Macroeconomics and the Phillips Curve Myth

Abstract

The story of the history of the Phillips curve up to the 1980s is reconsidered. It is argued that none of the principal components of conventional stories is correct. Phillips did not discover the negative relation of wage change (or inflation) and unemployment—that was well known long before him, and better analysed by several of his contemporaries. Samuelson and Solow did not present or advocate a ‘tradeoff interpretation’ of the curve. Almost none of the econometrics of the 1960s concerned exploration or refinement of the relation, and practically no one from that period saw the Phillips curve or any similar relation as offering an inflation-unemployment tradeoff. The ‘expectations argument’ conventionally attributed to Friedman and Phelps and dated to the middle or late 1960s was commonplace well before then. The Phillips curve econometrics of the 1970s was not mainly concerned with assessing that idea. Almost no one saw any basis for accepting more than minimal inflation, and policymakers were amongst the least susceptible to those arguments. The question of how the Phillips curve story came to be told is considered, and the answer found in a variety of factors, including simple confusion about terminology, as well as some retelling of history motivated by anti-Keynesian sentiment; but also in aspects of the poor presentation of the Keynesian position by some of its advocates. A rewriting of the history of thought is offered: it is suggested that the development of thinking in this does not conform to common ideas about scientific progress, and that economists of the early post-war period were never so foolish as they have been regarded.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 125 citations exist upstream.