Wage changes in a sample (1979-86) of Canadian wage contracts are analyzed. The sample is split into cost-of-living allowance (COLA) and non-COLA contracts. Econometric estimation corrects for sample selectivity bias. In non-COLA contracts, the unemployment rate and a proxy for expected inflation are significant. Expected industry price changes and productivity changes exert smaller but significant effects. In COLA contracts, ex ante inflation coverage ranges between 60% and 100% of expected inflation. Catch-up for previous uncompensated inflation, the unemployment rate, and expected inflation are also significant. A common wage structure across COLA and non-COLA contracts is rejected by the data.

This paper presents an empirical analysis of the determinants of negotiated wage changes using a pooled sample of time series observations for fourteen Canadian manufacturing industries. Data on individual contracts are used in an attempt to allow for the discontinuity of wage adjustments given the predominance of collective bargaining and variable contract length. Using a nonlinear formulation, profit levels, the unemployment rate, the rate of change of prices, and other variables are found to be statistically significant. DURING THE PAST decade a considerable amount of econometric research has been devoted to the explanation of movements in wages. Most empirical studies have used a basic disequilibrium model, first suggested by Phillips [11] in which the change in money wage rates is related to the level of unemployment. By relaxing some of the more rigid theoretical assumptions, the basic Phillips curve explanation has been expanded to include a number of other variables such as profits, prices, productivity, employment mix, etc. While many of these studies have provided valuable insights into the wage determining process, the statistical approaches used have often failed to deal adequately with the institutional features of the labor market. These statistical problems are briefly discussed in Section 1 of the paper and our own empirical analysis using data on individual contracts in Canada is presented in Section 2. The main implications of our study for the aggregate Phillips curve are given in Section 3.

Louis N. Christofides, Robert Swidinsky, David A. Wilton, A Microeconometric Analysis of Spillovers within the Canadian Wage Determination Process, The Review of Economics and Statistics, Vol. 62, No. 2 (May, 1980), pp. 213-221