Asleep at the Switch: The Political Economy of Federal Research and Development Policy since 1960 9780773596535

Table of contents :
Cover
CARLETON LIBRARY SERIES
Title
Copyright
Contents
Tables and Figure
Abbreviations
Acknowledgments
Introduction The Canadian Federal State and Domestic Technological Development
1 Promoting Domestic Technological Capacities: State Strategies and Social Antagonisms
Part One Permutations of Dependent Technological Development: From Early Fordism to Neoliberal Restructuring
2 Entrenching Dependent Technological Development: Canadian Fordism in the Early Twentieth Century
3 Reasserting Dependent Technological Development: Canadian Fordism in the Postwar “Golden Age”
4 Another Form of Dependent Technological Development: Post-Fordist Accumulation in the Neoliberal Era
Part Two Permutations of the Glassco Framework: Promoting R&D from Diefenbaker to Chrétien
5 Beginning the Process: The Diefenbaker Tories and R&D Incentive Programs, 1957–63
6 Internal Struggles: Left-Liberals, the Glassco Framework, and R&D Policy, 1963–68
7 Further Contestation: The Gray Initiative, 1968–71
8 Limiting Change: Industrial Restructuring and Social Forces, 1971–73
9 Extending the Glassco Framework: R&D Policy in the 1970s
10 Last Challenge to Transnational Capital: Left-Liberals and State-Led Strategies, 1980–81
11 Moving to the Right: The Trudeau Liberals and R&D Incentive Programs, 1981–84
12 The Glassco Framework in an Era of Free Trade: The Mulroney Tories and R&D Policy, 1984–93
13 Final Episode: The Chrétien Liberals, Transformative Strategies, and the Glassco Framework, 1993–2000
Conclusion The Impasse of the Federal State and Canadian Industrial R&D
Notes
Index

Citation preview

as l e e p at t h e sw it c h

26410_Smardon.indd 1

2014-09-17 10:53:21

CA RLETON LIB RARY S E RIE S The Carleton Library Series publishes books about Canadian economics, geography, history, politics, public policy, society and culture, and related topics, in the form of leading new scholarship and reprints of classics in these fields. The series is funded by Carleton University, published by McGill-Queen’s University Press, and is under the guidance of the Carleton Library Series Editorial Board, which consists of faculty members of Carleton University. Suggestions and proposals for manuscripts and new editions of classic works are welcome and may be directed to the Carleton Library Series Editorial Board c/o the  Library, Carleton University, Ottawa K1S 5B6, at [email protected], or on the web at www.carleton.ca/cls. board members: John Clarke, Sheryl Hamilton, Jennifer Henderson, Laura Macdonald, Paul Litt, Stanley Winer, Barry Wright

cls

209 The Fighting Newfoundlander A History of the Royal Newfoundland Regiment G.W.L. Nicholson 210 Staples and Beyond Selected Writings of Mel Watkins Edited by Hugh Grant and David Wolfe 211 The Making of the Nations and Cultures of the New World An Essay in Comparative History Gérard Bouchard 212 The Quest of the Folk Antimodernism and Cultural Selection in Twentieth-Century Nova Scotia Ian McKay 213 Health Insurance and Canadian Public Policy The Seven Decisions That Created the Canadian Health Insurance System and Their Outcomes Malcolm G. Taylor 214 Inventing Canada Early Victorian Science and the Idea of a Transcontinental Nation Suzanne Zeller 215 Documents on the Confederation of British North America G.P. Browne 216 The Irish in Ontario A Study in Rural History Donald Harman Akenson 217 The Canadian Economy in the Great Depression (Third edition) A.E. Safarian 218 The Ordinary People of Essex Environment, Culture, and Economy on the Frontier of Upper Canada John Clarke

26410_Smardon.indd 2

219 So Vast and Various Interpreting Canada’s Regions in the Nineteenth and Twentieth Centuries Edited by John Warkentin 220 Industrial Organization in Canada Empirical Evidence and Policy Challenges Edited by Zhiqi Chen and Marc Duhamel 221 Surveyors of Empire Samuel Holland, J.F.W. Des Barres, and the Making of The Atlantic Neptune Stephen J. Hornsby 222 Peopling the North American City Montreal, 1840–1900 Sherry Olson and Patricia Thornton 223 Interregional Migration and Public Policy in Canada An Empirical Study Kathleen M. Day and Stanley L. Winer 224 How Schools Worked Public Education in English Canada, 1900–1940 R.D. Gidney and W.P.J. Millar 225 A Two-Edged Sword The Navy as an Instrument of Canadian Foreign Policy Nicholas Tracy 226 The Illustrated History of Canada 25th Anniversary Edition Edited by Craig Brown 227 In Duty Bound Men, Women, and the State in Upper Canada, 1783–1841 J.K. Johnson 228  Asleep at the Switch The Political Economy of Federal Research and Development Policy since 1960 Bruce Smardon

