By Pedro S. Amaral and James C. MacGee
We document sectoral differences in changes in output, hours worked, prices, and nominal wages in the United States during the Great Depression. We explore whether contractionary monetary shocks combined with different degrees of nominal wage frictions across sectors are consistent with both sectoral as well as aggregate facts. To do so, we construct a two-sector model where goods from each sector are used as intermediates to produce the sectoral goods that in turn produce final output. One sector is assumed to have flexible nominal wages, while nominal wages in the other sector are set using Taylor contracts. We calibrate the model to the U.S. economy in 1929, and then feed in monetary shocks estimated from the data. We find that while the model can qualitatively replicate the key sectoral facts, it can account for less than a third of the decline in aggregate output. This decline in output is roughly half as large as the one implied by a one-sector model. Alternatively, if wages are set using Calvo-type contracts, the decline in output is even smaller.
In recent years, aggregate models have brought interesting answers to light regarding the origin and continuation of the Great Depression, starting with the work of Cole and Ohanian (1999). Cole and Ohanian (2001) extend it to a two sector model with wage rigidities in one sector and show that this contributed little to the Great Depression. Bordo, Erceg and Evans (2000) use a one-sector model with Taylor contracts and find that it can explain a significant part of the Great Depression. Amaral and MacGee combine the two and find only a third of the decline can be explained by these wage rigidities. Is this the end of this debate?