Swiss Institute for Empirical Economic Research (SEW)
University of St. Gallen
It has long been observed that married men earn higher wages than their single counterparts. In this paper, we document that, in the last decades, an analogous pattern has emerged for women. Married women experienced a wage penalty until the 1990s, whereas nowadays there is a sizable premium. To measure the causal effect of marriage on wages presents three main challenges: a significant part of the female population does not participate in employment (sample-selection bias), there might be some variables that are relevant for both wages and the propensity to marry that are not observable (omitted-variable bias), and wages may also affect marriage decisions (simultaneity bias). We apply a variety of techniques, along with a novel instrument based on local social norms towards marriage, to show that marriage has a positive causal effect on wages for both genders, although a sizable part of the observed correlation is spurious. We also show that the effect of marriage on wages is heterogeneous. Further, we present evidence that the main hypotheses discussed in the literature to explain the marriage wage premium for men, household specialization and employer discrimination, have little support in the data.
We study unemployment insurance in a framework where the main source of heterogeneity among agents is the type of household they live in: some agents live alone while others live with their spouses as a family. Our exercise is motivated by the fact that married individuals can rely on spousal income to smooth labor market shocks, while singles cannot. We extend a version of the standard incomplete-markets model to include two-agent households and calibrate it to the US economy with special emphasis on matching differences in labor market transitions across gender and marital status as well as aggregate wealth moments. Our central finding is that changes to the current unemployment insurance program are valued differently by married and single households. In particular, a more generous unemployment insurance reduces the welfare of married households significantly more than that of singles and vice-versa. We show that this result is driven by the amount of self-insurance existing in married households and, thus, we highlight the interplay between self- and government-provided insurance and its implication for policy.
In the US economy, black males on average receive lower wages than their white counterparts. This difference in wages increases over the working life. At the same time, the probability of being employed is lower for black than for equally educated white males. Notably, the black-white gap in employment is almost constant over the lifecycle. These two facts suggest that the determination of the earnings gap is related to on-the-job human capital accumulation. We propose a model of on-the-job human capital accumulation with labor market frictions to quantitatively assess how much of the black-white earnings gap can be accounted for by differences in employment probabilities versus pre-market factors. Conditional on education, we find that differences in employment probabilities between blacks and whites account for 26% of the aggregate labor earnings gap over the working life. Together with differences in the education distribution, employment probabilities can explain almost half the gap.
In this paper we document that married individuals face a lower unemployment rate than their single counterparts. We refer to this phenomenon as the marriage unemployment gap. Despite the dramatic demographic changes in the labor market over the last decades, this gap has been remarkably stable both for men and women. Using a flow-decomposition exercise, we assess which transition probabilities (across labor force states) are behind the marriage unemployment gap. We find that, for men, the higher attachment to employment of married males is the main driver of the gap. For females, we find that the participation margin plays a crucial role.