2014-09-17 10:53:21

Asleep at the Switch The Political Economy of Federal Research and Development Policy since 1960

Bru c e S m a r d o n

Carleton Library Series 228 McGill-Queen’s University Press Montreal & Kingston • London • Ithaca

26410_Smardon.indd 3

2014-09-17 10:53:21

©  McGill-Queen’s University Press 2014 ISB N ISB N ISB N ISB N

978-0-7735-4426-0 (cloth) 978-0-7735-4427-7 (paper) 978-0-7735-9653-5 (e P DF ) 978-0-7735-9654-2 (e P UB)

Legal deposit fourth quarter 2014 Bibliothèque nationale du Québec Printed in Canada on acid-free paper that is 100% ancient forest free (100% post-consumer recycled), processed chlorine free McGill-Queen’s University Press acknowledges the support of the Canada Council for the Arts for our publishing program. We also acknowledge the financial support of the Government of Canada through the Canada Book Fund for our publishing activities.

Library and Archives Canada Cataloguing in Publication Smardon, Bruce, 1955–, author Asleep at the switch: the political economy of federal research and development policy since 1960 / Bruce Smardon. (Carleton library series; 228) Includes bibliographical references and index. Issued in print and electronic formats. ISB N 978-0-7735-4426-0 (bound). – IS BN 978-0-7735-4427-7 (pbk.). – ISB N 978-0-7735-9653-5 (eP DF ). – IS BN 978-0-7735-9654-2 (eP U B ) 1. Research, Industrial – Government policy – Canada – History – 20th century.  2. Technological innovations – Government policy – Canada – History – 20th century.  I. Title.  II. Series: Carleton library series; 228 T177.C2S56 2014   607.2'7109045    C 2014-904353-8 C 2014-904354-6

This book was typeset by Interscript in 10.5/13 Sabon.

26410_Smardon.indd 4

2014-09-17 10:53:21

To Jan

26410_Smardon.indd 5

2014-09-17 10:53:21

26410_Smardon.indd 6

2014-09-17 10:53:21

Contents

Tables and Figure  ix Abbreviations xi Acknowledgments xiii Introduction:  The Canadian Federal State and Domestic Technological Development  3   1 Promoting Domestic Technological Capacities: State Strategies and Social Antagonisms  29 Pa rt o ne P e r mutat i ons o f D e p e n d e n t Tec h nol ogi c a l D e v e l o p me n t : F ro m E arly F o rd i s m to N e o l i b e r a l R e st ruc t ur i n g   2 Entrenching Dependent Technological Development: Canadian Fordism in the Early Twentieth Century  51   3 Reasserting Dependent Technological Development: Canadian Fordism in the Postwar “Golden Age”  77   4 Another Form of Dependent Technological Development: Post-Fordist Accumulation in the Neoliberal Era  99 Pa rt t wo P e r mutat i o ns o f t h e G l as s co Fr a me wo r k : P ro mot i ng r &d f ro m D i e f e n bake r to C h r é t i e n   5 Beginning the Process: The Diefenbaker Tories and r& d Incentive Programs, 1957–63  135

26410_Smardon.indd 7

2014-09-17 10:53:21

viii Contents

  6 Internal Struggles: Left-Liberals, the Glassco Framework, and r &d Policy, 1963–68  156   7 Further Contestation: The Gray Initiative, 1968–71  190   8 Limiting Change: Industrial Restructuring and Social Forces, 1971–73 215   9 Extending the Glassco Framework: r& d Policy in the 1970s  233 10 Last Challenge to Transnational Capital: Left-Liberals and StateLed Strategies, 1980–81  262 11 Moving to the Right: The Trudeau Liberals and r& d Incentive Programs, 1981–84  296 12 The Glassco Framework in an Era of Free Trade: The Mulroney Tories and r & d Policy, 1984–93  308 13 Final Episode: Transformative Strategies, the Glassco Framework, and the Chrétien Liberals, 1993–2000  333

Conclusion: The Impasse of the Federal State and Canadian Industrial r & d  359 Notes 391 Index 479

