Testing Keynesian Unemployment Theory against Structuralist Theory: Global Evidence of the Past Two Decades

Abstract

The theory of unemployment accorded the status of orthodoxy in Western macroeconomics — the theory expounded in the textbooks — is, in a word, Keynesian. It derives from (among others) Keynes, Hicks, Tobin, Patinkin, the ‘natural rate’, and either a ‘new microeconomics’ apparatus without rational expectations or a New Keynesian apparatus compatible with rational expectations in order to generate inappropriate responses or inertia in nominal wages or prices. Some would allow hysteresis making the natural rate of unemployment ‘path-dependent’. The monetarists use the same theory, though with different emphases, and in a sense the New Classical theory is a special (and more fully elaborated) case of it. So entrenched has it become in conventional economic thinking that people who know no economics at all have learned, like Samuelson’s parrot, to say ‘weaker demand’ or ‘supply shock’ with every slump no matter how prolonged it is or how obscure its source or sources.

This paper was written without complete access to my own files so the report of the statistical results was left somewhat imprecise but, I think, not inaccurate. My thanks to Alan Malz for computations and to a number of economists for ‘leads’ in the search for evidence, including Annalisa Cristini, Dennis Snower, and Giuseppe Tullio. A grant from the US National Science Foundation is gratefully acknowledged.