Report of the Advisory Panel on Efficiencies:- Chapter 3 Changes in the Canadian economy since 1969

In 1969, Canada’s population stood at 22.3 million and its gross domestic product (GDP) at $450 billion in today’s dollars. In 2005, there are 50 percent more Canadians than there were 35 years ago — 32.2 million — and Canada’s GDP is approximately three times more than it was then, at $1.36 trillion.

Many forces contributed to the evolution of the Canadian economy, forces that were indeed active in all advanced economies. The most important ones were globalization and technological progress and, in particular, the explosion in the use of information and communications technology, which has significantly changed how business is done. Also significant was the changing role of government and government policies, such as economic regulations and trade policy. Moreover, services steadily increased their importance in the economy.

The Economic Council of Canada’s 1969 conclusions about competition policy and efficiencies were based on the conditions that prevailed at that time. This chapter examines to what extent these conditions have changed and whether specific consideration of efficiencies is still warranted in competition policy.

Trade liberalization and Canadian manufacturing

In 1969, Canada accounted for about 1.9 percent of world GDP and 0.6 percent of world population. The Canadian economy was relatively small (Maddison 2003) and protected by high tariffs, 15 percent on average (Canada 1969: 74–79). Although the economy was fairly open, free trade with the United States was not on the political horizon.⁸ The so-called National Policy, which featured significant government intervention and an array of trade barriers, guided federal economic policy. It was in these circumstances that the Economic Council promoted the idea of incorporating efficiency considerations into the competition framework: “The continuation of substantial tariff and non-tariff barriers for some time into the future is another reason behind the need for public policies geared to promote efficiency” (Canada 1969: 78).

But significant changes in the global trade environment were about to occur. Tariff barriers started coming down significantly around the world, first under the General Agreement on Tariffs and Trade and then under the World Trade Organization. In Canada, the Macdonald Commission (1985) came out in favour of free trade with the United States. Two years later, negotiations for a free trade agreement began.

The Canada–U.S. Free Trade Agreement was implemented in 1989. In 1994, it was extended to include Mexico and became the North American Free Trade Agreement (NAFTA). These two agreements dramatically opened the Canadian economy, eliminating all tariffs between the three countries and establishing non-discriminatory policies.

As a result of global trade liberalization, Canada’s exports increased dramatically. Figure 1 illustrates Canada’s trade exposure (imports plus exports, as a share of GDP), which doubled in the three decades that followed the Economic Council’s report. The largest increase occurred in the aftermath of the free trade agreements, between 1989 and 2000, when trade exposure peaked at 85 percent of GDP. The growth in Canada’s trade exposure was significantly larger than that of any other country in the Organisation for Economic Co-operation and Development (OECD), and Figure 1 presents the situation of three of them, namely Germany, Japan and the United States.

Much of Canada’s pre-2000 surge was brought about by the elimination of tariffs between Canada and the United States. Figure 2 shows the evolution of the tariffs on manufactured goods in Canada from 1983 to 1996. Starting in 1989, Canada progressively eliminated its tariff with the United States, while lowering its tariff with the rest of the world at a significantly lower rate. Figure 3 shows the same pattern for the United States.

The impact of the Canada–U.S. Free Trade Agreement and NAFTA on Canada and the U.S. was far from symmetrical. Canada’s trade was much more exposed to U.S. trade and tariff reduction than the other way around. The U.S. and Mexico accounted for about 78 percent of Canadian foreign trade in 2001, whereas Canada and Mexico only accounted for about 27 percent of U.S. foreign trade. As a result, Canada’s export sector was the object of much more trade pressure than its U.S. counterpart.

The net effect was to greatly reinforce the already significant level of integration of Canadian manufacturing with that of the United States, as shown in Figure 4. Canadian exports to the U.S. amounted to 15 percent of Canadian GDP in 1969, 18 percent in 1989 and 30 percent in 2004. Imports from the U.S. amounted to 15 percent of Canadian GDP in 1969, 18 percent in 1989 and 23 percent in 2004. Since 2000, Canada’s trade exposure with the U.S. (and other countries) has deteriorated under the combined impact of the post-2000 economic slowdown, 9/11, SARS and mad cow disease, all of which directly affected Canadian trade (Canada 2001–2003).