26410_Smardon.indd 8

2014-09-17 10:53:21

Tables and Figure

T abl e s 2.1 Value of manufacturing production by size: Canadian production establishments with outputs of $200,000 and over  61 2.2 Value of manufacturing production by size: US production establishments with outputs of $100,000 and over  62 2.3 Tier 1 industries in the United States compared with Canadian industries with production establishments of $1 million and over 64 2.4 Tier 2 industries in the United States compared with Canadian industries with production establishments of $1 million and over 65 3.1 Percentage of employees in establishments with 500 workers or more: Canada and United States, 1929–72  87 4.1 Productivity breakdown, United States and Canada, 1970–71 to 1977–78 110 4.2 Productivity breakdown, United States and Canada, 1978–79 to 1984–85 111 4.3 Productivity breakdown, United States and Canada, 1985–86 to 1992–93 112 4.4 Imports as a percentage of the domestic market, 1965–70, 1979, 1989–91: machinery, transport equipment, electrical and electronic products  115 4.5 Imports as a percentage of the domestic market, 1980 and 1993: high-technology products – G7 countries  116

26410_Smardon.indd 9

2014-09-17 10:53:21

x

Tables and Figure

6.1 Federal assistance to industry for r & d and innovation, 1959–60 to 1967–68, in millions of dollars  176 6.2 Expenditures on r & d by major federal agencies, 1960–61 to 1966–67, in thousands of dollars  184

f i g u re 3.1 r & d financed by government and industry, United States, United Kingdom, and Canada, 1939–59  97

26410_Smardon.indd 10

2014-09-17 10:53:21

Abbreviations

b c n i b er d c dc c fi c ic c lc c u fta dds dip dir dr ee dr ie edp fir a g att g er d g n p ic t ir a p ir dia ir dp istc it&c msed mser d mosst n a b st

26410_Smardon.indd 11

Business Council on National Issues business enterprise expenditure on r& d Canada Development Corporation Canadian Foundation for Innovation Committee for an Independent Canada Canadian Labour Congress Canada–United States Free Trade Agreement Defence Development Sharing Defence Industry Productivity Defence Industrial Research Department of Regional Economic Expansion Department of Regional Industrial Expansion Enterprise Development Program Foreign Investment Review Agency General Agreement on Tariffs and Trade gross domestic expenditure on r& d gross national product information and communications technology Industrial Research Assistance Program Industrial Research and Development Incentives Act Industrial and Regional Development Program Industry, Science and Technology Canada Department of Industry, Trade and Commerce Ministry of State for Economic Development Ministry of State for Economic and Regional Development Ministry of State for Science and Technology National Advisory Board on Science and Technology

2014-09-17 10:53:21

xii Abbreviations

n a fta n dp n c e n ep n rc oec d pa it pmo ppb r & d r d&i sr &ed srtc smes tpc wto

26410_Smardon.indd 12

North American Free Trade Agreement New Democratic Party Networks of Centres of Excellence National Energy Program National Research Council Organisation for Economic Co-operation and Development Program for the Advancement of Industrial Technology Prime Minister’s Office planning, programming and budgeting research and development research, development and innovation Scientific Research and Experimental Development Scientific Research Tax Credit small and medium-sized enterprises Technology Parnerships Canada World Trade Organization

2014-09-17 10:53:21

Acknowledgments

A book rarely results from the efforts of just one person, and this book is no exception. Many people and organizations contributed to its completion. I was assisted in my research by the many people in the Government of Canada who helped me gain access to documents in the National Archives and in the Privy Council Office. Various interviews with former state officials also contributed to my work. A special thanks is due to Stephen Clarkson, who provided documents concerning the Trudeau years that were invaluable to my conclusions. The Department of Political Science at York University was the place where the ideas in this book developed. It is a unique environment that perhaps is equalled nowhere else. I want to thank Leo Panitch, Greg Albo, and Ken McRoberts who were influential in my early experiences in the department. Leo’s incisive work, both as a supervisor of my thesis and as a scholar, has shaped my thinking in many ways. I also want to acknowledge the conversion program at York University under the CU P E Unit 2 collective agreement. This program, which provides part-time ­faculty with the opportunity to be converted to full-time, tenure-stream appointments, was integral to my ability to complete this book. Sabbaticals from York University in 2006–07 and 2012–13 were crucial in allowing me to formulate the ideas that form the core of the first part of the book, and to complete necessary revisions. I want to thank McGillQueen’s University Press for its commitment to this book, particularly Kyla Madden, who, as editor, saw its potential from the beginning. I want to acknowledge permission to republish material in chapters 2 and 3 that was originally published in Studies in Political Economy no. 85 (2010): 179–208 and no. 87 (2011): 143–72.

