Inequality over the Business Cycle:
Estimating Income Risk using Consumption Data

Giorgio Primiceri and Thijs van Rens

We use CEX repeated cross-section data on consumption and income, to evaluate the nature of increased income inequality in the 1980s and 90s. We decompose unexpected changes in family income into transitory and permanent, and idiosyncratic and aggregate components, and estimate the contribution of each component to total inequality. The model we use is a linearized incomplete markets model, enriched to incorporate risk-sharing while maintaining tractability. Our estimates suggest that taking risk sharing into account is important for the model fit; that the increase in inequality in the 1980s was mainly permanent; and that inequality is driven almost entirely by idiosyncratic income risk. In addition we find no evidence for cyclical behavior of consumption risk, casting doubt on Constantinides and Duffie's (1995) explanation for the equity premium puzzle.

First draft: July 2002
This version: October 2004 [download pdf]

Presented at:

Society for Economic Dynamics, annual conference, 28-30 June 2002 in New York

Graduate Student Macro/International Workshop, 11 March 2003 at Princeton

Inter-University Student Conference, 11 April at Yale

Midwest Macroeconomics Conference, 16-17 May 2003 at the Federal Reserve Bank of Chicago

European Economic Association, 18th Annual Congress, 20-24 August 2003 in Stockholm


Thijs van Rens | IDEAS/RePEc | Google Scholar | ResearchGate