My dissertation is a study of wage stickiness in the Great Depression in the United States. The central research questions are threefold: Were wages in fact sticky during the Depression? What were the consequences of that stickiness in terms of employment? Why were wages sticky during the Depression?
The first central chapter deals with the first two questions. My answer to the first question is a qualified yes. The conventional wisdom that wages were sticky in the Depression relies largely on the standard time series on average hourly earnings. In response to the charges of some researchers that compositional changes in the work force create a false picture of wage rigidity in those standard time series, I compute estimates of the various aggregation biases in these data and find them to be small and often offsetting. The changes in the aggregate earnings data, then, give a reasonable picture of the movement of wages, and the picture is by and large one of sticky wages.
Much of my research into the second question involves a comparison of wages and employment in the Great Contraction of 1929-33 and in other contractions, including those of the 1890s, early 1920s, and early 1980s. Economists have long believed that wage stickiness was the mechanism by which aggregate-demand shocks were transmitted into the Great Depression. I find, however, that the degree of wage stickiness was no greater in the 1929-33 contraction than in the other contractions studied. In particular, I find the conventional wisdom of flexible wages in the slump of 1920-22 to be unsupported by the data, as the growth of real product wages in that contraction was actually larger and more permanent than that of the Great Contraction. These comparisons indicate that the relative severity and length of the major contractions of the past century owe little, if anything, to the degree of wage rigidity in those contractions.
In my second central chapter I test various theories of wage stickiness against available quantitative and qualitative data from the Great Depression. Labor-supply theories, such as Lucas and Rapping's "misperceptions" model, fare poorly. Morale-based versions of efficiency-wage and implicit-contract theories fare somewhat better. I conclude with an informal theory of my own, in which risk-averse employers initially respond to downturns with a combination of layoffs and hours reductions but avoid nominal wage cuts as potentially damaging in the long term to employee morale and productivity.
The third central chapter is a narrative account of employers' wage policies in the quarter-century from 1915 to 1940, using my morale-based theory of wage rigidity as the backdrop.