26410_Smardon.indd 13

2014-09-17 10:53:21

xiv Acknowledgments

I would like to acknowledge support from the Faculty of Liberal Arts and Professional Studies Minor Research Grants Fund and the Faculty of Liberal Arts and Professional Studies Publishing Subvention Grant, both of which assisted with the publishing costs of this book. The Carleton Library Series also provided financial support. Finally, I want to thank Jan Kainer. I can say with absolute certainty that without her infinite love and patience this book would never have been completed.

26410_Smardon.indd 14

2014-09-17 10:53:21

as l e e p at t h e sw it c h

26410_Smardon.indd 1

2014-09-17 10:53:21

26410_Smardon.indd 2

2014-09-17 10:53:21

I n t ro du c t io n

The Canadian Federal State and Domestic Technological Development

For two and a half decades after the Second World War, new domestic and international structures were developed that underlay a “Golden Age” of capital accumulation in the leading capitalist nations. Central to these new structures of accumulation was the creation of a variety of technological capacities. Under the impact of productivity missions sent from various countries to observe American industrial practices, the ­defeat of domestic opposition by workers to the introduction of more intensive Taylorist approaches to production, and a range of demandand-supply conditions that supported mass-production industries, the Fordist technological practices developed earlier in the twentieth century by US capital were established in various European economies.1 As I discuss in chapter 3, the extension of Fordism to Europe occurred as part of a wider restructuring of industry in which a range of new research and development (r&d) and original innovative capacities were developed in countries such as Britain, France, Sweden, and West Germany. A similar extension of technological capacities occurred in the Japanese economy. A cluster of factors – export-driven accumulation under the impetus of US support for Japanese exports to the US market, American purchases of Japanese supplies in the Korean War, and a particular set of class relations leading to the formation of company unions and a more restricted domestic market – brought about systematic changes in the domestic ability of Japanese capital to engage in original innovation.2 Also discussed in chapter 3 is the move of Japanese industry, gradually over the 1960s and then rapidly over the 1970s, away from its heavy reliance on technology transfers from other countries, particularly the United States.

26410_Smardon.indd 3

2014-09-17 10:53:21

4 Introduction

When combined with the intensive further development in the United States of industrial research and original innovation in the period after 1945 to the late 1960s – discussed in chapter 3 – the result was a significant change in the level of domestic technological capacities in various advanced economies and new international structures of competition marked by a more rapid pace of domestic innovation.3 Although, as Mowery notes, domestic structures of innovation and forms of “crossborder technology transfer” existed in the late nineteenth and early twentieth centuries, as well as in the period leading up to the Second World War, the breadth and intensity of domestic innovation and its connection with international processes of technology transfer increased considerably in the two and a half decades after the war.4 This trend only intensified with the unravelling of the Fordist “Golden Age.” The disruptions and crises that started in the latter part of the 1960s and continued through the 1970s led to the abandonment of the Keynesian welfare state doctrines that had informed policy-making in the advanced economies and to the development of a new global architecture of neoliberal development.5 Central to this architecture was an intensification of globalized production, trade, and investment as transnational corporations continued to restructure and develop their operations in various countries.6 The movement to greater global integration did not, however, remove the importance of nationally based institutions in managing domestic economies: forms of territorially based geopolitical competition were not eliminated, even though, as discussed in chapter 1, the manner in which national states managed their economies was very much informed by global political economic structures. As Lacher notes, “Capital is becoming increasingly globally integrated, though it remains enmeshed in national institutions, networks, cultures and traditions.”7 Neoliberal globalization was thus characterized by a dual movement encompassing both global forms of integration and national forms of differentiation. Within these new global structures of national competition, “supplyside strategies of ‘liberal-productivism’ for each firm, region or country” became dominant approaches to understanding how “to win a place in the competitive battle for world market shares and to solve the unemployment crisis.”8 In contrast to the focus on maintaining domestic aggregate demand in the previous Keynesian era, growth strategies were defined in terms of how to promote greater “international competitiveness” and “flexibility of industry” by improving capacity to move into

