Abstract

Our aim in this paper is to obtain a measure of the potential benefit of re- ducing the likelihood of economic crises. We define an economic crisis as a Depression-style collapse of economic activity. Based on the observed frequency of Depression-like events, we estimate this likelihood to be approximately once every 83 years for the US. Even for this small probability of transiting into a Depression-like state, the welfare gain from setting it to zero can range between 1.05 percent and 6.59 percent of annual consumption, in perpetuity. These large gains arise because even though the probability of encountering a Depression- like state is small, it is highly persistent once it occurs. We also find that for some calibrations of the model, uninsured unemployment risk contributes significantly to the size of these gains.

Citations