The allocation after an unanticipated event (often called an "MIT shock") is different from the allocation of a corresponding complete-market model that explicitly considers the possibility of the shock, even when the probability of the event approaches zero.

We create a novel measure of job search effort starting in 1994 by exploiting the overlap between the Current Population Survey and the American Time Use Survey. We examine the cyclical behavior of aggregate job search effort using time series and cross-state variation and find that it is...

This paper reconsiders the cost of business cycles under incomplete markets. Primarily, we focus on the heterogeneity in the cost of business cycles among agents with different skill levels. Unskilled workers are subject to a much larger risk of unemployment during recessions than are skilled...

This paper examines the causes of the observed increase in the average duration of unemployment over the past thirty years. First we analyze whether changes in the demographic composition of the U.S. labor force, particularly the age and gender composition, can explain this increase. We then...

This paper examines the causes of the observed increase in the average duration of unemployment over the past thirty years. First we analyze whether changes in the demographic composition of the U.S. labor force, particularly the age and gender composition, can explain this increase. We then...

This paper reconsiders the cost of business cycles under incomplete markets. Primarily, we focus on the heterogeneity in the cost of business cycles among agents with different skill levels. Unskilled workers are subject to a much larger risk of unemployment during recessions than are skilled...

This paper analyzes the implications of plant-level dynamics over the business cycle. We first document basic patterns of entry and exit of U.S. manufacturing plants, in terms of employment and productivity between 1972 and 1997. We show how entry and exit patterns vary during the business...

This paper formulates Rosenberg's (1982) "learning by using" as a stochastic process. The producer of machines learns from the experience of users. Due to this learning, the quality of machines improves over time. It turns out that the process of this improvement approximately takes an...