Explaining the Evolution of Government Size in the Canadian Provinces

Using time-series and time-series cross-section data from the 10 Canadian provinces, the authors test several competing explanations of government growth. Multivariate analyses reveal that the results are sensitive to how government size is measured. The authors find strong empirical support for the intergovernmental grants, the bureau voting, and the electoral budget cycle explanations irrespective of the measure of government size used. The party control explanation, the fiscal supply explanation, and part of Wagner's law give superior results when they are evaluated using real (adjusted) measures of government size than when they are evaluated using nominal (unadjusted) measures. The authors find virtually no support for the personal income part of Wagner's law.