Labor Supply with Social Interactions: Econometric Estimates and Their Tax Policy Implications

Our research fleshes out econometric details of examining possible social interactions in labor supply. We look for a response of a person's hours worked to hours worked in the labor market reference group, which includes those with similar age, family structure, and location. We identify endogenous spillovers by instrumenting average hours worked in the reference group with hours worked in neighboring reference groups. Estimates of the canonical labor supply model indicate positive economically important spillovers for adult men. The estimated total wage elasticity of labor supply is 0.22, where 0.08 is the exogenous wage change effect and 0.14 is the social interactions effect. We demonstrate how ignoring or incorrectly considering social interactions can mis-etimate the labor supply response of tax reform by as much as 60 percent. Paper originally published in 2005, revised in 2006.