January 1, 2004 will mark the fifteenth anniversary of the implementation of the Canada-US Free Trade Agreement (FTA), which largely formed the basis of the North American Free Trade Agreement (NAFTA), including Mexico as a partner, that came into effect five years later.
Both agreements were seen as epochal, and the subject of intense scrutiny and debates in Canada (at least with respect to the FTA), and the United States and Mexico (at least with respect to the NAFTA). In retrospect, they also must be put in context of the broader trend toward the increased liberalization of global merchandise trade, investment, services and peoples’ flows after the second world war, as well as technological changes affecting transportation and communications. These trends toward more deeply interconnected world economies called for new rules, beyond simply those affecting trade in goods at the border, that had been the mainstay of the rounds of negotiations on the multilateral General Agreement on Tariffs and Trade until the 1970s. In some ways the FTA and NAFTA served as role models for the subsequent agreement on the Uruguay Round of multilateral trade negotiations.
Given that various proposals have recently surfaced to facilitate even further the exchanges between the US and Canadian economies, it may be useful to look back on what we know about how the FTA and the NAFTA did or did not work and about economic integration in North America more generally, and to ask what questions remain or what lessons can be drawn as we look to the future. To that effect, this brief text reviews some of the recent analysis and data relevant to the evaluation of how the FTA and the NAFTA have affected Canada, and attempts to draw some conclusions relevant to the future of Canada’s economic position in North America.
Economic integration is typically analyzed by looking at four broad types of cross-border flows : trade in goods, trade in services, investments and people, and indeed this distinction is useful in terms of examining the breadth of North American integration across a range of activities. I summarize the evolution of these four variables between Canada and the other two NAFTA countries in Table 1 below. The table compares data for the years 1988 (the year before the FTA came into effect), 1993 (the year before NAFTA came into effect), and 2002. For investment, I show the accumulated stock of foreign direct investment (FDI, as distinct from portfolio investments) in Canada by NAFTA nationals, and by Canadians in other NAFTA countries. A sub-category of services transactions called “commercial” services, that exclude travel, transportation (which often reflects trends in goods trade) and government-provided services, is also shown. Commercial services include, for example, cross-border flows of royalties and licence fees, financial services, management services, and architectural, engineering and other professional services.
These data are, in a first block of rows in the table, expressed as a share of Canada’s GDP, except the data on temporary worker movement (only shown for 1993 and 2002), expressed as a share of Canada’s total employment. The picture that emerges from these lines is unambiguous : all the key trade and investment cross-border flows with NAFTA countries, as well as the temporary presence of Canadians working in other NAFTA countries, have assumed a much bigger importance relative to Canada’s economy since 1988.
These flows overwhelmingly represent transactions with the United States. Canada’s merchandise imports from Mexico, Canadian direct investment to Mexico, and two-way flow of temporary workers between Canada and Mexico have grown markedly faster than the corresponding transactions between Canada and the United States since 1988, but they did so from a very small base, and hence they have slightly reduced but not challenged the dominance of U.S. flows for Canada as far as intra-NAFTA relations are concerned.
These intra-NAFTA data can also be expressed relative to similar transactions between Canada and all countries, and I have done this for trade and FDI flows in a second set of rows in the table. Of note here is how much more subdued, when not in fact submerged, the just noted increased importance of intra- NAFTA transactions becomes, when compared to the increasing flows that Canada has also experienced with the rest of the world (note however, that 2001 and 2002 were years of exceedingly slow growth for the United States economy, and also ushered in noticeable increases in the costs of transacting across the US border due to the fallout from the September 11, 2001 terrorist attacks).
In other words, while our interdependence with NAFTA increased across all key indicators, Canada’s relationships with the rest of the world in some cases grew even faster, notably with respect to flows of Canadian direct investment abroad. Furthermore, even though Canada’s goods trade with its NAFTA partners rose much faster than that with the rest of the world, Canada has not actually increased its share of the total US import market during the period.
It could be tempting to conclude on that basis that North American free trade has had little overall impact, and that most of the booming trade is due entirely to the booming American economy for much of the period. But this would not likely be the correct conclusio n. We cannot expect North American integration to proceed in isolation from other rapidly evolving global trends and events, and the broad numbers may hide significant shifts on the ground. For example, while Canada has not gained share in total United States imports, it has in fact done quite well, relative to many traditional competitors, since the FTA. If one excludes the emergence of Mexico on the US market since the NAFTA (and the 1994 peso devaluation), and the huge progress that a rapidly modernizing China is making in the world’s major markets (including Canada, where China has displaced Japan as the number two importer), Canadian producers have done well relative to the rest of the field (see table). Indeed, early signs clearly had shown that by 1995 Canada- US trade was booming specifically in those sectors that were liberalized by the FTA, and relative to trade with other countries, suggesting a positive impact on trade flows from the trade agreement.
