Abstract: The agricultural sector has certain distinctive features over the business cycle: it is more volatile than and not positively correlated with the rest of the economy and its employment is counter-cyclical. Because of these features and even though the agricultural sector represents less than 2% of the U.S. economy, we show that agriculture plays an essential role in understanding aggregate business cycles. The inclusion of agriculture into standard business cycle analysis resolves the longstanding problems of the standard theory in matching the observed volatility of aggregate labor and the correlation of aggregate labor and productivity (the so called <93>Dunlop-Tharshis<94> observation). In addition, the role of agriculture in the economy can account for the substantial differences observed in business cycle patterns across countries. This novel implication of the model is consistent with the systematic relationship observed between business cycle patterns and the share of agriculture across countries. Our theory has two important implications. First, the model implies that as the size of the agricultural sector falls, business cycle properties across countries should converge. Second, the role of agriculture provides a simple, measurable, and contrastable explanation for the historical properties of aggregate business cycles documented by Backus and Kehoe (1992).
Keywords: Business Cycles, Agriculture, Two-sector Model.
JEL Classification: E32