We are currently in a very interesting period of time for the Mexico-Canada-US economic relationships. NAFTA has been in place for over a decade and has worked its effects on the economies of the three participants, and the FTA was implemented in 1989. Monetary integration between Canada and the US is proposed by some but is seen by many as a step that is fraught with negative consequences for Canada and is of little consequence for the US. It will be many years before Mexico will be included even in speculation about monetary integration. The much noted «bicycle theory» of economic integration would suggest that some new initiative must be introduced soon or the default forces of protectionism will begin their assault on what has been achieved thus far, but the time may not be ripe for large scale initiatives. These latter forces are very real and are espoused in high places, especially in Washington, and many Mexicans have strong reservations about the net benefits to their economy from trade liberalization. So, while Canadians have become less protectionist in the past few years, the issue of a halting or reversal of trade and financial market liberalization is far more than a think-tank fantasy.

In Ottawa, Paul Martin has replaced Jean Chrétien and has pledged that he will try to achieve something positive in Canada’s economic relations with the US soon in his term as prime minister. In Washington, the Bush administration is focused primarily on establishing the Republican Party as the dominant political force for the foreseeable future and, concomitantly, on electoral success in 2004. International economic policy is clearly subordinate in the US to domestic political party strategy. Mexico’s President Fox has been disappointed in the fact that President Bush has ignored  exico’s concerns and his electorate is becoming disenchanted by the lack of progress in the Fox agenda of reform. This suggests that the next few years will not bring much significant change in the economic relationship. 

Nonetheless, it is useful to review the Mexico-Canada-US economic relationship at this moment and to speculate on the possibility that some major development will take place in one or both of the two aspects of that relationship. Specifically, is it likely that Mexico, Canada and the US will resolve their outstanding trade dispute areas and that they will move beyond NAFTA to implement a customs union or some form of NAFTA- or FTA-plus? This issue is a contentious one and it has been discussed extensively by economists and political leaders for many years. In this paper we will examine the consequences to date of liberalization of international trade flows, and speculate on the likelihood of further development in the near  future.

Before discussing the likelihood and consequences of a customs union, it will be instructive to review the impacts on the Mexican and Canadian economies of the existing free trade agreement. Specifically, we will be interested in the impacts of NAFTA on Mexico’s, and FTA on Canada’s, trade and investment relations with the US and the rest of the global economy. It goes without saying that the dominance of the US in these flows has generated considerable comment in both Mexico and Canada and one would want to be clear as to the impacts of past actions before proceeding with additional ones. We will also be interested in the impact of the FTA on the structure of production in these two economies – what is the impact on the mix of output between manufacturing and raw materials? The US accounts for abut 80 per cent of Mexico’s and Canada’s international trade, so the impacts of trade liberalization on these two economies will be powerful. However, Canada accounts for only 20 per cent of US trade and Mexico’s share is only 12 per cent, hence since many goods were traded without tariffs prior to the agreements the impacts of NAFTA on the US economy have been negligible. In this short paper only the impacts on Mexico and Canada will be explored. 

CHANGES IN FLOWS OF GOODS AND SERVICES AND INVESTMENT

The central element in a country’s economic relations with the rest of the world are, of course, its trade and investment flows. When the opening of markets is the dominant force, two different things should be observed: first, the ratio of these flows to gross domestic product should increase, and, second, this increase should be noted in the country’s linkages with several parts of the global economy. That is, exports and imports of goods and services and incoming and outgoing foreign direct investment should be increasing with the major industrialized and third world economies. Tables 1 and 2 make clear the distinctive pattern that has marked the experience of the Canadian and Mexican economies with the FTA during the past two decades.

