We revisit the prominent finding that women’s incomes are disproportionally often observed just below the income of their partner. So far, this bunching has been explained by couple formation or couples’ labor market decisions. We propose an additional mechanism: income misreporting in surveys. Drawing on survey and administrative data, we show that income misreporting accounts for the discontinuity in the distribution of women’s relative incomes just below the point where a woman outearns her partner. This misreporting is best explained by the role of gender norms in individuals’ self-portrayals and self-perception.
gender norms women’s relative income shares combination survey and administrative data