26410_Smardon.indd 4

2014-09-17 10:53:21



The Canadian Federal State and Domestic Technological Development 5

new areas of production or restructure existing operations. In all of these strategies, a key means of promoting “competitive strength” was the expansion of the technological capacities of industry. The ability of domestic capital to innovate and to integrate new technologies into production received increasing attention as a primary way of gaining a competitive edge within the world economy. Strategies of growth through technological innovation were presented as “win-win” outcomes in which all sectors of society would share in the gains by virtue of higher growth rates, higher employment, and higher incomes. As I review in chapter 1, there was a proliferation of literature promoting innovation-based strategies, particularly in approaches stressing national and regional systems of innovation, over the three decades from the 1980s to the first decade of the twenty-first century, and this emphasis was reflected in state policy. A range of programs and policies were developed in various countries within the Organization for Economic Co-operation and Development (oec d) with the goal of promoting the level of innovation in their domestic industries.9 A central part of these domestic innovation strategies involved the level of industrial r & d . Various states implemented incentive programs intended to boost the original science and technology capabilities within their domestic industries and to increase the rate of technological innovation in comparison with other economies. The methods employed by states to boost r & d and innovation varied over the decades, ranging from “strong government leadership of economic growth and development through extensive funding of technology programs during the 1960s and 1970s,” through a gradual process of withdrawal from direct involvement in structuring innovation processes in the 1980s, to a neoliberal posture in the 1990s in which “governments increasingly moved to assume enabling and regulatory roles, emphasizing freedom of choice for consumers in product and factor markets alike.”10 Throughout these different periods, the stress on innovation-based strategies and expanded domestic r & d to promote growth and development remained a constant. Even in the face of the austerity measures implemented by various states after the major increase in deficits brought on by the financial crisis of 2008–09, state funding for r &d and innovation remained stable or grew in a number of economies, leading to an overall increase within the oe c d area when government budget appropriations and outlays for r & d were measured as a proportion of gross domestic product (g dp) .11

26410_Smardon.indd 5

2014-09-17 10:53:21

6 Introduction

t he f e d e r a l s tat e a n d in dus tri al r&d: f ort y y e a rs o f ac t iv e p ro moti on A focus on promoting domestic r & d and innovation can certainly be seen in the Canadian case and, from the beginning, has been informed by the need to increase the commitment of Canadian industry to r& d and innovation in relation to industry in other economies. As I discuss in my review of federal r&d and innovation policy in part 2 of this book, it was during the 1960s that the federal state became committed to developing more domestically based research-intensive industries as a central policy objective and began to fund industrial r& d on a broad basis through direct grants and tax incentives. In chapters 5 through 7, I discuss how, under various governments including the Diefenbaker Conservatives, the Pearson Liberals, and the Trudeau Liberals, the federal state established several new programs: the Defence Industrial Research program; the Industrial Research Assistance Program, which was housed in the National Research Council (n rc); the Program for the Advancement of Industrial Technology; and the Industrial Research and Development Incentives Act program. The commitment of the federal state to expanding domestic sources of industrial innovation extended throughout the 1970s and grew larger in terms of overall cost under the Trudeau Liberals, as I review in chapter 9. On top of a restructured grant program, called the Enterprise Development Program, major new tax credits for industrial r& d were added in 1977. These incentives quickly grew to become the core source of federal support for r & d in Canadian industry. Another major federal initiative that was designed to promote expanded technological capabilities in the 1970s was the “make-or-buy” program, which was established in 1971 and focused on contracting out federal research needs to the private sector. r &d and innovation continued to receive a lot of attention in the 1980s and 1990s, and more initiatives were implemented at the federal level. As I discuss in chapter 11, under the Trudeau Liberals the Industrial Regional Development Program (created as part of the new department of Regional Industrial Expansion) replaced the Enterprise Development Program and led to expanded funding in the early to mid-1980s. The ill-fated Scientific Research Tax Credit led in the same period to more federal money being focused on tax incentives. In chapter 12, I discuss the initiatives undertaken by the Mulroney Tories after they came to power in 1984, including the InnovAction initiative in 1987, which led

26410_Smardon.indd 6

2014-09-17 10:53:21



The Canadian Federal State and Domestic Technological Development 7

to a new focus on “research networks” and to the Networks of Centres of Excellence program (involving networks of research connections between industry and the universities) in the latter part of the 1980s; and a new federal department called Industry, Science and Technology Canada, which was formed to manage the promotion of r& d and innovation in Canadian industry. The focus on promoting domestic technological capacities continued into the early 1990s, when the Mulroney Tories established the Prosperity Initiative. Reflecting a movement away from grant-based forms of assistance in the post–free trade period, grants and contributions to industry from the n rc and from the newly formed Industry, Science and Technology Canada declined in the latter part of the 1980s and remained at a lower level in the early 1990s. But they were compensated for by  ­rising levels of spending on r & d contracts with industry. The Conservatives also sustained the commitment of the federal state to r & d tax credits. Although the Tories eliminated the Scientific Research Tax Credit in the face of widespread abuse, they created the broader Scientific Research and Experimental Development tax credit, which, as discussed below, was among the most generous in the advanced capitalist world in the 1980s and the 1990s. The focus on r & d and innovation did not change when the Chrétien Liberals formed the government in 1993. In his budget speech in February 1994, Finance Minister Paul Martin focused on the need to promote greater domestic research and innovation as a central priority and stated, “Our challenge is to take existing resources and redirect them towards the creation of a national system of innovation.”12 In the same budget speech, an extensive policy review called the Science and Technology Review – another in a long line of federal reviews, starting with the Glassco Commission report in 1963 and continuing through to the Tories’ Prosperity Initiative in 1992 – was announced. After the review, the Chrétien Liberals established the Technology Partnerships Canada program in 1996. The program was created and received expanded funding, even though the Chrétien Liberals cut funding significantly in a number of other areas after the 1995 budget. Indeed, as I discuss in chapter 13, r & d grants and contributions to the business enterprise sector rebounded from their lower level in 1995–97 to levels higher than in the period before the 1995 budget. A range of other new initiatives built on the Tories’ earlier emphasis on research networks and reflected the increased influence of “systems of innovation” thinking. These included the “Partnerships in Knowledge”

