An Empirical Investigation into Government Spending and Private Sector Behaviour

We examine whether there is a significant relationship between government and private consumption for Canada. We derive estimating equations between the two types of consumption under both cointegration and no-cointegration assumptions. This distinction seems to have been largely ignored in previous work in the literature. Our results suggest that this distinction is an important one. For government spending on goods and services, our pretests do not allow us to firmly conclude whether government and private consumption are cointegrated or not. Therefore, we estimate the relationship under both cointegration and no cointegration assumptions. Under the cointegration assumption we find the two types of consumption to be complements, whereas under the no-cointegration assumption we find them to be substitutes. For government investment spending we are unable to find any evidence consistent with cointegration. Under the maintained assumption of no cointegration we find no statistically significant relationship but an economic relationship that implies they are complements rather than substitutes.