Research Database

Economists have offered many theories for the U.S. Great Depression, but no consensus has formed on the main forces behind it. Here we describe and demonstrate a simple methodology for determining which theories are the most promising. We show that a large class of models, including models with various frictions, are equivalent to a prototype growth model with time-varying efficiency, labor, and investment wedges that, at least on face value, look like time-varying productivity, labor taxes, and investment taxes. We use U.S. data to measure these wedges, feed them back into the prototype growth model, and assess the fraction of the fluctuations in 1929–39 that they account for. We find that the efficiency and labor wedges account for essentially all of the decline and subsequent recovery. Investment wedges play, at best, a minor role.

This article is a progress report on research that attempts to include one type of market incompleteness and frictions in macroeconomic models. The focus of the research is the absence of insurance markets in which individual-specific risks may be insured against. The article describes some areas where this type of research has been and promises to be particularly useful, including consumption and saving, wealth distribution, asset markets, business cycles, and fiscal policies. The article also describes work in each of these areas that was presented at a conference sponsored by the Federal Reserve Bank of Minneapolis in the fall of 1993.

The paper first locates quarters in the early 1970s at which the covariance matrices of the innovation vectors have shifted for the real GNPs of the USA, West Germany and Japan treated as univariate series. The paper then exhibits differences in the impulse response time profiles of the two models estimated from the data primarily before and after the break as a concise summary of the changes in dynamic interactions of the three real GNPs.

The motive to hold inventories purely in the hope of profiting from a price increase is called the speculative motive. This motive has received considerable attention in the literature. However, existing studies do not have a clear implication for how large it is quantitatively. This paper incorporates the speculative motive for holding inventories into an otherwise standard real business cycle model and finds that empirically plausible parameterizations of the model result in an average inventory stock to output ratio that is virtually zero. For this reason, we conclude that the quantitative magnitude of the speculative role for holding inventories in this model is quite small. This suggests the possibility that the study of aggregate economic phenomena can safely abstract from inventory speculation.

We introduce procyclical labor and capital utilization, as well as costs of rapidly increasing employment, into a business-cycle model. Plausible variations in factor utilization enable us to explain observed variability of real GNP with considerably smaller economy-wide disturbances. The costs of adjustment create very interesting and realistic lead and lag relationships: Employment does not peak until a full quarter after output; workweeks, effort, capital utilization, and productivity all sharply lead the business cycle.

In the 1930s, Dunlop and Tarshis observed that the correlation between hours worked and the return to working is close to zero. This observation has become a litmus test by which macroeconomic models are judged. Existing real business cycle models fail this test dramatically. Based on this result, we argue that technology shocks cannot be the sole impulse driving post-war U.S. business cycles. We modify prototypical real business cycle models by allowing government consumption shocks to influence labor market dynamics in a way suggested by Aschauer (1985), Baro (1981, 1987), and Kormendi (1983). This modification can, in principle, bring the models into closer conformity with the data. Our results indicate that when aggregate demand shocks arising from stochastic movements in government consumption are incorporated into the analysis, and an empirically plausible degree of measurement error is allowed for, the model’s empirical performance is substantially improved.