26410_Smardon.indd 7

2014-09-17 10:53:21

8 Introduction

program; the Synergy Awards program; the creation of Industrial Research Chairs; Strategic Projects for pre-competitive research involving universities and industry; and Research Partnership Agreements among universities, industry, and government organizations. In addition to these initiatives were various sectoral strategies, such as the Canadian Biotechnology Strategy13 and “new program and spending initiatives” undertaken by the Chrétien Liberals in the latter part of the 1990s, ranging from the creation of the Canadian Foundation for Innovation, which provided funding for new research infrastructure projects in university and non-profit institutions, to expanded funding for existing programs, such as the Networks of Centres of Excellence. The extent of support provided by the Chrétien Liberals for research and innovation led one commentator to observe that “the combined impact of the four budgets from 1997 through 2000 represented one of the most significant investments in s& t [science and technology] spending in many decades.”14 Through these various programs and initiatives over a forty-year period, the federal state spent substantial sums of money in pursuit of the goal of expanded industrial r & d and greater domestic technological capabilities, particularly through the creation of r& d tax incentives that provided higher levels of funding than in other advanced capitalist countries. A study conducted in 1994 found that federal tax incentives were higher than in the United States, Japan, and Western Europe. When provincial incentives for r & d were added into the calculations, the study found that Canadian tax incentives were the highest in the world.15 These findings were unchanged from those of previous studies done in  the early and late 1980s.16 Canadian levels of assistance for r& d through the tax system in this period were thus unequalled in the world. Indeed, they were high enough to compensate for the very high levels of direct subsidization of r & d that existed in other countries. During the 1980s, with the development of the Scientific Research Tax Credit, the level of tax assistance for r & d more than compensated for the levels of direct subsidization of r & d that existed in other countries. As the Macdonald Commission pointed out in 1985, “Had the Scientific Research Tax Credit remained in force as originally established, the Canadian system would be generous by any standard.”17 Even without the srtc , the Canadian level of assistance was much closer to the levels provided in other countries if state-funded r& d contracts were viewed as less than a 100% subsidy.18 Because the state usually retains the rights to the technology developed under r & d contracts, the assistance provided through these contracts is not as beneficial as that provided by

26410_Smardon.indd 8

2014-09-17 10:53:21



The Canadian Federal State and Domestic Technological Development 9

subsidies or tax credits that give the company full property rights in the resulting technology. On this basis, the Macdonald Commission concluded that, if research contracts were not regarded as a full subsidy, the overall level of Canadian support for r& d compared quite favourably with other countries. This situation did not change in the 1990s. The Scientific Research and Experimental Development tax credit, along with provincial credits, provided higher levels of tax incentives than in other countries and was augmented by the major new investments made by the Chrétien Liberals.

can a d ia n in du s t r ia l r &d: achi e v in g a b r e a k t h ro ugh? The federal state was thus engaged from the early 1960s to the end of the 1990s in promoting a more innovative Canadian economy through substantial levels of financial support, and an upward movement in the relative position of Canadian industry seemed to occur, particularly in the 1980s and 1990s. Some commentators were led to argue that Canadian innovation had either reached a breakthrough or was on the verge of one. Reviewing the Canadian national system of innovation at the turn of the twenty-first century, Niosi claimed that, after an initial period of state-led encouragement of r& d and domestic innovation both through “horizontal policies,” such as tax incentives and grant programs, and the targeting of specific industries, the Canadian national system of innovation could, by the 1990s, sustain itself on “automatic pilot after an energy-intensive take-off”; that is, “only horizontal policies to keep incentives operative” were needed.19 De la Mothe echoed the claims of Niosi concerning the shifting position of Canadian industry when he stated, “It can be documented that since 1993 – indeed, the period probably began in 1986 – Canada’s relative innovative performance has improved across a wide variety of fronts … Despite still ranking fourteenth in the o e c d , Canada experienced the fastest growth in r & d intensity over the 1981–2000 period … In patenting, Canada achieved the fastest rate of growth in the G7 in terms of patent applications.”20 Finally, after reviewing the various initiatives by the Chrétien Liberals in the latter part of the 1990s, Wolfe stated that “many of the essential elements for the enhancement of Canada’s innovative capacity are currently in place.”21 It is ironic that, when these claims were being made, Canadian r& d and innovation was already beginning to move in the direction of