This is not to say that subsequent competition from Mexico or China is to be taken lightly. On the contrary, it is important both to understand who the competition is, and to adjust our strategies accordingly, a point I will return to below.
A similar observation applies with respect to FDI. China and other large and promising markets have opened themselves up to foreign investors since the FTA came into effect. It is not suprising that these new and fast growing markets exert a profound attraction on global firms, including Canadian ones, leading to a neglect (but only a relative one) of markets that in any event are already well “mined” by foreign investors, such as Canada’s.
Another factor might be at play in the Canada-NAFTA investment relationship. Foreign direct investment used to be, in part, the means by which foreign firms could “jump” over a country’s tariffs and access its market. With tariffs between Canada and the United States falling to zero for the vast majority of products, some of the increased cross-border trade could have substituted for products that might have otherwise had to be made locally by a subsidiary. In other words, trade reduced the need for FDI as a means of accessing markets, which would be consistent with the increasingly standard finding that investment is now in large part a complement trade, not a substitute for it.
Having said this, there appears to have been a structural shift in Canadian investments in the United States toward services industries, rather than goods production. During the first year of the FTA, in 1989, 33% of Canadian FDI in the United States was in services. Today, that ratio is close to 65%. This shift (which seems to be continuing with recent announcements of substantial Canadian acquisitions of US firms in the insurance and retail sectors), may be a sign that many Canadian services companies are pursuing, through investments, the vertical integration achieved through trade by their colleagues in manufacturing. Indeed, Statistics Canada has shown that, relative to manufacturers, Canadian service providers tend to serve their foreign clients more through majority-owned affiliates in foreign countries (Marth 2003). In part, this is because the nature of certain services makes it difficult to provide them without having a significant physical presence in the market. Having said this, the US services market remains more open to Canadian services exports than most others around the world.
Clearly, vertical integration (countries specializing in various stages of production in a wide range of products, as opposed to each country specializing in different industries) has been more intensively pursued than specialization by industry between the NAFTA partners. Indeed, a key trend there has been the increased use of imports from each others as inputs into our exports to each other. Thus, the typical Canadian export was made up of 33% imported goods and services in 1999, up from 26% in 1988 (Cross 2002). This increased ratio (evident across resources, manufactured goods and commercial services exports), underlines the greater sensitivity of North American production to potential border disruptions, as well as the increasingly self-defeating nature of protectionist measures between countries. A similar phenomenon has sprung up from more open trade between Mexico and the United States.
The temporary movement of people for economic reasons has also experienced a significant leap during the period. Here again, one may speculate that these temporary exchanges of talent may well be the sign of more vertically integrated economies and are often supportive of Canadian exports or Canadian investments.
Greater integration has had a positive effect on Canada’s productivity performance. The importance of productivity (producing more or better goods and services with a given amount of resources) is hard to overstate as a key underpinning of well-being, regardless of whether one thinks that standards of living are sufficiently approximated by GDP (economic output) or not.
A number of micro -economic studies show that greater trade integration with the United States has been positive for Canada’s productivity performance. They suggest, simply, that free trade fulfilled its mission by rasing the productivity of Canada’s manufacturing industry beyond what it would have been otherwise. A by now classic reference is Trefler (2001). In the same vein, a more recent Statistics Canada study concludes that Canadian manufacturing plants that have increased their exposure to export markets accounted for three-quarter of manufacturing productivity growth in the 1990’s, even thought they accounted for less than 50 percent of manufacturing employment (Baldwin and Gu 2003).
While Canada’s manufacturing employment continued on its secular downward trend relative to total employment, its decline since the FTA was nothing like the decline in manufacturing employment that characterized the US economy during the same period. The establishment survey conducted by the US Bureau of Labor Statistics shows that US manufacturing employment now stands at less than twelve percent of total US employment, compared to seventeen percent in 1988. Canada appears, relatively, like a manufacturing powerhouse, with average weekly earnings in manufacturing (including overtime) rising by five percent in Canada since the free trade agreement, relative to earnings in other sectors of the economy. Furthermore, the structure of Canada’s goods exports under the FTA, measured in constant 1997 dollars, has shifted markedly away from the traditional resource base, and toward non-resource exports, now accounting for some fifty-five percent of all exports (and no longer absolutely dominated by automobiles).
To be sure, there are serious clouds on the horizon. As useful as free trade has been in helping Canada shed its image of hewer of wood and drawer of water, there are dangers in the position that Canada finds itself in. It is clear that the low Canadian dollar during much of the period has also been a very significant factor in sustaining the Canadian manufacturing success. The recent rise in the Canadian dollar is now testing the basis of this success. And in a way this will be a good thing, for as Mexico is finding out, low costs are not a sustainable basis for competitiveness in the face of emerging economies. If Canada is to sustain its penetration of the US market, it will have to be on the basis of innovative products and superior quality, a point to which I will return below.
