Cohort Size and the Marriage Market: Explaining Nearly a Century of Changes in U.S. Marriage Rates

PWP-CCPR-2018-001

Abstract

In the paper, we propose an explanation for almost a century of changes in the U.S. marriage rates. We do this in three stages. In the first stage, we use reduced-form methods to provide evidence on the following two results. First, a single variable, cohort size, can account for almost all the variation in marriage rates since the early 1930s for both blacks and whites. Second, an increase in cohort size reduces marriage rates, whereas a decline in cohort size has the opposite effect. The most convincing evidence on the relationship between cohort size and marriage rates is obtained by using as a source of exogenous variation differences across states in mobilization rates during World War II. In the second stage, we develop a dynamic search model of the marriage market that can generate the observed patterns. Using the model we show that, qualitatively, it can generate the negative relationship between cohort size and marriage rates. We then derive a testable implication from the model: an increase in cohort size reduces the age difference at marriage and vice versa. In the last stage, we investigate whether the model is consistent with the data. We first use the derived implication to test the model and fail to reject it. We then estimate the model and evaluate whether it can quantitatively match the documented variation in marriage rates. We find that the search model can explain most of the observed variation.