Aggregate money demand functions in five industrial countries: Are they cointegrated?

In this paper we take issue with the claim made in some recent empirical studies that real money balances, real income and interest rates are cointegrated, or, alternatively, that velocity is a stationary variable, which is in contrast with the well known stylised facts about the behaviour of monetary aggregates in theUKand other industrial countries. We show that in fact this surprising result can be explained away in terms of statistical bias. It is only because in these studies inference is based on a mis-specified VAR that the null of no cointegration can be rejected - the standard result that money demand functions exhibit instability and that velocity is a non-stationary variable is confirmed when the analysis is carried out within a correctly specified system