Abstract. As a matter of fact, the consensus of empirical evidence on the wage-price causal relationship reveals that there are two opposing groups of economists supporting conflicting hypotheses with respect to the flow of causality. Equally importantly, the literature review suggests that there is at least some consensus in the fact that inferences of researchers depend on the sample length, type of data employed, applied econometric model, or at the same time the relationship may be subject to the dynamics of economic cycles. In the light of these arguments, this paper conducts an empirical investigation of the wage-price causal relationship in UK by utilizing VAR and VECM models. Prior to designing and estimating econometric models the relevant stationarity tests for the wage, price and productivity variables have been performed. In summary, all three time series data are non-stationary and thus need to be differenced once in order to render them stationary. Correspondingly, the relevant cointegration tests have been performed and they provide robust evidence in support of a strong cointegration relationship between wages and prices. Additionally, respective restrictions on the parameters of estimated models have also been applied in order to derive the most parsimonious model. Regardless of the fact that VAR models tentatively indicate unilateral causality running from prices to wages, the VECM analysis only suggests a strong cointegration relationship and negates short-run causality in any direction. Accordingly, the estimated models have been subjected to diagnostic testing procedures, and these tests firmly indicate that estimated results are statistically robust. At the same time the restricted VECM model provides estimate on the basis of which it can be argued that the assumption of rational expectations in the wageprice relationship is perfectly valid.

Keywords: Granger Causality, Cointegration, VAR, VECM.