With reference to the first of the two criteria, exports and imports of both goods and services have increased as percentages of GDP in both Canada since 1988, the year prior to the FTA, and Mexico since 1992, the year of NAFTA. Thus if one stopped with the first criterion it would be safe to conclude that Mexico and Canada were experiencing the anticipated and desired consequences of liberalization of goods and services. But when we look at data for the second criterion, both exports and imports of goods and of services suggest that the North American experience has been one of regionalization rather than of globalization. Canada’s exports have gone decreasingly to the EU, to Japan and to the rest of the world, while the US share has risen significantly both before the FTA (1989) and after. The growth of imports from the US is less dramatic but the growth rate nonetheless is positive. Imports from both the EU and Japan rose prior to the FTA but diminished considerably thereafter. Canada’s imports from Others have exhibited more short-term fluctuation than long-term trend. A similar pattern is found for Mexico: the US share of exports has risen significantly beginning with 1992, the year of NAFTA, while little change in imports has occurred. What is most interesting with Mexico’s imports is that the US share has declined in recent years, while that of Asia has increased. This effect is probably indicative of the importance of imports of Asianmade components of motor vehicles made by US plants in Mexico. The operation of these plants should supplant imports from the US with those from Asia and increase Mexico’s exports to the US.

It should also be noted that both Canada and Mexico have had positive impacts on their balance of trade with the US. The two graphs that follow show that for both countries the Balance of Trade with the US turned dramatically to surplus following the two trade agreements. It is also the case that Mexico has experienced a much more positive BoT with Canada. This latter flow is of such minor importance that there is little real consequence of this development.

Graph 1

Graph 2

In Table 3 a somewhat similar pattern is shown by the stocks foreign direct investment in Canada. Between 1990 and 2001 the US grew slightly as a source of direct investment, the EU showed no trend, but the UK, Japan and Others all declined. It is only with Canadian DI abroad that the globalization pattern is exhibited, albeit not as one would have anticipated. Here the US share shows a downward trend throughout the period, but the UK and Japan also show declines, and once again the EU shows no movement. However, Canada’s DI in the rest of the world expanded significantly. For foreign direct investment in Mexico the figures are annual inflows rather than stocks, due to the availability of data. Neither the EU nor Japan shows much trend as a supplier of direct investment. Canadian FDI in Mexico shows a significant increase since NAFTA, although the absolute amount of the flows is quite small. It is the US that exhibits the dramatic increase in FDI in Mexico, again following implementation of NAFTA. 

Thus we note that for the trade and direct investment indicators for Mexico and Canada some pattern other than that of multilateral expansion is evident. While one might be tempted to refer to this as “de-globalization”, in the North American context the customary term is “continentalism” or “continentalization”. Both Canada and Mexico have long been concerned about the nature of the structure of North American economic and political relations and the relationships of power that are inherent in them, but since this is a Canadian Studies conference I will focus my remarks on Canadian commentators. Canadian discussion goes back at least as far as the late 19th century when Goldwin Smith and Erastus Winman argued for a continental entity based on a ‘natural’ north-south economic orientation rather than the ‘artificial’ east-west imperial one and a cultural commonality between two peoples who had shared a common experience of ‘conquering’ the continent. The process of continentalization became central to the discourse during the 1970s and 1980s when writers from Kari Levitt and Mel Watkins on the left to George Grant and poet Dennis Lee on the right bemoaned the perceived or imminent loss of Canadian sovereign control over the economy, corporate governance, culture, the political process and so forth consequent of the liberalization of trade and investment relations between Canada and the US. The first step was the Canada-US Auto Pact in 1965 which gave birth both to intra-industry economies and other efficiency gains, and to a sizeable industry that was devoted to proselytizing either for or against further liberalization. When Prime Minister Mulroney and President Reagan agreed in Quebec City in 1986 to pursue a free trade agreement, critics felt their concerns had been justified, and saw the FTA as ‘the first step on a slippery slope.’ The FTA would eventually be followed, they argued, by a customs union, cross border harmonization and standardization in the manner of the European Economic Community, perhaps a monetary union and then transformation of Canada into the 51st state or the 13th Federal Reserve district, depending on whether one’s preference was for political or economic metaphors. Some of these concerns and forecasts struck many as being a bit fanciful at the time, but today we are beginning to have serious discussion of both movement from a free trade area to a customs union and adoption of some form of a common currency.

