Co-integration, causality and Wagner's law: tests for selected Caribbean countries

In this paper, six versions of Wagner's Law were empirically tested employing aggregate annual time-series data on nine Caribbean countries. The results indicate that a long-run equilibrium relationship between income and government expenditure does not exist for the countries studied, with the exceptions of Grenada, Guyana and Jamaica for a particular formulation of Wagner's Law. However, the direction of causality runs from income to government expenditure only for Guyana, while for the other two, the causality runs in the other direction. Results for short-run causality are mixed, but the predominant causal relationship appears to run from income to government expenditure. In light of the empirical results in this paper, one may tentatively conclude that Wagner's Law finds broad support in these islands. These results run counter to what has been previously reported for a subset of the islands studied in this paper.