The changes free trade triggered in the Canadian economy are also evident in a graph known as the “L” curve (Figure 5). This graph maps the evolution of international trade exposure (y axis) against interprovincial trade exposure (x axis). From 1981 to 1991, as Canada debated free trade, there was little change in the international trade exposure of manufactured goods. But starting in 1991, as the effects of free trade were felt in the economy, international trade exposure increased significantly without a corresponding decline in interprovincial trade.⁹

The opening of markets for goods and the surge of imports in the post-free trade era forced Canadian manufacturers to adapt to new competitive conditions. The impact on productivity was significant, as Canadian manufacturing firms had to become significantly more efficient to survive.

Despite significantly freer trade, not all barriers to trade have been eliminated in the goods sector. Quotas still exist in some agricultural industries (e.g. dairy and poultry). Tariffs against manufactured goods from countries other than the United States and Mexico remain.¹³ Moreover, the cross-border movement of goods is still subject to an array of regulatory requirements (Canada 2004c, 18). On the whole, though, manufacturing is probably more open in Canada than it is in any other country of similar size.

Productivity in the Canadian services sector

Services were much less affected by the two free trade agreements, because they are not traded to the same extent as are goods. On the whole, the services sector has grown at a faster pace than has manufacturing: from 50 percent of GDP in 1969 to 67 percent in 2001. In 1969, manufacturing was the biggest component of the economy, accounting for 22 percent of GDP, followed by financial services (more specifically, finance, insurance, real estate and rental and leasing services) at 14 percent. Today, these two have switched places. In 2001, financial services accounted for 19 percent of GDP while manufacturing represented only 17 percent.14

The growth of the services sector is also reflected in employment data. In 1976, the services sector accounted for 65 percent of total employment. By 2004, that figure had climbed to 75 percent. Manufacturing employment decreased from 19 percent of total employment to 14 percent. Health care and social services, in contrast, grew from 8 percent to 11 percent of total employment.

Paradoxically, the increasing share of employment represented by the production of services was brought about partly by the significant productivity advances in goods production, especially manufacturing (Baldwin, Durand and Hosein 2001). Fewer resources are required to produce goods as productivity increases. As a result, the reduction in the real prices of goods over time has more than offset the increase in volume produced, which has led to the shrinking of the relative size in the economy of the goods-producing sector and of manufacturing in particular (Baldwin, Durand and Hosein 2001). Rising incomes and lower real prices for goods have left consumers with more income with which to purchase services, which has contributed to the growth in the GDP share of the services sector. In addition, service industries have grown as manufacturers outsource their non-core functions to specialists.

As Figure 10 shows, services remain a small component of international trade, accounting for less than 15 percent of Canada’s trade. This is because services are intangible and thus less transportable than goods. Most services, such as retail and personal services, are delivered locally, with production and consumption occurring simultaneously in many cases, such as in the case of a haircut. Further, the delivery of many services depends on the individual service provider. A hairdresser, for example, cannot export his or her services, short of travelling.

International trade in services is also hampered by the existence of numerous non-tariff barriers and regulations that impede competition in and entry into the services sector. A case in point is restrictions on foreign direct investment. A recent survey found that Canada has the second highest level of restrictions on foreign direct investment in the OECD (Baldwin and Sabourin 2004). These restrictions primarily affect service activities, curtailing foreign entry into industries that the government deems strategic, including broadcasting and telecommunications (2.8 percent of GDP), banking (3.2 percent) and air transport (0.4 percent). The government justifies foreign ownership restrictions in media and book distribution by citing the need to develop and protect national identity.

All together, industries in which foreign entry is constrained represent a small but significant portion — about 6.4 percent — of the economy, as Table 2 indicates.

Table 2

The importance of the constrained¹⁵ services, as a percentage of real GDP, 2004

Other areas of the service economy are subject to significant government ownership (e.g. electric utilities, the postal service) or restrictions associated with their public financing (e.g. health care, education). Competition in these areas is usually absent or heavily regulated, effectively pushing them outside the market economy. These public and para-public services represent more than 16 percent of the economy and include public administration (5.5 percent), health care and social assistance (5.9 percent) and public and post-secondary education (4.3 percent).