26410_Smardon.indd 9

2014-09-17 10:53:21

10 Introduction

significant decline. Over the first decade of the twenty-first century, there was a major deterioration in the relative position of Canadian industry with respect to its commitment to domestic r& d . Business enterprise expenditure on r & d (b e r d ) as a percentage of g d p declined consistently after 2000, from 1.15% (75% of the o e cd average) to 0.93% (58% of the o e c d average) in 2010.22 This returned the ratio to levels reached almost twenty years earlier, in 1993/94.23 The pattern was quite different in the wider o e c d region. From 2000 to 2008, the o e cd average increased in every year until the financial crisis of 2008/09 hit all of the oec d economies, leading to some decline between 2008 and 2010 (from 1.63% to 1.58%). However, the decline was not as great as in Canada (from 1.04% to 0.93%), which in combination with the earlier decreases led to a steep drop (from 75% to 58%) in the Canadian be rd :g d p ratio as a proportion of the o e c d average over the decade. Industry-financed gross domestic expenditure on r & d (ge r d ) as a percentage of g d p peaked at 1.03% in 2006 (72% of the oe c d average) and then declined rapidly to 0.84% in 2010 (58% of the o e c d average).24 In the latter part of the 2000s, b er d at 2005 prices and purchasing power parity declined absolutely by 15% between 2006 and 2010.25 Moreover, preliminary figures for 2011 showed further deterioration. The be rd :g d p ratio declined to 0.89% – and thus to levels previously reached in 1992 – while the industry-financed ge r d:gdp ratio fell to 0.81% and the level of b er d declined by another 1.76%.26 The falling commitment to r & d in the private sector pulled down the overall level of spending in the Canadian economy. After peaking at 2.04% in 2005, the ge r d :gdp ratio declined over the rest of the decade. Reporting in 2012 on the situation in Canada, the o e cd noted that “g er d in constant dollars has declined by 1.2% a year during the latter half of the past decade to usd 24 billion and 1.74% of g d p in 2011. It fell sharply in 2008 and again in 2010.”27 By 2011, the g er d:g dp ratio had moved back to the levels that existed in the latter part of the 1990s and was much closer to the levels that had historically existed in the Canadian economy. From the latter part of the 1960s to the mid-1990s, this ratio had “stubbornly remained around the 1.5% level.”28 The deterioration in the first decade of the twenty-first century, particularly as it gathered momentum in the last half of the decade, ­returned Canadian r & d as a proportion of the Canadian economy to levels marginally higher than this, while eliminating a large part of the gains made in earlier periods. Moreover, there was a growing gap

26410_Smardon.indd 10

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 11

between Canada and other oe c d economies. The o e cd average consistently increased over the 2000s, from 2.20% in 2000 to 2.41% in 2009, and did not follow the Canadian pattern of persistent decline after 2005. The limitations on r & d spending created by the economic crisis of 2008/09 led to some reduction in the g e rd :g d p ratio in the wider oec d area (from 2.41% to 2.38% between 2009 and 2010), but not to the same extent as in Canada. As a result, between 2005 and 2010, the g er d:g d p ratio in Canada declined from 91% of the o e cd average to 77%.29 The shifting relative position of Canadian industry in the first decade of the twenty-first century reflected the fact that, contrary to the views of various commentators on Canada’s innovation situation, the growth years of the 1980s and 1990s, and the new innovation policies, programs, and frameworks implemented at the federal level, were not successful in altering fundamental dimensions of Canada’s domestic ­ technological capacities. The growth experienced in specific areas in the 1980s and 1990s, such as with the expansion of Nortel and j d s Uniphase in information and communications technologies, masked underlying structural weaknesses and led to the mistaken impression that Canadian innovation had moved to a qualitatively different level. The underlying vulnerabilities in the system were quickly revealed, however, with the major fall in r & d spending when these core companies began to struggle after the dot.com meltdown in 2000. The Canadian system was neither extensive enough nor diversified enough to accommodate these losses without going into steep decline. At the centre of the inability of the Canadian economy to absorb the loss of key firms was a highly concentrated structure of r& d and innovation in which research-intensive companies were the exception rather than the rule in Canadian production. This resulted from a deeply ingrained pattern of development in which Canadian industry consistently relied on technological innovation in other countries, particularly the United States. As detailed in chapters 2 through 4 of part 1, Canadian industry followed a dependent model of technological development ­defined by limited levels of r & d within the Canadian subsidiaries of transnational capital, relatively few domestic networks focused on the development of new technologies, and consistently high levels of imports in key areas of capital-goods production. Contrary to Niosi’s claims concerning the “energy-intensive take-off” of the Canadian innovation system, the dependent model was not removed in the period from the 1960s to the 1990s. In the 1990s, when the

