I study the interaction between discrimination and investment using a directed search model where firms decide the capital intensity of their production technologies before being matched. Discrimination makes some workers cheap to hire. As a consequence, some firms might save on capital costs adopting labour intensive technologies. This framework allows to reconcile search models with three well-known facts regarding the labour market outcomes of minority workers: low wages, high unemployment and occupational segregation. Furthermore, the model questions the role of equal pay legislation in reducing inequality since removing this restriction, i.e., allowing firms to post type-contingent wages, eliminates the negative effects of discrimination on investment and wages.

In this paper we investigate the sorting of workers within and across firms to understand gender gaps in labor market outcomes. Using Danish employer-employee matched data, we find a stronger tendency toward positive sorting in job-to-job transitions for female workers with respect to men, while it is weaker for promotions, especially after motherhood. Our findings suggest that career impediments in certain firms cause the most productive female workers to seek better jobs in firms where there are less gender biases. Hence, firms heterogeneity with respect to their attitude toward female workers is crucial to understand gender differences in sorting.

In this paper, we study a local public good game in an endogenous network with heterogeneous agents. We consider two specifications in which different networks arise. When agents differ in the cost of acquiring the public good, active agents form hierarchical complete multipartite graphs; yet, better types need not have more neighbors. When agents' benefits from the public good are heterogeneous, nested split graphs emerge in which investment need not be monotonic in type. In large societies, few agents produce a lot and networks dampen inequality for most agents under cost heterogeneity and increase it under heterogeneity in benefits.

This paper investigates the impact of school peers on an individual's adult romantic relationships. In particular, we consider the effect of quasi-random variation in the share of black students within an individual's cohort on the percentage of adult relationships that are with blacks. We find that the share of blacks in a given cohort has a positive impact on the share of relationships with blacks later in life. The results remain when we consider cohabiting couples, and also when we focus on a subsample of relationships that were formed in a different state from that where an individual attended school. This suggests that the racial mix of schools has an important impact on racial attitudes.