Research

Directed search on the marriage market - JOB MARKET PAPER

This study applies a static two-sided directed search model to the marriage market to unravel male and female preferences over partner characteristics (i.e., educational attainment) and terms of marriage (i.e., intrahousehold labor supply division). The analysis focuses on racial differences in marital sorting by employment status in the U.S., estimating the model separately for White, Black, and Hispanic populations. Using ACS 5-year PUMS data (2015–2019) and exploiting regional variation in gender ratios across the U.S., the model recovers equilibrium parameters governing preferences and market conditions. Counterfactual analyses then assess the extent to which observed racial differences in matching outcomes are driven by (i) differences in preferences, (ii) differences in the educational distribution, and (iii) differences in marriage market structure (capturing gender ratios and search frictions). By combining structural estimation with targeted counterfactuals, the paper provides new evidence on the origins of racial variation in marital sorting and the role of market forces in shaping intrahousehold specialization.

Consideration sets on the marriage market - SUBMITTED

This paper integrates the notion of consideration sets in the revealed preference framework of Cherchye, Demuynck, De Rock, and Vermeulen (2017). Using a revealed preference methodology based on the assumption of marriage stability within the collective household model, this framework can identify the intrahousehold decision structure, such as the sharing rule. Based on this methodology, a comparative study is conducted across various marriage market specifications, ranging from two polar cases, the “conservative” approach (where only one’s current partner is considered) and the “full-market” approach (where everyone considers everyone), to six intermediate cases. The latter include the Cherchye et al. (2017) approach, where potential partners are considered based solely on age, as well as specifications based on five additional categories (BMI, wage, education, children, and religiosity). These categories are selected based on the assortative mating literature. The specifications are evaluated based on the level of stability, as well as the bounds and tightness of the sharing rule.

WORK IN PROGRESS