[Twitter; supplement] We analyze sectoral labor reallocation and the reversal of urbanization in the United States during the Great Depression.
The widespread movement to farms, which serves as a form of migratory insurance during the crisis, is largely toward farms with low levels of mechanization. In contrast, the mechanized agricultural sector sheds workers, many of whom reallocate into low-productivity or subsistence farming.
The crisis perverts the normal process of structural change in which workers displaced by farm equipment are released into more productive occupations, suggesting that macroeconomic fluctuations are an important factor determining the labor market consequences of technological change.
…Our main empirical strategy makes use of a novel instrument for modernized agricultural production. We show how land topography—specifically the average slope, or ruggedness, of the land—influences the suitability for large-scale mechanized agriculture. Farm areas with smoother, less rugged land are more amenable to mechanization and thus exhibit more capital-intensive production. During the crisis, these areas experience relative declines in population and farm employment. We argue that these effects are driven by the characteristics of the farm sector, and we attempt to rule out alternative explanations. We also show that this relationship between ruggedness and farm migration arises only during the severe downturn of the Great Depression, suggesting that the impact of technological change on labor markets depends on broader macroeconomic conditions.
These findings highlight the importance of interactions between short-term macroeconomic fluctuations and the longer-run process of structural change. Instead of releasing labor to the non-farm sector, as predicted by models of structural transformation, the workers driven off mechanized farms actually reallocate into the lower-productivity subsistence agricultural sector. Thus the “normal” process of sectoral reallocation is obstructed by the crisis. It is not simply that the process slows down or stalls; instead, it takes a perverse form that may actually impede both the economic recovery and the longer-run development process.
