A summary of the sources of gain in the Wonnacott estimate are shown in Table 2. The major source of the benefit (left side of the table) would come from increased productivity of labour and other factors of production in manufacturing. The benefits would be realized in the form of reductions in relative prices and increases in wages and returns to other factors of production. The estimates are based on a comparable degree of resource utilization before and after free trade. The estimates in Table 2 are based on the year 1974 and a key part of the estimate is based on a closing of the productivity gap in manufacturing between Canada and the United States. This would still leave the key manufacturing plants in Ontario and Quebec slightly below adjacent plants in the U.S. northeast. To allow for possible errors, a range of 7 to 9 percent of GNP is favored in the study (Wonnacott, 1975, p. 178 and surrounding pages). An alternative approach to estimate the gains from free trade was used
by Harris and Cox (1983). They developed and used a computable general
equilibrium model (C.G.E. henceforth). This approach developed equations
for individual industries within manufacturing for both the demand side
and the supply side. They distinguished between industries that could be
approximated by a competitive model (where constant returns to scale may
be appropriate) and industries where oligopoly and monopolistic competition
(with product differentiation) was present. In the latter group of industries,
economies of scale were allowed for, and such industries would experience
more specialization, more outsourcing of components and more two-way flow
of international trade of the intra-industry form. Estimates were developed
for 20 industries within manufacturing and 9 in the non-manufacturing group.
Such class of models could be adapted to simulate changes in employment
and unemployment with free trade in addition to gains in real income. Their
estimate of the gains from free trade was 8.6 percent of GDP, very similar
to the estimate by Ron Wonnacott for a similar year in the 1970's, as shown
in Table 1 above.
Source: Wonnacott (1975, p. 178)
What does the aggregate evidence on the gains from freer trade show to date for Canada? Canada has been undergoing a series of tariff reductions since 1947. This paper will concentrate on the period since 1977, after a brief comment on the automotive agreement. The tariff reductions that have been implemented and were still having an influence over the last two decades would include:
These all involved a phasing in of the tariff reductions, normally spread over ten years. There had been a fairly persistent drop in the ratio of duties to total imports for Canada. However, in the mid 1980's, the rate of price protection for manufactured products was still 6.5 percent (primarily from tariff rates, but some quantitative restrictions also (Department of Finance 1988, pp. 15-20). The period from the mid 1970's to date thus include important further tariff reductions, including most of the CUSFTA reductions (the last few reductions were implemented January 1, 1998). The period of the mid 1970's also covers the two major studies referred to in Table 1 that show large gains from freer trade for Canada. A further consideration is that the estimates of real output for manufacturing in the U.S. have not yet been revised for the years prior to 1977. It should be noted that the estimates of real output per hour in the automotive industry had seen the disappearance of the earlier gap in real output per hour worked of more than 30 percent in 1966 below the U.S. to 2 to 4 percent by the late 1970's and 1980's (Fuss and Waverman, 1992, p. 201). On the international trade side, there has been a major increase in the two way flow of trade between Canada and the U.S. in motor vehicles and parts. Plants in Canada produce one (or a few) models and they are then sold in the whole North American market. These results were in line with the predictions of large gains in real output in relation to labour and materials and capital inputs made by the supporters of free trade. Table 3 provides some aggregative evidence on productivity and real wage comparisons for Canada and the United States for selected years over the last two decades. The changes over time in real GDP per employed person are relatively small, after a significant narrowing up to about the mid 1970's. Real output per hour in manufacturing shows a widening from a Canadian level about 25 percent below the U.S. in 1977 to more like 40 percent below in 1996. The tariff reductions over that two decade period from the GATT negotiations and the CUSFTA
(with implementation starting on January 1, 1989) have not yet had the
predicted result of a narrowing in the productivity gap in manufacturing.
It continues to be wider in manufacturing than for the economy as a whole
in spite of higher capital stocks per employer in manufacturing in the
last comparison available.2 A significant degree of foreign
ownership persists, with about 40 percent of total shipments being made
by plants that were totally foreign controlled (primarily in the U.S. )
(CALURA, 1994, pp. 77-84). This foreign ownership has encouraged the relatively
easy flow of new technology and managerial and industrial practices into
branch plants in Canada.
