Time-series implications of the permanent income hypothesis on durable goods consumption

Mankiw's rational expectations-permanent income hypothesis (RE-PIH) model on consumer durable expenditures predicts that the change in durable expenditures should follow an MA (1) process. Empirical tests of this joint hypothesis, however, have been rejected using the postwar U.S. data. This thesis presents a time-series representation of the RE-PIH model that is capable of explaining the quarterly aggregated durable goods expenditure series. A novel feature of the analysis is that the observed infrequent purchases of durable goods by the consumers are incorporated into the model and that the time aggregation problem is explicitly addressed to investigate the aggregate dynamics of the durable expenditures. The time-series implication of the base model is that the change in durable expenditures is a function of the durable goods purchase interval. Mankiw's MA (1) model is shown to be a special case where the purchase interval is one quarter. Estimation results show that the RE-PIH model is capable of explaining the quarterly aggregate dynamics of the consumer durable goods expenditures once the infrequent microeconomic action is incorporated into the model and the time aggregation problem is explicitly taken into account.