Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Is consumption more or less variable than predicted by the permanent-income hypothesis? To answer that question, the author develops a procedure based on a long-run restriction implied by the consumer's intertemporal budget constraints. In contrast to previous work, the approach here (1) does not require any assumptions on the stochastic properties of labor income, (2) does not impose restrictions on the consumer's information set, and (3) is robust to departures from the permanent-income-hypothesis model. The application of the procedure to postwar U.S. data suggests that consumption is smoother than the permanent-income-hypothesis model predicts. Copyright 1991 by American Economic Association.