John Manley has remarked that success on the federal deficit has allowed Canada to make choices that will allow it to become a ”œ”˜northern tiger’ ”” a preferred des- tination for knowledge workers, trade and investment, and a center of excellence in innovation, science, research and education.” At the same time, Industry Canada has com- missioned two University of British Columbia professors to convene a conference and prepare a subsequent volume that will focus on Canada’s ability to attract the following ”œinternationally mobile factors”: foreign direct investment (FDI), R&D, and human capital. The data on at least two of these fronts, FDI and R&D, is disappointing.

Whereas in 1970 Canada’s inward FDI stock was four times its outward, today, outward FDI exceeds inward. As a result, although Canada has been able to maintain its share of the rap- idly growing stocks of world outward FDI, its share of world inward FDI stocks has been falling (Figure 1, page 33). These  patterns are especially of concern given the US economy has been maintaining its inward share of world FDI and decreasing its outward share (Figure 2, page 33). That is, the propensity (rel- ative to global trends) for US multinational enterprises (MNEs) to locate abroad has fallen, whereas that for Canadian MNEs to locate abroad has not. On the other side, the propensity for foreign firms to locate in Canada has fallen, whereas the propensity foreign MNEs to locate in the US has not.

In addition, Canada’s R&D performance has been chroni- cally weak, ranking far behind those of most of our trading part- ners. This poor R&D performance has been linked to both our productivity gap with the US economy and, until recently, the depreciating value of the Canadian dollar. But this is not a new development. Since the 1980s, Canada has been hobbled by the lowest R&D levels of any G7 country in terms of its gross expen- diture on research and development (GERD). A single company, Nortel Networks, has for many years been responsible for about one-fourth of all the private sector R&D in Canada.

Although there seems to be a con- sensus that policy makers must address the comparatively poor productivity and R&D performance of the Canadian economy, there is less of a consensus regarding both the cause and the effects of Canada’s changing FDI patterns. As a result, informed policy recommendations are not avail- able. This article discusses these pat- terns, both at the aggregate and industry levels, and argues that although Canada’s reduced attractive- ness to foreign investors should be of some concern, the surge in Canada’s outward FDI is a positive development for the Canadian economy.

The importance of FDI to the economy must be understood before policy recommendations can be for- mulated to address the changing FDI trends noted above. There is a rich dis- cussion of the costs and benefits of both inward and outward FDI to home and host countries.

First, both international trade and FDI are important channels of R&D diffusion. It is well established that a majority of the world’s formal R&D efforts are located in the OECD coun- tries, with about one-half in the United States. The immediate ques- tion that arises is, are the benefits of R&D as concentrated as the expendi- tures? The answer is clearly no. But how are R&D efforts in one country disseminated to other countries? In their seminal 1995 work, David Coe and Elhanan Helpman establish that international trade is an important channel through which R&D under- taken in one country is diffused to other countries. When a country imports goods and services from another country, it is also importing the embedded technology. The more a country imports from countries that undertake a lot of R&D, the more R&D will be embedded in the imports. By modeling international trade as the channel by which R&D in one country is transmitted to other countries, they establish that the R&D efforts of any domestic economy improves domestic productivity as well as the productivity of trading partners. In many instances, and especially for smaller economies, the impact of trading part- ners’ R&D efforts on domestic produc- tivity are larger than the domestic countries own R&D efforts on its own domestic productivity. That is, for these smaller economies, access to the R&D undertaken in other countries is often more important for domestic productivity growth than is the coun- try’s own R&D efforts.

In subsequent work, Walid Hejazi and A. Edward Safarian extend the Coe and Helpman framework of analy- sis to allow FDI in addition to interna- tional trade to be a channel by which R&D is diffused internationally. Using data for OECD countries, they show that FDI is the dominant channel of R&D diffusion ”” roughly two-thirds of R&D spillovers occur through FDI, and one-third through trade. My 2001 article in Policy Options confirms these results for Canada.

