Working Papers

If affirmative action policies induce more exposure of whites to minorities on the job, then this can theoretically improve race relations making these policies only necessary in the short run. However, it has been argued that affirmative action policies can have negative long-term consequences on the minorities they are designed to help if the policies result in resentment among members of the majority group. This paper explores the broad relationship between affirmative action policies and race relations by examining how interracial marriage patterns -- a readily available measure of race attitudes -- change in response to the enactment of state affirmative action laws. Specifically, this paper exploits time and state variation in initiating affirmative action laws along with fact that state affirmative action policies directly affect only public-sector employees. Using a triple difference model, I find that the probability of a white male having a black wife increases by 0.07 percentage points in response to being exposed to affirmative action policies given when he married, where he lives, and whether he is a public sector worker. Results for black males are sensitive to estimation technique. Furthermore, interracial marriage decisions for females, regardless of race, are not affected by state affirmative action policies. Taken together, these results suggest that affirmative action laws improve race relations but with important gender and racial differences.


Theoretical models have ambiguous predictions on how workplace gender composition affects the incidence of marriage. Marital search theory suggests that having more opportunities for interactions between members of the opposite gender increases the likelihood of marriage. Yet, according to overload choice theory, people with more options could actually delay or forgo marriage if the increase in the number of choices makes it more difficult for them to make marriage decisions. I explore how changes in the gender composition within occupation and industry over the past 40 years affect marriage decisions. I find that a higher share of opposite gender coworkers within a person’s occupation-industry is associated with a decreased likelihood of ever having been married.

Affirmative Action Bans and Black-White Wage Gaps: Evidence from Public Sector Workers (with Kenneth Couch)

Virtually all debates about affirmative action policies end by raising the topic of whether affirmative action need not be permanent. Since the 1990s, nine states have banned affirmative action in public sector employment or public educational institutions, which provides a natural experiment to test this argument. This paper estimates the impact of banning affirmative action in the public sector on Black-White wage gaps. Using data from the 1990-2019 Current Population Survey (CPS), we employ a triple-difference model and find that affirmative action bans widen Black-White wage gaps in the public sector. Remarkably, the bans decrease Black male wages in the public sector by 3.8 percentage points compared to their White counterparts. We argue that affirmative action had been effective while in place but failed to create persistent effects in reducing inequality.

Works in Progress