Due to the fact that many reliable indicators of further inflationary pressures do not seem to work anymore, finding whether or not wages Granger because prices is an important concern for policymaking. However, international evidence on the relationship between wages and prices does not show strong evidence in favor of causation in the direction of prices. The results presented here for Colombian data point to the same direction. This paper differs from previous ones published in Colombia in two aspects. First, we include the Unit Labor Cost (productivity adjusted wages) as a more sensible measure of wages. Second, we base our analysis on a price markup expectation augmented Phillips curve in which we include indicators of aggregate demand and supply shocks, thus avoiding omitted variables bias in our inferences. We worked under alternative stationary/ non stationary VAR models. We found evidence in favor of Granger causality from prices to wages but no evidence of Granger causality in the direction of prices. These results hold only when unit labor cost is used as the wage indicator and under alternative measures of aggregate demand and under different assumptions on the integration properties of the series. The policy implication of these results points to the very careful use of wages as leading indicator of inflation.