One of the concerns of these critics is particularly relevant to the current topic. It was argued that liberalization of trade with the US would stimulate Canada’s movement into the position of being “a hewer of wood and a drawer of water”, that is that Canada would increasingly specialize in and export raw materials and would import from the US manufactured goods. There are legitimate concerns if such a transformation should in fact take place. Raw materials exports are less advantageous to a national economy than are finished manufactures with regard to income elasticity of demand, trends in price movements, market access, productivity increases, and price fluctuations. Thus, if trade liberalization were to stimulate raw material production at the cost of manufactured goods a legitimate objection to freer trade would be rational. A very summary examination of the composition of Canada’s exports and imports is given in Graph 4. It is apparent that between 1989 and 2001 Canada did not become a supplier of raw materials. In fact, although there is no discernable movement over the period in the composition of Canada’s imports, there is a reversal in the composition of exports. Exports of raw materials declined from 53.9 per cent of the total to 43.2 per cent, while manufactured goods exports rose from 47.9 to 56.1 per cent. In Mexico there has been a similar concern voiced, but data in Graph 5 indicate that, as with Canada, it is exports of manufactured goods that have shown the most striking increase in the composition of Mexico’s trade – following NAFTA in 1992. As is befitting in a relationship in with intra-industry trade is so significant, imports of manufactured goods have also shown an increase. These latter imports are, as noted above, that concurrent with imports from Asia of components for manufactured goods. Thus it would appear neither Canada nor Mexico has become a supplier of primary goods to US manufactures and an importer of finished products and, if anything, both have seen their manufacturing sectors enhanced by their trade agreements, primarily with the US. 

Graph 5

The effects on trade we have just examined, while positive, should not be taken as proxies for the entire economic situation that has developed as a consequence of NAFTA and the Canada-US Free Trade Agreement. Of at least as great importance are impacts on employment, the structure of employment, incomes and income distribution, the state of the agriculture and service sectors, and a host of other factors. While important, and examination of them would be tangential to the primary focus of this paper. Thus it is still quite possible for one to conclude that on balance both Canadians and Mexicans are correct in their opposition to these agreements and to further developments along these lines.

Following Benedict Anderson, we can appreciate that national borders exist in the mind as well as on the map. His conceptualization of the nation as “an imagined community” is crucial to our understanding of the rationality of the Canada-United States border and of the existence of two distinct nations. (Anderson, 1983). The economic integration process, beginning with the Canada-US Auto Pact of the mid-1960s was by its adherents supposed to bring us a ‘seamless border;’ the power of Anderson’s approach is made evident when we realize that today, after 40 years of trade liberalization we have less a seamless border than we do a ‘borderless seam.’ That is to say, the bi-national border may have been reduced dramatically (at least until September 11) but there is still the border in the mind that affects our behavior. (Goff, 2000, pp. 340-350). 

This can be seen in the recent work that has been done on the economic significance of the Canada-US border. John McCallum examined the impact of the border on shipments of goods between Canadian provinces and between the two countries. He used a gravity model in which trade is taken to be a function of the GDP of the two countries and distance, and his conclusion was that the national border matters a great deal. One example is trade between Ontario and Quebec, and both California and British Columbia. The distances are roughly equal but the GDP of California is ten times that of British Columbia; therefore one would assume that Ontario and Quebec would ship to California ten times what would be shipped to BC. In fact, interprovincial trade with BC was three times the international trade with California. McCallum did this exercise in two studies eight years apart and his conclusion is that, other things equal, trade between two provinces was 20 times that between a province and a state, using data for 1988, and 12 times in 1996. (McCallum, 1995, p. 616; 2000, p. 3). Charles Engel and John H. Rogers examined the same question but used a different methodology. In a one market economic space prices of individual goods will be the same. Distance will allow price differentials to develop but they conclude that the national border can have the same effect. Specifically, Engel and Rogers state that the Canada-US border has an impact on price differentials that is the equivalent to 1,780 miles of distance. (Engel and Rogers, 1996, p. 1120). Engel and Rogers’ study uses data for 1978-1994 and they suggest that there was a “slight tendency” for the border effect to be larger in the post 1989 period. This is a much weaker conclusion than is that of the two studies of McCallum which indicate that a significant change has occurred between the year prior to the FTA and several years thereafter. The effect of the border is a function of a variety of factors, such as local lending practices by financial institutions, cultural similarity/ dissimilarity, personal and social affinities, imperfections in access to information, misperception of the ‘other,’ tax policies, and biases in aspects of the communication and transportation infrastructure.

I am not aware of any similar work that has been done on the Mexico-US border effectbut, given that the factors that have just been mentioned are even stronger here that on the Canada-US border, the effect is certainly not weaker and is probably significantly stronger.