Journal Article Labour Supply and Demand, Absenteeism, and Union Behaviourh Get access Ali M. Reza Ali M. Reza University of Pittsburgh Search for other works by this author on: Oxford Academic Google Scholar The Review of Economic Studies, Volume 42, Issue 2, April 1975, Pages 237–247, https://doi.org/10.2307/2296531 Published: 01 April 1975

JN a recent article Hall (1972) examined the relationship between wage and unemployment rates in twelve cities. He observed a positive relationship between the two: high hourly wages were paid in high unemployment rate cities, low hourly wages were paid in low unemployment rate cities. Moreover, this relationship, he argued, was a characteristic of equilibrium of the aggregate economy. The empirical results obtained by Hall were based on twelve observations for the year 1966. Since his model was designed to study the characteristics of equilibrium, the question may justifiably be raised as to whether the year 1966 represented an equilibrium state of the economy; might it not be possible that what Hall observed was a characteristic of disequilibrium instead of equilibrium? This is a particularly important issue, for Tobin (1972, p. 10) has tended to regard the aggregate economy to be in a state of perpetual disequilibrium. Furthermore, as Robert A. Gordon has pointed out, since Hall's results were obtained on the basis of only twelve observations, could not his results be reversed if some of the cities were excluded from his study?' Hall's conclusions were based on the results of the least squares regression of the wage rate on the unemployment rate, an appropriate procedure if the direction of causality runs from the latter to the former. Yet his theoretical analysis (correctly) implied that the two variables were jointly determined, an issue to which we will return in section II. But if the unemployment and wage rates are jointly determined then orthogonal regression should be used in order to determine, the quantitative relationship between them. The aim of this work is to re-examine and extend Hall's empirical results so as to determine the extent to which they are affected by his estimation technique, his selection of the cities and the year 1966. Although the necessary data for a direct, straightforward extension of Hall's work are not available, data do exist that permit us to look into these issues.