er640104

SOME RECENT DEVELOPMENTS PHILLIPS CURVE ANALYSIS Thomas M. Humphrey

IN

Economists’ views of the Phillips curve concept have changed drastically in recent years. The original interpretation of the Phillips curve as a stable trade-off relationship between inflation and unemployment has given way to the view that no such trade-off esists for policymakers to esploit. As a result, some economists now argue that economic stabilization policies are incapable of influencing output and employment, even in the short-run. Instrumental to this change were several key developments in Phillips curve analysis, most notably the so-called natzrral rate and rational expectations hypotheses. The purpose of this article is to esplain these developments and their policy implications and to show how they altered economists perceptions of the Phillips curve. Accordingly, the first half of the article traces the evolution of Phillips curve analysis focusing particularly on the natural rate hypothesis. The second half concentrates on the rational expectations idea, currently the most hotly-debated aspect of Phillips curve analysis. Early Versions of the Phillips Curve Phillips curve analysis has evolved through at least five major stages since its inception in 19%. The first stage involved the formulation of a simple, stable trade-off relation between inflation and unemployment. The initial Phillips curve depicted a relationship between money wage changes and unemployment. But the assumption that product prices are set by applying a constant mark-up to unit labor costs permitted the Phillips relationship to be transformed into a pricechange equation of the form1 (1) where p is the overall excess markets-this inverse of the p=ax

cient espressing the numerical value of the trade-off between inflation and excess demand. This equation expresses the early view of the Phillips curve as;i stable, enduring trade-off permitting the authorities to 0l)tnin permanently lowel rates of unemployi~lcnt in eschange for l~ermnnently higher rates of inflation or vice-versa. Put differently. the equation was popularly interpreted as offerinflation-u~ien~l~loylnent 111g a menu of alternative coml~inntions from which the authorities could choose. Being stable. the menu never changed. Economists soon discovered, however, that the menu was not as staljle as originally thought ant1 that the Phillil~s curve had a tendency to shift over time.2 Accordingly. the equation was augmented wit11 additional variables to account for such movements. Introduction of Shift Variables The addition of shift variables to the trade-off equation marked the second stage of Phillips curve analysis. The inclusion of these variables meant that the Phillips equation could now be written as (2) P=ax+z

percentage rate of price inflation, x is demand in labor and hence product excess demand being proxied by the unemployment rate-and a is a coeffiterm contained is disregarded in in

where z is a vector of variables-

productivity, profits, trade union effects, unemployment dispersion and the like-capable of shifting the inflation-excess demand trade-off. Absent at this stage were variables representing price expectations. Although the past rate of price change was sometimes used as a shift variable, it was rarely interpreted as a proxy for anticipated inflation. Not until the late 1960’s were expectational variables fully incorporated into Phillips curve equations. By then, of course, inflationary expectations had become too prominent to ignore and many analysts were perceiving them as the dominant cause of observed shifts in the Phillips curve. The Expectations-Augmented Phillips Curve and the Adaptive-Expectations Mechanism Three innovations ushered in the next stage of Phillips curve 2 Indeed, Phillips himself in his 1958 article had recog-

1 For simplicity, the additive constant most empirical Phillips curve equations equation 1.

nized the possibility of such shifts. FEDERAL RESERVE BANK OF RICHMOND

15