A New Approach to the Equilibrium Labour Force

The time series, labour force participation equation has received considerable attention in the recent literature. This seems largely to be due to interest in the macro-policy implications of the equation and to the need for such an equation in the labour sector of econometric models. The major finding of this research is that the labour force varies inversely with the unemployment rate (the effect) (Tella, 1964, 1965; Dernburg and Strand, 1966; Barth, 1968; Bowen and Finegan, 1969; Vroman, 1970). The discouraged worker equation has also been used as a component of labour market models (Black and Kelejian, 1970; de Menil and Enzler, 1972), although there are exceptions (Fair, 1971; Lucas and Rapping, 1969; Wachter, 1972). The labour force-unemployment relationship is used to define and measure hidden unemployment, a variable which has played a role in money wage or Phillips curve equations (Simler and Tella, 1968; Gordon, 1969). Hidden unemployment, in turn, is the pivotal notion in the concept of a full-employment labour force, a variable which is included in most recent research on potential output (Okun, 1962; Kuh, 1966; Thurow and Taylor, 1966). Hidden unemployment has also figured in public policy discussion on the seriousness of the unemployment problem (Dernburg and Strand, 1966). In this paper alternative aggregate labour force participation equations are introduced. The models tested on quarterly data for the United States for the period 1948-70 provide a better explanation of the macro-data than does the discouraged worker model. The models presented in this paper have three main independent variables. The first variable reflects the rate of inflation and is included to allow for money illusion or uncertainty effects. The labour force is found to respond positively to this variable. The second variable measures the influence of unemployment on labour supply. This paper finds that the long-term unemployment rate (workers unemployed 15 weeks or more) provides a better theoretical and empirical variable than the traditionally used total unemployment rate. The final variable is a relative wage term (today's wage in relation to past wages) which may be a proxy for either a transitory wage effect or, alternatively, a standard of living effect. The evidence suggests that the latter influence prevails today. That is, a decrease in the current wage in relation to past wages causes workers to enter the labour force in order to maintain a 1 Many helpful comments were provided by Richard A. Easterlin, Benjamin M.