Sources: Column 1 from B.L.S. (1998), Table 3 based on 1993 Benchmark EKS PPP; Column 2 from Daly MacCharles (1986a, p.65), updated from B.L.S. (1998); Column 3, price level comparisons from Summers, Kravis and Heston (1980), updated from B.L.S. Aug. 15, 1997. The failure of the productivity gaps to narrow further in the 1980's and 1990's is partly related to the wider gaps in the unemployment rates and output gaps between Canada and the United States that were present in the later periods. Slack demand is reflected in lower levels of output in relation to capital and labour inputs, as well as higher levels of unemployment. (Canadian Public Policy, Supplement, February, 1988) . It should be noted that press and television reports over the last decade or more have emphasized plant layoffs and closures, corporate downsizing by major companies (including foreign subsidiaries in Canada), the substitution of capital for labour, increased outsourcing by large companies, and the increased use of automation and computers in business. Similar developments are also taking place in the United States, of course. How can one reconcile the widespread information on these developments in large plant and firms in Canada with the absence of any significant narrowing in the productivity gap in Column 2 in Table 3 above? An interpretation and some evidence on this point will be provided in the next section. The evidence on real wage levels in the two countries in Column 3 of Table 3 is striking. Real wages in Canadian manufacturing are within three percent of the U.S. in all the years shown - markedly less than in the productivity comparisons for either manufacturing or the economy as a whole. For both Canada and the United States the increases in real wages in manufacturing have been significantly less than the increase in real output per hour in the same broad industry group over the last two decades. For Canada, the increase from 1977 to 1996 was 7.3 percent compared to 35.9 percent in output per hour. For the U.S. over the same period the increase was only 3 percent in real wages compared to an increase of 60 percent in output per hour (BLS, August 15, 1997). There was thus a slight narrowing in the gap in real wages while there was an important widening in the productivity gap in manufacturing. Japan has been experiencing a significantly smaller increase in real hourly wages in manufacturing than in real output for many decades. This has continued over the last two decades with an increase in real wages of 50 percent, whole output per hour more than doubled, (BLS August 1997) Japanese manufacturers were passing a major part of the productivity gains in manufacturing along to the buyers of manufactured products within Japan, but also internationally. This had permitted Japanese manufacturers to get an increased share of the rapidly growing market for manufactured products, especially up until 1985. Since then, the appreciation of the Yen relative to many other currencies has drastically slowed the growth in their volume of manufactured exports. Developments in Canadian trade in manufactured products have been in
line, or have even exceeded, the high growth predicted by those supporters
of free trade who had predicted large economic gains. The increased specialization
at the plant level associated with more outsourcing of small parts and
components to smaller organizations domestically would also be reflected
in more intra-industry trade. The earlier increase in the two way flow
of trade in motor vehicles and parts has broadened to include other manufactured
goods. During the 1990's, exports of "other manufactured products" has
exceeded that of motor vehicle and parts in 1986 dollars. By the mid 1990's,
exports of manufactured products in 1986 dollars had begun to exceed total
exports of food, energy materials and other natural resource materials.
These changes have been fully analyzed in Schwanen (1997). It should be
noted, however, that large increases in imports of manufactured products
have also been taking place and Canada continues to have a large net deficit
in trade in manufactured products. For the purposes of this paper the question
is how this increase in both manufactured exports and import competition
affects the small Canadian-owned establishments in contrast to the large
multinationals with one or more large plants.
