Markets, Productivity, and Happiness in a Historical Perspective

The new classical microfoundations of aggregate labor supply

Le Fur Tanguy, New York University Abu Dhabi
Clerc Pierrick, HEC Liège

In this article, we document how the discrepancy between the large labor supply elasticity required by the canonical real business cycle model to generate credible employment fluctuations and the lower empirical estimates from microeconometric studies fostered discussions on aggregation, heterogeneity, and the use of a representative agent in aggregate general equilibrium models in the 1980s. Debates initially pointed to the inadequacy of the representative agent to account for adjustment along the extensive margin of labor and the heterogeneity characterizing the labor market, but also underscored epistemological disagreements on the degree of aggregation needed to account for macroeconomic phenomena. Despite calls to build full-fledged heterogeneous agents models that would allow to map a variety of individual behaviors into aggregate variables and permit the importation of microeconometric estimates of preference parameters in general equilibrium models, new classical macroeconomists defended the use of a representative agent as a useful abstraction. To vindicate their position, they proposed a theory of aggregation of individual labor supply decisions showing that the elasticity of the representative agent is disconnected from individual elasticities, and can be substantially greater. We document how this aggregation theory set itself in the context of the defense of the real business cycle model following the first salvo of criticisms it faced, and show how it was largely adopted as a satisfying answer to the discrepancy. We nevertheless argue that this episode marked a substantial shift in the interpretation of the representative agent. While the representative agent was initially introduced to describe the behavior of optimizing individuals, its labor supply elasticity came to be seen as the reduced-form of some labor market phenomena causing an aggregation bias while still being treated as a structural parameter. Although this article challenges the view that new classical macroeconomists eluded the question of aggregation during the rise to prominence of the real business cycle model, we emphasize that such a change in the definition of the representative agent was a concession with respect to their microfoundational ambitions. We conclude by reviewing how developments in heterogeneous agents models have led to the reexamination of the aggregation theory behind the labor supply elasticity in the 2000s, but highlight how questions regarding the interpretation of representative agent models and their relevance to account for heterogeneity in the labor market remain salient today.


Keywords: Labor supply elasticity, microfoundations, aggregation, real business cycles