Labor Supply Estimation Biases from Disregarding Nonwage Benefits
Published in: Economic Inquiry, Volume 55, Issue 2 (April 2017), pages 1064-1090. doi: 10.1111/ecin.12405
Posted on RAND.org on May 10, 2017
|PDF file||1.5 MB||
Use Adobe Acrobat Reader version 10 or higher for the best experience.
Read MoreAccess further information on this document at Economic Inquiry
This article was published outside of RAND. The full text of the article can be found at the link above.
Labor supply models and research underpinned by labor supply decisions often assume that workers' choices are functions of wage and wage offers. However, the literature shows evidence that such decisions at least partly depend on nonwage benefits encompassed in jobs and occupations. In this paper, I develop and estimate a stochastic dynamic model of occupation and job choice, where nonwage benefits are directly incorporated into the decision alongside wages (a full model). Nested within the full model is a wage model, which represents the common practice in the literature of basing selection solely on wages and disregarding nonwage benefits. I separately estimate the full model and the nested wage model to compare the implications (biases) of omitting nonwage benefits. I compare the two models' estimates of elasticities and an inequality reduction intervention policy. I find that disregarding nonwage benefits generally causes biases. There are cases when the two models predict very similar outcomes and have close estimates, such as in occupation-specific elasticities and job transition elasticities. But these special cases are products of canceling biases. In most cases, ignoring nonwage benefits will bias estimates by overestimating the importance of wage in the selection process and by disregarding changes in relative prices between wage and nonwage benefits.