Who picks up the slack? Understanding spousal responses to unemployment spells

Abstract

We use a large panel of married households to update estimated added worker effects. In response to a primary earner’s job loss, secondary earners are 1.1 to 2.4 percentage points more likely to work and compensate for 3.6 to 5.1 percent of the displaced worker’s lost earnings. When a secondary earner is displaced, spousal employment is unchanged but there is a substantial earnings reduction. These small compensatory responses are explained by an increased probability that the nondisplaced spouse exits employment, either through correlated unemployment shocks or retirement. Conditional on relative-earner status, sex-based differences in added worker effects are small.