New Considerations on Empirical Analysis of the Determinants of Canadian Provincial Government Health Expenditures, 1966-1998
Government of Canada, Health Canada, Working Paper No. 02-06
41 Pages Posted: 20 Sep 2007 Last revised: 24 May 2010
Date Written: January 2003
This paper uses data on real per capita total income, the proportion of the population 65 years of age or over, and the ratio of the deficit or surplus to gross domestic product (GDP) for the 10 Canadian provinces for the period 1966 to 1998 to study how these factors impact the Canadian provincial governments' real per capita health expenditures. Stationarity and cointegration issues are addressed, leading us to investigate a cross-section time-series model with a time trend, heteroskedasticity and autocorrelation in the residual terms. An income elasticity of 0.88 is found, suggesting that health care is not a luxury good under the assumption of constant prices. However, taking into account the Baumol effect would invalidate this hypothesis, and the estimated income elasticity could be an overestimate or an underestimate of the underlying true income effect, depending on the relative price of health care. For the ratio of the fiscal deficit or surplus to GDP, a significant coefficient of 1.45 is obtained, but the coefficient for the proportion of the population 65 years of age or over is non-significant.
Keywords: Health expenditures, Canadian provinces, Stationarity, Cointegration, Panel data, Income elasticity, Baumol effect
JEL Classification: I10, C22, C23, E30
Suggested Citation: Suggested Citation