The Canadian services sector has evolved significantly since the late 1960s but it has been not been subjected to the same forces of change as was the Canadian manufacturing sector, in particular, the opening up of the Canada-U.S. border. Moreover, a significant share of the service economy is shielded either from competition by being in the public or para-public sector, or shielded from foreign competition because of public policy decisions. Thus, one would not expect productivity to have evolved in the services sector in the same fashion as in the manufacturing sector. Indeed, John Baldwin of Statistics Canada estimates that Canada’s productivity in the services sector is about three quarters of U.S. services sector productivity, whereas the gap is much narrower in the manufacturing sector, because of its openness to the U.S. economy.16

Canada’s productivity challenge

One of the puzzles of the Canadian economy is why, despite the massive opening of its economy to North American integration, its productivity remains significantly below that of the United States and, indeed, has been slipping relative to that of other countries.

Figure 11 shows two measures of the Canadian productivity gap with the U.S. One measure is GDP per capita and the other is labour productivity, or real output per hour worked, both expressed as a percentage of the U.S. figures. Relative output per capita increased in the 1970s and then fell until the mid-1990s. (There has been a small rebound since then.) Labour productivity measured in terms of hours worked fell steadily throughout the period.

Canada has also fared badly in comparison with other OECD countries, as Table 3 indicates. On labour productivity specifically, Canada fell from the fourth rank in 1969 to the sixteenth in 2004. Canadian output per hour worked is now at 79 percent that of the United States, significantly lower for instance than that of Ireland and France and lower than that of the U.K., Sweden and Australia, as shown in Table 3. In 1969, Canada’s labour productivity was higher than that of all these countries.

Table 3

Productivity ranking, selected OECD countries (out of a list of 21)

Source: www.ggdc.net

Moreover, the data from 1999 to 2004, also presented in Table 3, suggest that Canada is not catching up, despite NAFTA. On labour productivity growth, Canada ranks 14th out of 21 OECD countries for the period from 1999 to 2001.

Countless papers have been produced to explain Canada’s sluggish productivity performance, since 2000 in particular. Andrew Sharpe, Executive Director of the Centre for the Study of Living Standards, and one of the recognized specialists on the issue, recently reviewed the literature on Canada’s productivity performance and identified several factors that explain the productivity gap (Sharpe 2003).17

Sharpe identified what he deems the five most important factors that explain the productivity gap:

• the lower capital intensity of economic activity in Canada;
• Canada’s innovation gap, which is reflected in lower research and development expenditures and less patent activity than in the U.S. and in slower adaptation of best practices techniques (see Chapter 4 for more on this);
• Canada’s smaller high tech and, in particular, information and communications technology industries, compared to those in the U.S., which drove recent productivity growth;
• a lack of human capital at the top end of the economy, as seen in the proportionally fewer university graduates and scientists and engineers in research and development activities in Canada; and
• the smaller scale of manufacturing plants.

Factors that Sharpe ruled out are industry structures, human capital other than at the top end of the economy, taxes, social policies, unionization, labour market policies and product market regulations.

The impact of borders

One of the major assumptions behind the efforts to remove trade barriers such as tariffs was that they constituted an overwhelming influence on cross-border trade. Removing these barriers, it was assumed, would increase competition from imports, putting sufficient competitive pressure on domestic firms such that productivity would increase.

The persistence of the productivity gap has led economists to consider the role played by the so-called border effect in shielding domestic producers from international competition. Research initiated in the late 1980s by John McCallum, when he was an economics professor at McGill University, first pointed out that national borders mattered much more to commercial exchanges than was commonly assumed at the time, even in the absence of tariffs (McCallum 1995). His research showed that in the absence of tariffs, the Canada-U.S. border was 10 times as much of an obstacle to trade as the actual cost of transportation suggested. (In the so-called gravity model used by economists to explain trade patterns, this difference was equivalent to multiplying by 10 the distance from a plant to a market whenever there is border crossing.)

