GOVERNMENT ACTIVITIES AND TESTS OF THE LONG-RUN IMPLICATIONS OF THE NEOCLASSICAL GROWTH-MODEL FOR CANADA

This paper shows that the conclusions of time-series analyses of the l ong-run properties of the one-factor neoclassical stochastic growth mo del under uncertainty are robust with regard to alternative ways of ha ndling government expenditures. In particular, when annual Canadian da ta from 1946 to 1993 are used, the hypothesis that the log ratios of c onsumption to output, and investment to output are stationary is rejec ted even when government expenditures are treated as perfect substitut es for private ones (and aggregated with the corresponding private var iables).