Aggregate Employment Fluctuations and Agricultural Share

Differences in employment volatility and the correlation of employment with output across countries are often cited as examples of the limitation of standard real business cycle (RBC) theory to reproduce the observed labor market facts. These observations have lead researchers to argue for the necessity of Non-Walrasian features to reflect the labor institutions in European countries. In this paper, we show that the same labor market evidence is observed in regional economies with the same labor market institutions. We conjecture that differences in agricultural activity can generate the observed differences in labor market behavior. We show that a standard two-sector RBC model with agriculture and non-agriculture can account for the observed labor market facts. In particular, as the size of agricultural activity increases, aggregate employment volatility and the correlation between aggregate employment and output decrease. Moreover, contrary to the Non-Walrasian approach to business cycles, agricultural activity can account for the correlation between aggregate employment and output as reported by Danthine and Donaldson (1993) for Europe and the U.S.