Having said this, as is widely known, and in spite of the boost provided by free trade, Canada’s overall productivity performance did not keep pace with that of the United States. OECD statistics show that labour productivity in the business sector increased by 26.5% in the United States since 1988, or 1.7% per year, compared with 20.3%, or 1.3% per year in Canada (OECD 2003). This result runs explicitly against the hope that, thanks to free trade, Canada’s productivity could catch up to that of the United States.
What went amiss on that score, then? Recent studies point to two possible explanations: a rise in self-employment, and a lack of innovation. Rather than reaping huge economies of scale from free trade as some had predicted, Canada’s productivity initially increased through the shedding of workers in non- competitive firms (Gu, Sawchuk and Whewell 2003). To be sure, workers remaining in industries that modernized in the face of foreign competition, and the myriad consumers benefiting from lower prices and greater choice, were “winners” from free trade. But, as the table shows, this was also a period of high unemployment due not only to those initial losses to liberalized imports, but to other factors that made it hard for employment to gain traction at the time, such as the punishing level of the Canadian dollar (undermining export efforts in the short run) and the beginning of serious fiscal retrenchment in the public sector.
It now appears that this period coincided with an extraordinary mushrooming of self-employment, with the self-employed not having the level of productivity as those remaining in the incorporated business sector. This factor alone can account for most of the productivity differential between Canada and the United States (Baldwin and Chowhan, 2003).
Another, possibly complementary, explanation focuses on innovation. Dachraoui and Harchaoui (2003) found that, while there was a revival of Canadian productivity after 1992 relative to US levels, this revival was due to the progressive adoption of best-practice (existing) technology in Canada, presumably in the face of greater international competition. However, while Canada was copying best practices, the United States was able to keep its lead by pushing the innovative frontiers in new industries.
While many of the studies of the impact of free trade just mentioned have focused on manufacturing wages and employment, there is no doubt that services were also profoundly affected. This was not necessarily due to any liberalization that was contained in the FTA or NAFTA, since many trade and investment restrictions remained in the service industries under these two agreements. Rather, the need to serve customers who were themselves coming under greater competitive pressure provided the backdrop for changes in the Canadian services sectors. CN is the prime example of this, having reoriented its entire rail and service infastructure from being geared to East-West trade to being able to serve North- South trade. While CN did shed workers in the process, there is evidence that free trade favoured employment and earnings of workers in trade-related services industries – those services most linked to both export and import activities (Schwanen, 2003).
This exploration of the impacts of the FTA on the nature of North American integration and on Canada’s economy, while short and inherently incomplete, raises a number of questions moving forward.
Thanks to more open trade, Canada has improved and in many cases realized its potential to become a more productive economy overall, than would otherwise have been the case. Thanks in part to the low Canadian dollar relative to its US counterpart, it has certainly been competitive in the NAFTA. But, as I have sketched out, the world is moving rapidly, and simply holding the line on competitiveness is not sufficient.
Mexico provides a sobering object lesson in this respect. Quite apart from the difficulties that had been predicted and that have sadly come true with respect to Mexico’s newly liberalized agricultural sector, with attendant social and migration and environmental problems, the Mexican economy faces formidable emerging competitors. Yet Mexico has not fully taken advantage of the opportunities afforded by NAFTA to transform itself into a dynamic economy that could, among other things, provide better employment for displaced agricultural workers. The main lesson there is the role that domestic policies do or do not play in ensuring that economies can properly compete under conditions of mo re open transactions with the rest of the world. In the Mexican case, an energy sector struggling for capital under state ownership, and a poor tax structure and tax collection that in turn starve the public sector of much needed funds for infrastructure and education, seem to be key culprits.
The point made above concerning the productivity travails of the Canadian economy under free trade also suggest some lack of flexibility and innovation domestically, preventing the capture of potential benefits from the larger market. It is remarkable that innovation has been such a buzzword of public policy for so long, and that governments at all levels have produced countless studies and strategies aimed at improving Canada’s innovation performance, and that we still find ourselves with a major shortcoming on that score. Yet there are strong reasons to think that, the larger and richer the markets in which we operate, the greater the returns to innovation and, conversely, the lower the relative incomes of those who do not have the skill or talent to innovate (Dinopoulos and Segerstrom 1999). In that light, every policy area and major public investment, from health care to cultural policies to the subsidies being sought by the automobile indutry, should be reviewed to ensure that it does not stifle or discourage innovation. Similarly, the case is strong that some elements of Canada’s tax mix are out of synch with building a more innovative and productive economy in the context of open trade, notably our relatively strong reliance on income- based rather than consumption-based taxes (Note that this is not necessarily a question of the overall tax levels needed to fund public services).
The emphasis I have just put on bringing domestic policies more in synch with the more open economy does not mean that we should not address the many flaws and « unfinished business » of existing trade agreements. But it is important to take the time to do this right, because the experience of the FTA and the NAFTA also shows that partial liberalization has a cost, and indeed can corner participants into clearly sub-optimal situations. For example, the incomplete liberalization of transportation services is a major element preventing the full realization of the benefits of trade in goods (Brooks 2003). This is just one illustration of the fact that it makes less and less sense to discuss strictly goods trade liberalization, for example under a customs union, without also considering how we can improve the cross-border relationship more generally (for example, concerning dispute avoidance and settlement) or address complementary flows of services, investments and peoples.
Thus, while the NAFTA remains a major asset in Canada’s economic arsenal, the lessons from our own experience and elsewhere is that the world keeps moving, and that continued adaptability, as well as taking greater account of the complementarity of policies in various areas, is the name of the game.
Notes and sources for data in Table. Most data are from Statistics Canada CANSIM data base, and author’s calculations. Data on temporary migrants are from Papademetriou (2003); his data on entry into the US reflect admissions, and the same individual may thus be counted several times. In calculating NAFTA trade in commercial services for 1988 and 2002, I actually used Mexican data from 1989 and 2001, the only ones publicly available currently. Both these numbers are from Canada Department of Foreign Affairs and International Trade (2003).
* This note is a compendium of points made for various recent presentations by the author on the impact of North American free trade. It is intended to be updated and expanded from time to time, and that more formal subsequent versions will incorporate a more complete set of sources and references. In the meantime the author would be happy to provide on request specific sources for the factual statements made or attributed in this note. All other views expressed and analysis contained herein are the author’s responsibility. This version : December 23, 2003.
Baldwin, John and James Chowhan (2003). The Impact of self-employment on labour-productivity growth : A Canada and United States comparison. Ottawa: Statistics Canada, Economic Analysis Research Paper Series No. 016.
Baldwin, John and Wulong Gu (2003). Participation in Export Markets and Productivity Performance in Canadian Manufacturing. Ottawa: Statistics Canada, Economic Analysis Research Paper Series No. 011.
Brooks, Mary (2003). Mapping the New North America : A Transport Sector Perspective. Presentation before the Study Group on Mapping the New North American Reality, Pan American Partnership for Business Education, école des Hautes études commerciales (HEC) Montréal, Novembr 14, mimeo.
Canada Department of Foreign Affairs and International Trade (2003). NAFTA@10 : A Preliminary Report. Ottawa : Minister of Public Works and Government Services Canada.
Cross, P. (2002). “Cyclical implications of the rising import content in exports.” Canadian Economic Observer, December. Ottawa : Statistics Canada.
Dachraoui, Ka¿s and Tarek Harchaoui (2003). A Frontier Approach to Canada-U.S. Multifactor Productivity Performance. Ottawa: Statistics Canada, Economic Analysis Research Paper Series No. 010.
Dinopoulos, E. and P. Segerstrom (1999). “A Schumpeterian Model of Protection and Relative Wages.” American Economic Review 89(3) :450-472.
Gu, Wulong, Gary Sawchuk and Lori Whewell (2003). The effect of tariff reductions on firm size and firm turnover in Canadian manufacturing. Ottawa: Statistics Canada, Economic Analysis Research Paper Series No. 011.
Marth, Michael (2003). Foreign affiliate trade statistics – Canadian operations abroad, 1999 to 2001. Ottawa: Statistics Canada, Economic Analysis Research Paper Series No. 013.
OECD (2003). OECD Economic Outlook. Vol. 2003/1 No. 73, June. Paris : Organisation for Economic Co- operation and Development.
Papademetriou, Demetrios G. (2003). “The Shifting Expectations of Free Trade and Migration,” chapter in J. Audley et al., NAFTA’s Promise and Reality : Lessons from Mexico for the Hemisphere. Washington, D.C. : Carnegie Endowment for International Peace.
Schwanen, Daniel (2001). “Trade Liberalization and Inequality in Canada in the 1990s” in Keith Banting, Andrew Sharpe and France St-Hilaire, eds. The Review of Economic Performance and Social Progress : The Longest Decade : Canada in the 1990s, Montreal and Ottawa: Institute for Research on Public Policy and Centre for the Study of Living Standards.
Trefler, Dan (2001). “The Long and Short of the Canada-U.S. Free Trade Agreement.” NBER Working Paper No. 8293. National Bureau of Economic Research.