Before turning to the evidence on the role of small business in the productivity gap in manufacturing, it may be helpful to highlight the role of large business and the multinationals - what one can appropriately term the Galbraith thesis. In The New Industrial State J. K. Galbraith put forth the thesis of the dominance of the large corporation. He emphasized its importance in manufacturing, petroleum, banking and finance and the important role of defense industries. He emphasized interlocking directorships in the private sector and the close relationships of business leaders with government in the military - industrial complex. A theme was the dominance of the large corporation in the economy and in public policy. When first written there was still limited international competition from imports on domestic manufacturing. It is not too strong to say that this has been the dominant paradigm in the economic and business literature over the last three decades. On the Canadian side this has been reflected in the large literature on foreign ownership and the multinationals. Illustrations of such studies are Government of Canada (1972), Rugman (1981), Rugman and McIlveen (1985), Eden (1994) and the many references in each study. The Porter and Monitor Corporation (1991) deserves special comment. It points out that 64.2 percent of manufacturing shipments in Canada were made by firms with less than 500 employees, while only 17.7 percent of U.S. manufacturing shipments were made by the same size of firms (Porter 1991, p. 293). In spite of the much greater relative importance of small firms in Canada, about 95 percent of the 468 page volume concentrated on large, rather than small firms! Some initial discussion of the role of small business began in the 1970's. Schumacher (1973) emphasized the social benefits of small business. Litvak (1971) did a study on the role of entrepreneurship in new firms. Peterson (1977) provided a comprehensive discussion of small business in Canada (with some comparisons with the United States), and issues of growth in size and financing. It received widespread distribution, partly through the Canadian Federation of Independent Business. John Birch stimulated interest in the role of small business in providing new jobs (Birch 1987 which expanded an earlier 1979 report for M. I. T.). Case (1992) provides a useful review of the debate that John Birch's initial study created and also has some interesting case studies of the role of small firms in the high technology field, emphasizing the flexibility and fast response that can be achieved in small entrepreneurial companies. There have also been a number of studies of small business in Canada by government, trade associations and research groups, including Amboise (1991), Ministry of Industry, Trade and Technology (1991), Industry, Science and Technology (1992), Canadian Chamber of Commerce (1992), Gray (1994), MacIntosh (1994), Sharwood Report (1994), Canadian Department of Industry (1994) and Industry Canada (1995). Studies of small business is almost becoming a growth industry. The big growth in small business studies came after the earlier work on free trade and was not really related to Canada's performance in manufacturing over the last two decades of reduced tariffs. For our purposes, there are two initial questions, namely:
Sources: Statistics Canada, Manufacturing Industries of Canada: Type
of Organization and Size of Establishment (Ottawa: No. 1975) p.19 and Statistics
Canada, Manufacturing Industries of Canada: National and Provincial Areas,
1993 (Ottawa: Dec. 1995), p.201.
It is clear that there has been a remarkable shift in the size distribution of employment, with a significant net reduction in employment in large establishments (over 200 employees), and a significant net expansion in employment in the smaller sized establishments. This, of course, reflects a very dynamic process of births and deaths and shifts in the size categories of continuing plants. This shift in employment towards smaller establishments has been well documented in additional detail by John Baldwin and associates in Statistics Canada. They have been able to follow changes for individual plants over time from their micro longitudinal panel data sets, which provide a perspective on dynamic changes not possible from published data. Baldwin and Picot (1994) found that net job creation for smaller establishments in Canadian manufacturing was greater than in large establishments from 1970 to 1990. Baldwin (1996) showed that the small plants in manufacturing experienced an increased share of employment from 1973 to 1992 in all five industry sectors (resource based, labour intensive, scale related, product differentiated, and science based). For the purposes of this paper, it is also significant that the growth in employment in smaller establishments in Canada than the United States (Baldwin-Picot, 1994, esp. Figures 1 to 6 and related discussion). It is also important that the small plants were becoming increasingly important during the 1980's. It might be noted that the data on employment by company size shows a similar pattern, with a decline in the median size of company from 1978 to 1988 for manufacturing, and almost all the individual industries within manufacturing (Wannell, 1991, esp. Charts 13 to 16 and related discussion). In light of the evidence on the increased importance of small establishments and small companies in Canada, it seems desirable to speculate on some of the reasons for this growth. One consideration is the improvement in communication and transportation. Fax's, E-mail, lower rates for long-distance telephone calls make it faster and cheaper to communicate orders and information. Truck transportation provides more flexibility and faster door to door delivery than freight trains. Serviced land costs can be lower a short distance away from major metropolitan areas, one storey buildings can be cheaper than the older higher rise factories of earlier decades. The lower costs of computers and flexible software packages for accounting, payrolls and personnel records permit small establishments and firms to stay close to the leading edge of new technology - both in the plant and the office. Small business can be more flexible in creating and adopting new technology (Audretsch 1996). Newspapers