Multivariate cointegration and causality tests of Wagner's hypothesis: evidence from the UK

Abstract

The purpose of this paper is to examine the issue of omitted variables in testing the long run validity of Wagner's hypothesis. Using UK data for the period 1948 to 1997, this paper first investigates the secular relationship between public spending and economic development in a bivariate system. In all cases considered, our bivariate cointegration tests indicate the absence of a long run equilibrium condition. However, the introduction of a third variable (money supply) re-establishes a cointegrating relationship between public expenditure and economic development variables. In addition, the results of the Granger's multivariate causality test indicate a unidirectional causality from income and money supply to public spending in the long run, thus providing support for Wagner's hypothesis.