in math has declined, and the gap, if any, has usually appeared only in adolescents over the last several decades. Recently, it has been reported that a gap appeared in children. Moreover, the gap widened with advancing grade-levels in school. Using the Early Childhood Longitudinal Study Kindergarten Cohort, I estimate the counterfactual distribution of girls’ math scores and measure the gap for each quantile. I find that girls at the top of the distribution fall behind boys but they catch up to boys later. At the same time, girls in the lower parts of the distribution lose ground. In fifth grade, a gap of 0.2 SD is observed across the entire distribution. The results demonstrate important dynamics of the gap that the mean gap fails to show but that are relevant for the purpose of policy.

“The Dynamics of the evolution of the black-white test score gap” (presented at an annual conference held by the Southern Economic Association in 2008)
Abstract: Recently the black-white test score gap came to a complete stop and even seemed to start reversing itself after a vigorous closing. Relying on the concept of the average treatment effects on the treated, I apply a quantile version of the Oaxaca-Blinder decomposition to estimating the counterfactual distribution of the test scores of black students. From the Early Childhood Longitudinal Study, Kindergarten Class of 1998-1999 (ECLS-K), I find that initially the gap appears only at the top distribution of test scores. As children age, however, the gap at the top shrinks, whereas the gap in the mid-part of the distribution grows. Moreover the gap due to differences in the effect, rather than the amount, of characteristics becomes important. The findings suggest that researchers look beyond the mean gap and the amount of characteristics to reduce the test score gap.

“The monopsony power in the US teachers labor market” (2008, Empirical Economics Letters, 7(6), 553-559.)
Abstract: Teacher shortages in the United States have not been resolved for more than a half century. When shortages persist in a labor market, monopsony can be one reason. The standard search model is used to derive a monopsony measure that is robust to the assumptions in the previous literature. Drawing on the Current Population Survey from 1994 to 2006, I find that the monopsony measure is substantially higher in the teacher labor market relative to other comparable labor markets. Additional evidence is provided that the monopsony power in the teacher labor market is not affected by the general economic activities.

 Abstract: Although people constantly search for partners, there are always some ones who are left as singles. This paper provides a simple model to demonstrate that this can happen even in the environment most conducive to partnership, when monitoring infidelity is not perfect. The intuition is that, when having a partner is preferred to being single, being single can be used as a discipline device to prevent infidelity.

 “School and subject choices in education” (2007, Economics Bulletin, 9(13), 1-7.)