Second, the international business literature argues that multinational enterprises (MNEs) must be more pro- ductive than domestic firms in order to be able to overcome the costs of doing business in a foreign economy. The empirical evidence supports this notion ”” several Statistics Canada studies as well as Daniel Trefler’s 1999 Policy Options paper document the superior productivity performance of foreign firms operating in Canada. These higher productivity firms have positive effects on the local economy vis-aÌ€-vis R&D spillovers, as well as on many other dimensions.

Third, there is a great deal of evi- dence to indicate that inward FDI con- tributes to domestic capital formation. However, the impact of both Canadian outward and inward FDI on domestic capital formation depends on the source of the FDI. For example, FDI into Canada from the United States has a smaller impact on Canadian cap- ital formation than does FDI from out- side North America. The reasons for this relate to the role of intrafirm trade ”” American MNEs supply much of their Canadian affiliates with interme- diate inputs produced in the United States, hence relying less on Canadian suppliers. In contrast, MNEs from out- side of North America rely more heav- ily on local Canadian suppliers. As a result, there is a larger impact on Canadian production and hence capi- tal formation when the inward FDI is coming form outside North America.

The impact of Canadian outward FDI on the Canadian economy also depends on the role of intrafirm trade within Canadian MNEs. If a Canadian affiliate locates in the United States, there will be a larger reliance on inter- mediate inputs produced in Canada in comparison to a Canadian affiliate locating outside of North America. The data on intrafirm trade by Canadian MNEs clearly shows this to be the case. Therefore, Canadian FDI into the United States actually stimulates production and hence capital formation in Canada, whereas Canadian FDI into developing countries and Europe do not. In fact the evidence indicates that Canadian FDI locating in developing countries actually reduces Canadian capital formation. It latter result that may be of concern to policy makers, a point to which I return below.

Finally, there is an emerging con- sensus in the international economics and business literatures that there is a complementary relationship between international trade and FDI. That is, when Canadian FDI to any given country increases, Canada’s trade to that same country also increases. Having a multinational presence abroad often stimulates trade. It has been shown, for example, that FDI abroad has a beachhead effect in pro- moting subsequent domestic exports. Reasons for this include FDI abroad markets home products and home- made inputs, and MNE retailers are more likely to sell home products.

This complementary relationship between trade and FDI is consistent with estimates produced by the United Nations Conference on Trade and Development (UNCTAD), which indi- cate that the top 500 MNEs account for 90 percent of the world’s stock of FDI and well over one half of world trade. In other words, Canada’s trade and hence our linkages to the global economy are very much linked to the role of MNEs.

This discussion clearly indicates therefore that FDI is very important in many dimensions of the Canadian economy. As a result, the FDI trends that have emerged over the past two decades must be of concern to Canadian policy makers as well as the private sector. These changing FDI patterns have had significant impacts on the Canadian economy, and their effects may persist.

In my last Policy Options article (2001), I argued that Canada relies too heavily on acquired R&D and too little on its own efforts. I argued that such an approach is detrimental to the Canadian economy. These arguments are consis- tent with those of the Roger Martin, dean of the University of Toronto’s Rotman School of Management, who has argued eloquently that Canadian firms depend too much on replication strategies and not enough on being innovative. The industry minister, Allan Rock, has acknowledged that the federal government has done a poor job of pro- moting the use of homegrown technolo- gies within our own borders.

Clearly, Canada’s relatively poor productivity performance has reduced Canada’s attractiveness to foreign firms as a location to invest, and at the same time, less investment by foreign firms, which are more productive on average than domestic firms, has also reduced the growth in Canada’s productivity. In other words, the FDI trends described above are both a cause and a reflection of Canada’s R&D and productivity per- formance. Improving Canada’s produc- tivity performance will go a long way in improving Canada’s attractiveness as a location to invest. At the same time, improving Canada’s attractiveness to foreign firms will improve Canada’s productivity performance.

The natural question that arises from these FDI trends is, why are Canadian MNEs increasingly locating abroad and foreign MNEs locat- ing in Canada less often? These trends are reflected in Figure 1 above. Together with my col- league A. Edward Safarian at the Rotman School, I answer this question using separate models for trade and FDI ”” and doing the analysis simultaneously.

