Archived — Perspectives on North American Free Trade: Paper Number 5: Effects of the Canada-United States Free Trade Agreement on Interprovincial Trade
by John F. Helliwell, University of British Columbia, Frank C. Lee, Industry Canada, and Hans Messinger, Statistics Canada, April 1999
In this paper, two types of evidence are used to assess the impact of the FTA on interprovincial trade. First we use a gravity model to explain interprovincial and province-state total flows of merchandise trade from 1988 through 1996. The results show clearly that the FTA increased province-state trade relative to interprovincial trade. However, even after adjustment to the FTA, interprovincial trade intensities remain twelve times higher than those of trade between provinces and states, down from around eighteen to twenty times higher before the FTA was introduced. The same model was used to predict what the 1996 interprovincial trade flows would have been had the trade structure remained the same as in 1988, but taking account of economic growth in both countries. After adjusting for economic growth, and for the increasing share of services in GDP, the model predicted that 1996 interprovincial trade would have been 13 percent higher than it actually was. This 13 percent reduction in interprovincial trade is one estimate of the effects of the FTA on interprovincial trade, since it took place at the same time as the FTA was being introduced.
We then used disaggregated data recently prepared by Statistics Canada to estimate the extent to which tariff changes in Canada and the United States help to explain inter-industry differences in the growth of interprovincial and international trade. The results suggest that the FTA-related reductions in Canadian tariffs led to increases in imports from the United States and to reductions in interprovincial trade, while the reductions in U.S. tariffs led to increases in exports to the United States and to increases in interprovincial trade. For the 47 commodity classes studied, the net FTA-related reductions in interprovincial trade were estimated to be 7 percent, or about half the total shortfall calculated previously using aggregate data.
Finally, we compared the post-FTA industry-by-industry changes in trade between Canada and the United States with those that were predicted by the general equilibrium model used to assess the FTA in the late 1980s. The industrial pattern of trade increases is correlated with the predicted pattern, but the actual increases are much greater than predicted, even after accounting for general economic growth.
- Date modified: