A Re-Examination of the Permanent Income Hypothesis Using Cross-Section Canadian Data

Abstract

Since Keynes made his break with the classical theory, the literature on the consumption function has been dominated by two contending traditions. The first, derived from Keynes (1936), views consumption as a function of current income; the second, derived from Friedman (1957), views consumption as a function of wealth or permanent income. The purpose of this article is to re-examine these alternative theories, which have different theoretical foundations as well as different policy implications. The reason for this reexamination is that tests of the permanent income hypothesis (PIH) conducted in the late 1950s and early 1960s were based on data which were 'largely inadequate and/or irrelevant' (Modigliani and Ando, 1957, 123). The two sets of data most widely used, the Bureau of Labor Statistics/Wharton School Survey (1957) and the Oxford Savings Survey (Klein, 1955), were both found to be unreliable in many respects. It was felt that a new set of tests using a new set of data was in order.