The comprehensive analysis takes account for traditional determinants of trade and FDI, such a GDP, GDP per capita, exchange rates, distance between countries, language and cultural similarities, and regional trade agreements. In addition, the analysis accounts for factor endowment differ- ences, R&D expenditures, financial mar- ket liquidity, government policies toward trade and investment, as well as measures of the quality of institutions. The period of coverage for this study is 1970 to 1998, and it covers Canada’s bilateral relations with 29 countries.

The results that are especially of interest include the following:

The inner workings of government
Keep track of who’s doing what to get federal policy made. In The Functionary.
The Functionary
Our newsletter about the public service. Nominated for a Digital Publishing Award.

  • The NAFTA has reduced the attractiveness of Canada as a destina- tion for foreign MNEs. Many non- North American firms now locate in the United States or Mexico, and have access to the entire North American market. This effect is very strong.

  • Countries that have high levels of R&D spending invest heavily in Canada. It is unclear, however, what effect Canada’s R&D has had on its outward FDI. What is clear, however, is that Canada’s outward FDI, on the margin, steered away from countries with high R&D intensities.

  • Canada’s outward FDI is increas- ingly being attracted to low wage locations.

  • The effects of the Foreign Investment Review Agency (FIRA), which was phased out in the mid- 1980s, have been significant and reduced Canada’s ability to attract FDI. The reputation effects of such policies persist for a long time.

In addition to changes in Canada’s FDI patterns as presented in Figure 1, there are also important changes in the distribution of Canada’s FDI bilaterally. Although the United States remains Canada’s dominant FDI partner, its importance has fallen dramatically on the outward side but less so on the inward side (Figure 3). In the mid-1980s, over 60 percent of Canada’s out- ward FDI was destined for the US. This has fallen to roughly 50 percent. On the inward side, there has been a steady though less dramatic decline in Canada’s share of FDI coming from the United States. In short, however, coun- tries outside North America and Europe are playing an increasingly important role in Canada’s FDI patterns.

Beneath the FDI trends discussed above are significant changes in its industrial distribution, especially on a bilateral basis. If we consider Table 1 (page 36). Panels A and B provide the distribution of Canada’s outward and inward FDI by industry for 1983 and 1995, respectively. Panel C provides the change in the distribution by invest- ment partner. We will focus our discus- sion here on Panel C. Over the period 1983 to 1995, the share of Canada’s outward FDI in the United States in Services increased by 19.2 percent, whereas the shares in natural resources and manufacturing fell by 11.1 percent and 8.1 percent, respectively. On the inward side, the share of inward FDI from the United States going to natural resources fell by 14.8 percent, whereas the shares going to manufacturing and services increased by 10.3 percent and 4.5 percent, respectively.

Canadian inward FDI is motivated by market access (services and nonser- vices), access to natural resources, and perhaps factor price differences, given the weak performance of the Canadian dollar over the period in question. It is clear from these data that natural resources are playing a decreasing role in attracting FDI into Canada. Given free trade between Canada and the United States, market access is becom- ing less important as a driver for American FDI in Canada, except of course in non tradeables (services), but even this did not grow rapidly over the 1983-95 period. In contrast, market access should play a major role in non- US FDI, although such MNEs can locate in the United States and export to Canada.

As indicated in Table 1, the share of inward FDI going to Canadian man- ufacturing has increased from all desti- nations considered, whereas the share in services is up only slightly. This may indicate that foreign MNEs are increas- ingly locating in Canada to produce manufactured goods, which in turn are used to supply both the Canadian and US markets. We also know the fol- lowing: foreign-controlled firms in Canada have twice the export orienta- tion as domestically controlled firms; foreign-controlled firms have higher productivity levels than domestic firms; and inward FDI complements capital formation regardless of the source of the inward FDI. Taken together, these data indicate that the positive impact manufacturing FDI is having on Canadian capital forma- tion, driven until recently by the low Canadian dollar and Canada’s revealed comparative advantage in some manu- facturing industries, has outweighed the negative effect on capital forma- tion that is the result of a reduced importance of natural resources.

