In the US, married households have more net worth per working-age adult than single households. In the 1960s, the gap was associated with single-earner married households. Today, dual-earner married households have become the wealthiest as women play an important role in the labour market. We analyse these trends, combining rich household portfolio data with a structural life-cycle model. The evolution of wealth within and across families with different structures has key implications for wealth inequality because of the high concentration of high-earning individuals in dual-earner households.

We leverage the quasi-random assignment of Swedish men to army units and its subsequent effect on peer exposure to study the causal effect of marriage on labour market outcomes. Social norms play a central role in the decision to get married, and army peers affect these norms in a crucial stage of life. We construct measures of exposure to marital norms and leverage the quasi-random exposure of Swedish men to these norms in an instrumental-variable setup to study the effect of being married on the labour market outcomes of men.