Race, class, and marriage: Components of race differences in men’s first marriage rates, United States, 1960–2019

Background: Wilson (1987) argued that race differences in the frequency of marriage from the 1960s to the 1980s resulted from a shortage of marriageable men in the Black community. A large literature used spatially defined measures of male marriageability to predict marriage rates of women. These studies concluded that the availability of marriageable men can explain only a fraction of race differences in marriage. I argue that this finding may reflect errors in the measurement of the availability of marriageable spouses. Methods: My analysis assesses marriage rates from the perspective of men instead of women, allowing direct assessment of men’s economic positions without resorting to fuzzy spatial indicators. I develop new measures of first marriage rates for the 1960–1980 censuses and combine them with survey-based measures for 2008–2019. I use classic decomposition methods to assess the changing relationship of economic composition to race differences in male first marriage rates over the 1960–2019 period. Results: The analysis shows that in the mid- to late 20th century, race differences in economic composition were sufficiently large to account for most race differences in first marriage rates, but the size of the economic component declined dramatically over time. Conclusions: With the decline of male-breadwinner families since the late 20th century, the role of male economic circumstances for race differences in marriage rates has diminished, but it remains substantial. Contribution: Leveraging a new research strategy and a new measure of first marriage, this analysis provides a consistent decomposition of race differences in first marriage rates over six decades.