and company reports indicate an increased amount of outsourcing by large companies and plants. Parts and components that were previously made internally might now be subcontracted out - a shift from hierarchy to markets. This can be partly due to the flexibility of smaller establishments or lower wages in small plants (see the next section). The net reduction in employment in large plants was not limited to blue collar workers, but there were frequent examples of early retirements and layoffs in middle management as well in the sluggish growth since mid 1975. Some of those laid off have set up their own small companies and plants. They may be willing to make smaller earnings but get the satisfaction of running their own business. The lack of employment alternatives in large plants and firms may encourage the growth of smaller organizations. Tax considerations can also be a contributing factor to the growth in small manufacturing businesses. The corporate tax rate on small incorporated companies is lower than on large companies, and it is also lower than the marginal tax rate under the personal income tax (for Ontario, for example). These tax incentives were designed to encourage smaller businesses and they may have contributed to these results. All of these considerations thus far can apply to the United States, as well as Canada. A new consideration for Canada is the improved access to the U.S. market made possible by the elimination of the last tariffs in the U.S. (We are aware of the continued risks to exports from contingency protection in the U.S. ) There were about twenty U.S. states within 400 miles of the manufacturing belt between Windsor and Cornwall and those states had levels of population and income five or six times all of Canada. Quebec is not quite as well situated by U.S. states within 400 miles, but the population and income in those adjacent states was two or three times the Canadian totals. This much larger market can permit a small plant to be highly specialized and sell in a market substantially greater than anything available historically. There are, however, greater risks in a highly competitive market, whose profitability can be quickly changed by exchange rate changes. There is also the increased scope for import competition (as can be seen in import statistics for manufactured products). This enlarged market from free trade can be a factor in the faster growth in small plants in Canada than in the U.S., and the acceleration in the growth of small plants in the 1980's. However, it is still puzzling that Canada now has relatively more small
plants than the U.S., where they have grown up with easy access to a larger
national market.
How has this increased role for small business in Canada affected the productivity gap in manufacturing? Figure 1 shows an important contrast between small and large plants. In 1973, large plants had levels of productivity about 15 percent above the national total, but two decades later it was about 40 percent above. On the other hand, small plants had levels about 15 percent below the national average, and this had widened to about 30 percent about two decades later. The gap between large and small had widened from about 30 points in 1973 to about 70 points in 1992 and the widening in the gap accelerated during the 1980's. Since the gap between Canada and the United States has widened a bit over this period (see Table 3 above), it would appear that large plants in Canada have had bigger productivity gains than the U.S. manufacturing total. This is in line with the studies of large gains from freer trade. On the other hand, the big growth in small plants with lower (and declining)
levels of productivity had not been anticipated by either the supporters
or the critics of free trade.
There is also some evidence that the lower levels of productivity are heavily concentrated in the Canadian-owned sector of manufacturing. This can be seen in Table 5 for the 1970's. Large Canadian-owned plants (more than 400 employees) were roughly comparable to foreign-owned larger plants, but smaller Canadian-owned plants were only half or two-thirds as productive as small foreign-owned plants (based on the same size and industry cells that contained both forms of ownership).
Source: D. J. Daly and D. C. MacCharles, Canadian Manufactured Exports: Constraints and Opportunities, (Montreal: IRPP, 1986), p.20.
Figure 2 shows the relative productivity by size class and ownership. The larger plant sizes have undergone increases in relative productivity levels in the later period, while smaller domestic plants (under 100 employees) have declined in relative levels. There is only a small difference in relative productivity levels of large and small foreign establishments, but small domestic establishments had productivity levels only half the foreign establishments with under 100 employees in the 1985-90 period. It is clearly the small domestically-owned plants where the relative productivity levels have experienced the greatest relative declines. Similar contrasts between large foreign-controlled establishments and small domestically controlled establishments emerged in an econometric study, using time series data from 1973 to 1993 for three size groups and four broad industry groups in manufacturing. Both marginal and average labour productivities were estimated. Average labour productivity was measured by total shipments divided by total employment, with shipments measured in real terms by the output price index at the corresponding 4-digit industry level. (Baldwin and Dhaliwal 1998). The study found stronger productivity growth in larger than smaller plants and the largest gains occurred in large foreign-controlled plants. Similar results were obtained when net output rather than gross shipments were used in the measurements of outputs. There are some differences between small and large establishments in
value added per employee in the United States, but the differences are
not quite as great as those shown in Tables 5 and 6 for Canada. However,
an even more important factor is that the smaller establishments are relatively
more important in Canada than in the United States. Establishments with
fewer than 500 employees contributed about 64 percent of shipments, compared
to only 55 percent in the U.S., as shown in Table 6.
