Is aggregate consumer borrowing consistent with the permanent income hypothesis?

Using US data covering from 1959 to 1994, we examine the consistency of aggregate consumer borrowing with the permanent income/life-cycle hypothesis (PI/LCH) and the predictive power of consumer borrowing, The PI/LCH implies that consumer borrowing should be an increasing function of the gap between permanent and current income, In addition, if consumers accurately estimate permanent income, large borrowing should be associated with rapid income growth in the future, Our empirical results support the PI/LCH; consumer borrowing increases with the estimate of permanent income and decreases with current income, The predictive power of consumer borrowing, however, is marginal; lagged consumer borrowing explains only a small portion of income growth and does not Granger-cause income growth.