The Effect of Adjustment Costs and Institutional Constraints on Labor Supply Elasticities: Evidence from Denmark
The effects of taxes and wages on labor supply are central inputs to macroeconomic models and policy analysis. Several decades of microeconometric research has found very small intensive margin elasticities of labor supply, often not significantly different from zero (e.g., Hausman 1983, Eissa and Liebman 1996, Blundell, Duncan, and Meghir 1998, Saez 2003; see Blundell and MaCurdy 1999 and Saez et al. 2009 for recent surveys).
One interpretation of this evidence is that individual preferences lead to very inelastic labor supply curves. In this paper, we investigate an alternative hypothesis: microeconometric elasticity estimates are attenuated because of adjustment costs and institutional rigidities in the labor market.
Microeconometric studies typically examine short'run responses to small changes in tax rates for a subgroup of the population (such as single women with children) in order to identify causal effects. If workers face adjustment costs qcosts of searching for a new job, changing consumption plans, or acquiring information about tax changes qtheir responses to such tax changes could be attenuated in the short'run because the adjustment costs exceed the benefits of reoptimization. Institutional constraints, such as adherence to a 40 hour work week to exploit complementarities in production, could further limit responses to tax changes levied on a narrow subgroup of individuals. In contrast, large, economy'wide, permanent differences in tax regimes could lead to changes in labor market institutions and make the benefits of reoptimization exceed adjustment costs. Hence, the identification strategies employed in microeconometric studies could lead to underestimation of the long run elasticities relevant for tax policy analysis and calibration of macro models.
The idea that search costs and hours constraints affect labor supply decisions has been proposed in earlier work (Pencavel 1986, Altonji and Paxson 1988, Dickens and Lundberg 1993, Ham 1991, Blundell and MaCurdy 1999). However, there is little evidence on whether and how such frictions affect estimates of labor supply elasticities. Here, we present non parametric, quasi experimental evidence that adjustment frictions and institutional constraints mediate the relationship between taxes and labor supply. We also propose techniques for estimating long'run elasticities in the presence of such frictions.
To motivate our empirical analysis, we first develop a stylized model of labor supply and taxation with search costs and endogenous hours constraints. We consider a model where firms have constant returns to scale Leontief production technologies and have all employees work the same number of hours to exploit complementarities in production. Firms post hours offers, generating a distribution of wage'hours packages in the economy as in Rosen (1976) and Blundell, Brewer, and Francesconi (2008). Individuals draw job offers from the hours distribution for their wage level. If an individual is unsatisfied with the job he draws, he can find another job that offers his ideal hours by paying a fixed search cost. In equilibrium, the distribution of jobs offered by firms matches worker preferences in the aggregate economy. However, because of search costs, individuals are distributed around their optimal level of hours. This model generates three testable predictions about the relationship between taxes and labor supply.
Posted in :