Sources: U.S. Department of Commerce, 1992 Census of Manufactures, General Summary, p. 1-178 and Statistics Canada, Manufacturing Industries of Canada: national and provincial areas, 1993, p. 200. Clearly the increased importance of employment in small business has been associated with a slippage in productivity levels of small Canadian-controlled establishments relative to the national totals. It is these developments that have prevented the productivity levels in Canadian manufacturing from closing the gap with the U.S. as expected. It is also clear that small firms in manufacturing pay less than larger firms (Morrisette 1993). From 1972 to 1991 there was also a tendency for large plants to experience increases in relative wages, and small plants to undergo relative declines (based on 7 classes of plant size). By 1991the relative wages of the smallest to the largest class size was wider by 1991 than it had been in 1973 in all five broad industry groups within manufacturing. These tendencies were more apparent in the 1980's than in the 1970's. (Baldwin 1996). The increased relative importance of small plants (and firms) had thus contributed to a smaller increase in real wages for total manufacturing as well as slowing the increase in productivity for total manufacturing in Canada. When the increases in employment are occurring in lower wage establishments,
there is an understandable reluctance for workers laid off in high wage
establishments to accept employment in the low wage establishments. When
this shift appears to be more apparent for Canada than the United States,
this could be important in the persistence of a higher unemployment rate
in Canada. This possibility does not seem to have been considered in the
studies published in the recent special issue of Canadian Public Policy
on the Canada - U.S. unemployment gap.
The increased relative importance of small plants and firms in Canadian manufacturing has clearly been a key factor in the growth of employment and the slower increase in productivity and money and real wages. However, it is not clear how the dramatic increase in the two way flow of trade in manufactured products between Canada and the United States has affected the small and large establishments and the Canadian and foreign-controlled establishments. Has the increase in exports for large foreign-owned establishments been above average? On the other hand, have the smaller Canadian-owned establishments been particularly hard hit by competition from the large economy to the south? (Daly 1990). It should be noted that a previous study had shown faster increases in exports of small Canadian-owned establishments than in foreign subsidiaries. This special tabulation covered the years 1970 to 1979 (Baldwin and Gorecki 1983). This project required a matching of establishment data from the census of manufacturing with the commodity data from international trade statistics. Changes in exchange rates (both up and then down for the Canadian dollar), the implementation of free trade, the increased relative importance of employment in smaller Canadian-owned establishments and the depressions and slower growth in Canada in the 1980's and 1990's make an update of that material essential. Another area is the desirability of re-examining the evidence on economies of scale. There has clearly been a decline in the average size of plants and the average size of firms in manufacturing over the last two decades. Possible factors contributing to this change were mentioned earlier, but the relative importance of the factors has not been studied to our knowledge. It may not be an easy project to do, of course. Another area is the topic of small business financing. Costs are relatively higher for small firms to access the capital market. Historically, the chartered banks lent to small business, but often relied on physical assets and inventories as a basis of lending. The increased importance of human capital, new technology and computer applications have become more important in the knowledge intensive industries of the present and future. It is much more difficult to assess credit risks in these new and smaller plants and companies. There is considerable variability between one small plant and another, with much higher rates of both exit and entry among small establishments. Corporate profit levels have been lower in Canadian manufacturing than in the U.S., especially for smaller firms (Wilson 1995) There is also very little correlation between the growth in employment in one period (1983-86) and the next period (1986-90) (Picot and Dupuy 1996). This makes the whole topic of financing of small business a very difficult one. Small business has been an important source of employment growth in Canadian manufacturing, but the contribution of small plants to real and money wage income, to profits and to productivity per employee has been to pull down the national totals. Small plants have clearly been a factor in checking the narrowing in
the productivity gap between Canadian and U.S. manufacturing that has been
taking place in the larger plants, especially the foreign-owned.