John Helliwell extended that research for all countries. In a series of papers published throughout the 1990s, he not only confirmed McCallum’s basic finding but also fully documented its effect, which pervades the economic structures of most countries, and of small countries in particular. (Helliwell 1998). In fact, Helliwell’s work shattered many of the assumptions behind the “open economy” models that economists and policy makers were using. He demonstrated that there is much more than tariff barriers associated with borders.20 Different laws and regulations, different service organizations, lower capital and labour mobility and, sometimes, different languages and cultures add up to a significant barrier to the smooth flow of goods and services across a national border. As a result, transnational organizations set up separate divisions in each country, creating an additional factor that strongly influences market definitions in multi-country situations.

Thus, the removal of tariffs on manufactured goods as a result of the Canada–U.S. Free Trade Agreement and NAFTA did not eliminate all obstacles to trade between Canada, the U.S. and Mexico. The border remains, with its trade-hindering effects. In manufacturing, then, there are sound reasons for firms to set up in Canada plants designed at a less-than-optimal global scale —  that is, not at the most efficient level that would minimize resources — despite the opening up of North American borders. These smaller Canadian plants can be competitive with imports from larger, more efficient U.S. plants, since the latter have to cope with border effects, which increase the landed price of their products in Canada.

Moreover, these border effects are likely to remain. In particular, the Internet will not eliminate them, since they result not only from differences in traditional trade inhibitors, such as regulations, but also from institutionally induced differences, such as the wider use of metric system in Canada. Thus, as long as countries differ, there will be border effects that will hamper trade in manufactured goods and somewhat protect domestic producers, allowing them to survive even when they are less productive or efficient than are their foreign competitors.

Helliwell estimates that border effects are twice as large for services as they are for goods. Not only does that explain why international trade in services has not grown as quickly as trade in goods but it also suggests that productivity differences may be significantly higher in the services sector than in the goods-producing sector (Helliwell 1998, 38). Different laws and regulations that specifically affect the provision of services, and subsequent differences in the way service businesses cope with these differences, are likely to have pervasive effects on services. For instance, laws and regulations that affect urbanization patterns influence the size of local markets and thus the size of the service establishments in these markets. In the constrained sectors, public policy prevents global corporations from entering the Canadian market, which affects the average size of the entities that operate in these sectors. Protection from direct foreign competition in the airline industry affects price structures in Canada. This, in turn, has an impact on demand and thus on resources used. All these non trade-related factors may contribute to lower productivity of services in Canada.

In other words, Canadian service providers are protected from competition from their U.S. rivals by the existence of important border effects. Because the Canadian market is smaller, Canadian service firms tend to be smaller than their U.S. counterparts and do not operate at the same efficiency when economies of scale are present.

As noted above, these border effects will not disappear and their continued existence will have dysfunctional effects on productivity and thus on efficiency21 (Sharpe 2003, 12).

Efficiency still matters

The Canadian economy has changed dramatically since 1969. It is now a very open economy, with one of the highest trade intensities among OECD countries. The impact of this growing openness is best felt in manufacturing. Canadian manufacturers are facing more U.S. competition than ever and are pulling their weight: they are competitive. In a sense, they have to be efficient or they would go out of business. But that does not mean that they are necessarily as efficient as their U.S. counterparts, because of border effects. Although the gap is probably not large, there is still cause for concern. A recent submission by Paul Darby, chief economist of the Conference Board of Canada, to the House of Commons Standing Committee on Banking and Commerce summarized the expert consensus on the issue: barriers to trade other than tariffs still have a “very significant impact on productivity in the primary and manufacturing sectors of the Canadian economy” (Darby 2005). Thus, despite the significant opening of the Canadian economy, the Economic Council of Canada’s 1969 argument may still be valid for manufacturing, implying that competition policy may still contribute to enhancing the efficiency of manufacturers.

This applies even more clearly to Canada’s services sector. Services are such that they are not a large part of international trade and have not, as a result, been significantly exposed to increased North American competition as tariffs were eliminated. Moreover, because of Canada’s particular pattern of urbanization, many Canadian service firms are smaller than their counterparts in the United States.

In addition, there is a large segment of the service economy that operates in a protected environment, very much like manufacturing did in the 1960s. Some of these industries are highly strategic in today’s economy: banking, telecommunications and media. It is possible that their protected status artificially maintains them at less than their optimal size and thus at less than full efficiency. Without the pressure from foreign firms, they are not as productive and innovative as they could be. Finally, many Canadian service firms are government-owned or -regulated (particularly in health and education) and as a result are somewhat shielded from competition.

