Papers:

"The Ins and Outs of Cyclical Unemployment," with Ryan Michaels, and Gary Solon. Forthcoming at the American Economic Journal: Macroeconomics. Click here for paper.

Older version also available as NBER Working Paper No. 12853.

Responds to a recent literature on the extent to which the increased unemployment during a recession arises from an increase in the number of unemployment spells versus an increase in their duration. Like the recent literature, we find an important role for increased duration. But contrary to recent conclusions, we find an important role for increased inflows to unemployment as well. Moreover, we find that increased inflows are dominated by increases in job loss at the start of an unemployment ramp-up. This suggests that recent emphasis on explaining increased unemployment duration around recessions can at best explain only part of the cyclical variation in unemployment.

"Marginal Jobs, Heterogeneous Firms, & Unemployment Flows," with Ryan Michaels. Click here for paper. Also available as NBER Working Paper No. 13777.

Much recent research has sought to explain the cyclical amplitude of unemployment fluctuations in the US. A common solution has been to make the unconventional assumption of rigidity in the wages of newly hired workers. This paper shows that amplification of the cyclical variation of unemployment can be obtained from adding two very conventional features to a simple model of the aggregate labor market, namely downward sloped short run labor demand and endogenous job destruction. This generalized model is able to more closely match the cyclicality of both job finding and employment to unemployment flows observed in US data. Contrary to the standard search model, the generalized model can match the data whilst maintaining realistic surplus to employment relationships. In addition, we uncover a novel source of amplification of cyclical shocks that is generated by the interaction of countercyclical unemployment inflows and job creation.

"Evaluating the Economic Significance of Downward Nominal Wage Rigidity." Click here for new and hopefully better version. Older version also available as NBER Working Paper No. 12611.

The existence of downward nominal wage rigidity has been abundantly documented, but what are its economic implications? This paper demonstrates that, even when wages are allocative, downward wage rigidity can be consistent with weak macroeconomic effects. This occurs because firms have an incentive to compress wage increases as well as wage cuts when downward wage rigidity binds. By neglecting potential compression of wage increases, previous literature may have overstated the costs of downward wage rigidity to firms. Using a broad range of micro--data from the US and Great Britain I find that firms do indeed compress wage increases as well as wage cuts when downward wage rigidity binds. Accounting for this reduces the estimated increase in aggregate wage growth due to wage rigidity to be much closer to zero, consistent with the predictions of the model. These results suggest that downward wage rigidity may not provide a strong argument against the targeting of low inflation rates, as practiced by many monetary authorities.

Suggestive evidence that wage setters do indeed think along these lines:

"General Motors Corp’s historic health care deal with the United Auto Workers will require active workers to forgo $1-an-hour in future wage hikes." Detroit News, October 21st 2005.

“[Business leaders] take account of the fact that, if they raise the level of pay today, it will remain high in the future. I hear a lot about this last point now. [...] Some say that they are not now increasing pay [...] because they know they will not be able to reverse the increases during the next downturn.” Bewley (2000), p.46.

"A Simplified Approach to Irreversible Investment & an Economic Interpretation" Click here for paper. [Preliminary: Comments Welcome]

This paper demonstrates that the optimal investment policy in the presence of costly reversibility can be derived using the familiar principles of discrete time dynamic programming as an alternative to the less familiar methods of smooth pasting. In doing so, it makes the analysis of irreversible investment accessible to a wider audience of economists. Second, it shows that the discrete time approach to the optimal investment problem developed in the paper admits a surprisingly clean economic interpretation of the optimal investment policy of a firm that has not, to my knowledge, been recognized in the literature.