A special thanks to John Baldwin for making available the numerous studies done by he and his colleagues and advice and encouragement on this project.
Amboise, Gerald D' (1991) The Canadian Small and Medium-sized Enterprise: Situation and Challenges (Halifax, Nova Scotia: The Institute for Research for Public Policy). Audretsch, David B. (1996) New Firms and Creating Employment (Berlin: Forschungsschwerpunkt Marktproze? und Unternehmensentwicklung). Baldwin, John (1996) Were Small Producers the Engines of Growth in the Canadian Manufacturing Sector in the 1980s? (Ottawa: Statistics Canada). Baldwin, John (199?) "Productivity Growth, Plant Turnover and Restructuring in the Canadian Manufacturing Sector" presented at a conference "Sources of Productivity Growth in the Eighties." Baldwin, John and Naginder Dhaliwal (1998) "Employment Responses to Output Growth: An Econometric Analysis of Differences Between Domestic and Foreign-Controlled Firms" (Ottawa: Statistics Canada). Baldwin, John and Garnett Picot (1994) Employment Generation by Small Producers in the Canadian Manufacturing Sector (Ottawa: Statistics Canada; Statistics Studies Branch). Birch, David L. (1987) Job Creation in America: How our Smallest Companies Put the Most People to Work (New York: The Free Press). Blackford, Mansel G. (1991) A History of Small Business in America (New York: Twayne Publishers). B.L.S. (1998) United States Department of Labor News, March 27. Canadian Chamber of Commerce (1992) Small Business Guide to Mastering Strategic Change (Ottawa). Canadian Department of Industry (1994) Growing Small Business (Ottawa: Communication Branch). Case, John (1992) From the Ground Up (New York: Simon & Schuster). Eden, Lorraine (1994) Multinationals in North America (Calgary, University of Calgary Press for the Minister of Supply and Services). Galbraith, J. K. (1967) The New Industrial State (Houghton Mifflin). Government of Canada (1972) Foreign Direct Investment in Canada (Ottawa: Information Canada). Gray, Douglas A. (1994) Complete Canadian Small Business Guide, 2nd ed. (Whitby, Ontario: McGraw Hill Ryerson, 1994). Industry Canada (1995) Small Business: A Progress Report (Ottawa). Industry, Science and Technology (1992) Exporting for Competitiveness: 10 Steps for Small Business: Report on Small Business in Canada (Ottawa: Entrepreneurship & Small Business Office, Industry, Science and Technology Canada). Litvak, Isaiah A. (1971) Canadian Entrepreneurship: A Study of Newly Established Firms (Ottawa: Industry, Trade and Commerce, Technology Branch). MacIntosh, Jeff (1994) The Financial Economics of Small Firms (Toronto: Canadian Law & Economics Association, Faculty of Law, University of Toronto). Ministry of Industry, Trade and Technology (1991) The State of Small Business: 1990 Annual Report on Small Business in Ontario (Toronto: Queen's Printer for Ontario). Morrisette, R. (1993) "Canadian Jobs and Firm Size: Small Firms Pay Less?" Canadian Journal of Economics, pp. 159-174. Peterson, Rein (1977) Small Business: Building a Balanced Economy (Erin, Ontario: Press Porcepic Ltd.). Rugman, Alan M. (1989) The Impact of Free Trade on Small Business in Canada (Toronto: Centre for International Business Research Program). Rugman, Alan M. (1981) Inside the Multinationals: The Economics of Internal Markets (New York: Columbia University Press). Rugman, Alan M. and John McIlveen (1985) Megafirms: Strategies for Canada's Multinationals (Toronto: Metheun). Schumacher, E. F. (1973) Small is Beautiful: Economics as if People Mattered (New York: Harper and Row). Sharwood Report (1994) "Canada-Small is Beautiful-but Growth is Better" (Toronto). Wannell, Ted (1991) Trends in the Distribution of Employment by Employer Size: Recent Canadian Evidence (Ottawa: Statistics Canada). Wilson, Jack D., Bill Potter and Art Kirkwood (1995) Corporate Canada/Corporate America: A Comparison of Financial Performance (Ottawa: Statistics Canada, August).
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
