Home Address:
  1214 Chapel Street, #23
  New Haven, CT 06511

Telephone: (203) 787-2931 (home)
                   (203) 432-5779 (fax)

Macroeconomics
Search Theory
Labor Markets

Macroeconomics
Labor Economics
Econometrics

May 2002 (Oral) Macroeconomics, Econometrics
May 2001 (Written) Microeconomics and Macroeconomic Theory

Worker Preferences, Precautionary Savings, and Equilibrium Unemployment

Professor Giuseppe Moscarini
Professor Björn Brügemann
Professor George Hall

May 2006

Ph.D., Economics, Yale University, expected May 2006
M. Phil., Economics, Yale University, May 2003
M.A., Economics, Yale University, May 2001
B.S., Economics and Mathematics (with honors), University of Washington, June 1999

Yale University Dissertation Fellowship, 2004
Yale University Graduate Fellowship, 2000–2004
Yale University Summer Fellowship, 2001 and 2002

Instructor
Intermediate Macroeconomics, Yale University, Summer 2004 and 2005
GMAT Preparation Course, The Princeton Review, Chicago, IL 2000
Freshman Interest Group, University of Washington, Fall 1998 and 1999

Teaching Assistant
Introductory Macroeconomics, Professors Gerald Jaynes and William Nordhaus, Fall 2003 and 2005
Ph.D. Macroeconomic Theory, Professors George Hall and Giuseppe Moscarini, Spring 2005
Ph.D. Macroeconomic Theory, Professors George Hall and Stefan Krieger, Spring 2003
Master’s Macroeconomic Theory, Professor Shin-ichi Fukuda, Fall 2002

Research Assistant, Professor Thomas Sargent, New York University, 2002–2004
Research Assistant, Professor George Hall, Yale University, Spring 2003
Research Assistant, Professor Donald Brown, Yale University, Summer 2002

Professor Giuseppe Moscarini
Department of Economics
Yale University
PO Box 208268
New Haven, CT 06520-8268
Phone: (203) 432-3596
Fax: (203) 432-2128
Email: giuseppe.moscarini@yale.edu

Professor Björn Brügemann
Department of Economics
Yale University Yale University
PO Box 208268
New Haven, CT 06520-8268
Phone: (203) 432-6217
Fax: (203) 432-2128
Email: bjoern.bruegemann@yale.edu

Mortensen and Pissarides’ (1994) job matching model has become a standard framework for understanding the behavior of labor markets from a macroeconomic perspective. However, as Shimer (2005) noted, the standard model is unable to generate the observed volatility of the vacancy-unemployment (v/u) ratio. Specifically, "in the U.S. the vacancy-unemployment ratio is 20 times as volatile as average labor productivity, while under weak assumptions, search models predict the vacancy-unemployment ratio and labor productivity have nearly the same variance." My dissertation addresses this shortcoming and demonstrates that changing worker preferences and allowing them to save over time increases the volatility of the vacancy-unemployment ratio.

The first chapter of my dissertation, Risk Aversion in a Search and Matching Model, examines the impact risk aversion has on workers’ wages and the vacancy-unemployment ratio. My main result shows it is possible to solve this problem and actually increase the volatility of v/u to any desired level by choosing suitably large values of risk aversion and the worker’s Nash bargaining coefficient.

The benchmark search and matching model assumes workers are risk neutral, precluding any need for consumption smoothing via insurance when faced with aggregate productivity shocks. I show that risk aversion induces "implicit risk-sharing" between the worker and firm in a standard model with Nash bargained wages and no worker savings. In particular, wages are affected along two dimensions. First, the standard deviation decreases because risk-averse workers prefer to smooth wages over the business cycle. This in turn increases the volatility of v/u as more firms are willing to post vacancies during a boom. Intuitively, firms face a lower wage premium after a productivity increase, making it more profitable to enter the market. As a result, vacancy creation increases and the v/u ratio becomes more volatile.

Risk aversion also decreases the mean wage over the business cycle. This is a direct result of the "implicit risk-sharing": workers accept a trade-off between the standard deviation and mean wage. Alternatively, unemployment is more painful with risk aversion, and workers willingly accept lower wages to prevent a potential bargaining breakdown that would lead to unemployment. Lower wages in turn decrease the volatility of the vacancy-unemployment ratio because they imply large firm profits, which mitigate the benefits of a given productivity increase. Therefore, by increasing the worker’s bargaining coefficient, I am able to increase the mean wage while maintaining its low standard deviation implied by risk aversion. Hence, the volatility of the vacancy-unemployment ratio can be made arbitrarily large.

The "risk-sharing" is implicit because it stems from the efficiency of Nash bargaining. As in the standard model, wages are continuously renegotiated and there are no explicit insurance contracts.

The second chapter of my dissertation, Precautionary Savings in a Search and Matching Model (work-in-progress), explores the impact worker savings has on wages and the volatility of v/u. The standard model with risk neutrality is nearly equivalent to one with risk aversion and perfect savings. Intuitively, if risk averse workers are allowed to save freely, they can self-insure against wage volatility and unemployment. Hence, they would behave similar to risk neutral workers. Given this, my model with risk aversion and no savings represents the extreme alternative to its risk neutral counterpart.

What happens as I bridge this gap between the two models by allowing incomplete savings? Do we simply move back steadily toward the risk neutral model, or is there a tradeoff among risk aversion, savings, and the bargaining coefficient that can be exploited? Evidently, precautionary savings affects wages along two dimensions: it increases both the mean and standard deviation. Yet there is no reason to believe these two effects always offset one another exactly. If, for example, the mean wage increases more relative to its standard deviation, for a given level of risk aversion, the volatility of v/u will increase. In that case it would be possible to match the volatility of v/u with a more moderate Nash bargaining coefficient.