Thus, the Economic Council’s 1969 concerns about inefficiency are still valid for services, although the sources of inefficiency are quite different from those affecting manufacturing in 1969.

Canadians should be concerned by the lower productivity of the Canadian economy. Although there is some evidence that much of the productivity gap between Canada and the U.S. can be attributed to the services sector, it is clear that the effects of freer trade were neither universal nor immediate. As time goes on, international competitive pressures may increase. In particular, cross-border competition is likely to become more prevalent in services and more non-tariff barriers are likely to be eliminated.

Still, border effects will remain and it is likely that they will always have some residual impact. Thus, the Economic Council’s 1969 argument will always be valid in some regard. It is the Panel’s view then that competition policy should continue to play a role in encouraging more efficiency in the Canadian economy.

8. Two brothers, Paul Wonnacott and Ronald J. Wonnacott, the latter teaching at the University of Western Ontario, were the best known of the few promoters of the idea among academic economists in the late 1960s (see Wonnacott and Wonnacott, 1967).

9. The “L” curve also shows the impact of the other forces shaping the economy. During the 1980s, as transportation and communications costs came down, interprovincial trade exposure shrank in Canada. Direct shipments became more important and plant rationalization started, as Canadian manufacturers started to shed their protection-based business models, with a resulting lowering of interprovincial trade.

10. For discussion of competitive pressure, see Martin (2003) and Institute for Competitiveness and Prosperity (2004 a, b).

11. See Caves (1998) for a review of the research on business dynamics. See Baldwin, Dunne and Haltiwanger (1998) for the comparison between Canada and U.S. data and in the evolution over time of business dynamics.

12. Other mentions included desiring to compete globally, government regulation, greater competition from outside of North America, NAFTA, changes in the exchange rate, and the relative lack of flexibility of Canadian workers compared to U.S. workers.

13. For example, tariffs on textile products from countries other than the United States average 9.6 percent; this figure is 14.1 percent for clothing products. Wyman (2005).

14. The Statistics Canada time series for GDP by industry in current dollars is currently only updated to 2001. However, the most up-to-date constant dollar series has data up to 2004, but only goes back to 1981 because of changes to industry classifications and measurement. For this particular comparison, the current dollar series was selected so that a single consistent time series could be used. For information purposes, in 2004 (constant 1997 dollars, basic prices), the goods-producing sector share was 31 percent, services-producing sector 69 percent, with manufacturing at 17 percent and finance, insurance, real estate and rental and leasing services (and management of companies and enterprises) at 20 percent.

15. The word constrained refers to foreign entry into this sector. Public services include public administration at all levels. Para-public services include health care and social assistance, postal service and urban transit.

16. Direct communications with the Panel, to be published in a forthcoming study of productivity in the Canadian services sector.

17. The International Productivity Monitor, a well-respected quarterly journal published by the Centre for the Study of Living Standards, an Ottawa think thank, is the main venue for debating productivity in Canada.

18. As in most countries, services are also regulated through qualifications, from those for professionals such as lawyers and accountants to those for people who work in skilled trades, such as electricians and plumbers. As a result of this controlled access, supply is restricted in these mostly domestic (e.g. non-traded) areas. Prices end up being higher and the businesses do not operate at their optimal efficiency. Consumer protection is the justification government uses for such controlled access, resulting in an implicit trade-off between regulated quality of services and their broader availability and lower prices.

19. Discussions with John Baldwin

20. A paper by Brown (2003) based on 1993 data suggests that in Canada the border effect is much less important than what McCallum and Helliwell found. Brown suggests a 2X factor instead of a 10X factor but does not challenge the existence of a border effect.

21. On the other hand, John Helliwell argues that the loss of efficiency may not be as a significant as thought under standard economic interpretations. His view is that border effects allow social trust to develop within a country, and social trust can greatly increase the overall efficiency of an economy by simplifying transactions and lowering their costs. The gains in efficiency from higher social trust may offset any losses from the lower scale of the “protected” production units. See Helliwell (1998, 124).