On the outward side, the impact on Canadian capital formation is very much a function of the underlying motivation. Specifically, outward FDI to the United States results in an increase in Canadian capital formation, as much of the production in the United States by Canadian MNEs use Canadian inter- mediate inputs. On the other hand, Canadian production in developing countries has resulted in a reduction in Canadian capital formation. Canadian MNEs are moving much of their low- value-added production to countries in East Asia and Latin America.

A significant productivity gap has emerged between Canada and the United States. This gap can be attrib- uted to two ”œproduct innovating” industries: computers and machinery. In contrast, Canada has done well in ”œprocess-innovating” low-end manu- facturing industries. That is, Canadian industries have been able to cut costs more effectively than US manufactur- ing industries. Furthermore, these are exactly the industries that experienced the largest tariff reductions in the Canada-US Free Trade Agreement. It appears from these data, therefore, that Canadian firms are expanding their manufacturing facilities in developing countries to exploit their pro- ductivity advantages in such low- value-added industries.

These manufacturing investments by Canadian MNEs in developing countries are significant. Consider the following observations. First, Canada’s outward FDI is growing rapidly. Second, the share going to developing countries is increasing. Third, the share in manufacturing is up slightly (Table 1, Panel C). Therefore, these invest- ments represent a significant portion of Canadian outward investments. I argue, however, that these investments are good for the Canadian economy, despite their negative impact on Canada’s levels of capital formation.

As these low-value-added activities move to developing countries, resources in Canada are allowed to move to higher-value-added activities. Therefore, in the long run, incomes, productivity and hence Canadian liv- ing standards will increase as a result. Of course, higher R&D spending and domestic productivity are important factors in ensuring that as these resources move from low- to high- value-added activities; they in fact move into the most productive activi- ties, that is, ensuring that the new jobs that are created are good jobs. This fur- ther underscores the importance of improving Canada’s R&D and produc- tivity performance.

This leaves several issues of con- cern. First, why are Canadian firms gaining a productivity advantage in these low-end ”œprocess-innovating” industries rather than in the high-end ”œproduct-innovating” industries? Second, why are foreign firms, both US and international, increasingly locat- ing in the United States rather than in Canada? These are two trends that pose a particularly difficult challenge for Canadian policy makers.

I have argued that in order to draw policy implications from an analysis of Canada’s changing FDI trends, we must first understand what impact these changing FDI patterns have had on the Canadian economy. To the extent increased Canadian investments abroad have a ”œpositive” effect on the Canadian economy, then such invest- ments should be encouraged. On the other hand, to the extent the effects of such investments are negative, the underlying cause of the increased out- ward FDI must be understood in order to direct policy formulation.

Consider the following examples on the outward side. If Canadian MNEs are increasingly locating abroad for efficiency reasons such as access to unskilled labour, then such invest- ments should be seen as beneficial to the Canadian economy in the long run: domestic resources will move to higher-value-added industries as these low-value-added activities move abroad. On the other hand, to the extent Canadian MNEs are being driv- en to locate abroad ”” say due to a lack of skilled labour, high taxes, or a poor R&D environment in Canada ”” policy changes may be forthcoming to reme- dy such deficiencies.

On the inward side, Canada’s reduced attractiveness to foreign MNEs as a location to undertake production should be of concern. Given the signifi- cant benefits that are derived from such investments ”” including capital forma- tion, job creation, and more important- ly, R&D spillovers ”” policy makers need to consider ways to improve Canada’s attractiveness. It must be stressed that improving Canada’s R&D performance will improve its attractive- ness to foreign investors. At the same time, increased foreign investments in Canada will improve its productivity. The potential benefits of improving Canada’s attractiveness to foreign investors are tremendous, as have been the costs of our reduced attractiveness over the past decade.

You are welcome to republish this Policy Options article online or in print periodicals, under a Creative Commons/No Derivatives licence.

Creative Commons License