Household Search or Individual Search: Does It Matter? Evidence from Lifetime Inequality Estimates

We develop and estimate an household search model to evaluate if ignoring that labor market decisions are taken at the household level - as usually done in search models of the labor market - has relevant empirical consequences. We evaluate the impact of this potential mispecification error by comparing parameters estimates under different specifications, by running a specification test, and by generating simulations to compute cross-sectional and lifetime inequality measures. We build on previous literature (Dey and Flinn (2008) and Guler, Guvenen and Violante (2011)) to propose a novel identification strategy of the risk aversion parameters. We find that ignoring the household as the actual unit of decision-making leads to estimate gender differentials in average wage offers two and a half times larger and to erroneously conclude that women experience higher inequality in wages and earnings than men.