26410_Smardon.indd 11

2014-09-17 10:53:22

12 Introduction

Canadian innovation system was supposedly on “automatic pilot” and had attained self-sufficiency, clear differences remained in trade patterns, in the extent of reliance on imported technology, and in the structural composition of the Canadian economy, particularly in relation to the US economy. In their comparison of the Canadian and American economies in the 1990s, Howse and Chandler pointed to “startling” differences between the two with respect to both their share of exports in the “highintensity research and development” category and their reliance on domestically produced technology.30 These differences extended beyond the US economy to oecd member countries more generally, and were reinforced with the decline in r&d and innovation in the Canadian economy in the first decade of the twenty-first century. As noted in the concluding chapter, in that period several reviews of Canadian innovation pointed clearly to the relatively low capacities of Canadian industry in a variety of areas. One review (after comparing the relative situation of Canadian industry across a range of indicators, ­including business r&d intensity, triadic patent families, venture-capital ­investment as a proportion of gdp, and the proportion of value-added represented by high- and medium-high technology manufacturers) concluded that “our performance in innovation is stunningly poor. We rank 14th out of 17 countries, the fourth from the bottom.”31 Despite all of the federal support for r&d and innovation in the Canadian economy, Canadian industry in the first decade of the twenty-first century remained mired at the bottom of the advanced economies in its innovation capacities and, far from improving, continuously spent less on r&d as a proportion of gdp from 2001 to 2010. This decline was not reversed by the high level of federal support for r&d and innovation throughout the decade, particularly through r&d tax credits. As noted by Phillips and Castle, in 2009/10 the federal state provided, through its various programs, a quarter of all the funds spent on r&d in Canada. This funding was in addition to the Scientific Research and Experimental Development tax credit, which in 2008 “was the second most favourable among the G7 countries for large firms and the most favourable for small and medium sized firms, with implicit subsidies of about 18% and 32% respectively.”32

t he s t ru c t u r a l bas is o f t echnologi cal de p e n d e n cy The major reversals in industrial r & d and the failure of the Canadian federal state to promote a restructuring of Canadian capital that could

26410_Smardon.indd 12

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 13

sustain a higher level of r & d and innovation in the Canadian economy raise a central question: Why has this dimension of the Canadian economy not fundamentally changed despite all of the attention that has been focused on it by the federal state for so many years? Answering this question provides the central organizing principle of this book, which is structured into two main parts. In part 1, I detail the entrenchment of dependent technological development as a core feature of the Canadian economy. I begin my discussion of this process in chapter 2, where I examine how the particular approach of Canadian industry to dependent accumulation was first established. I argue that this dimension grew out of the development of the Canadian economy, specifically the region of southern Ontario, as the earliest example of the systematic extension of Fordism outside of the United States. Canadian industry was at the forefront of the advanced economies in incorporating the institutions and practices of American Fordism by the end of the first decade of the twentieth century. A specific approach to economic development was introduced that was supported by both manufacturing capital and the Canadian state, and built on the emerging forms of US Fordist capital in the latter part of the nineteenth and early twentieth centuries. As part of this framework of development, the expansion of the Canadian economy was tied closely to sustaining flows of technology from the heartland of Fordist development in the United States. From the beginning, technological dependency was incorporated into the specific version of Canadian Fordism and its associated ways of structuring growth. In chapter 3, I argue that this approach continued with the reconstruction of another version of Fordism after the Second World War. In what Jenson has termed “permeable Fordism,” all of the previous core dimensions – resource extraction, imports of capital and technologies from the United States, infrastructural investments by the federal state, “government oversight of the actions of firms and unions in private collective bargaining relations,” and the growth of mass consumption – were combined once again in the Canadian economy.33 As part of this discussion, I provide a comparison of the Canadian version of Fordism with the versions established in other advanced economies in the United States, Europe, and Japan. This comparison illustrates the specific dimensions of the Canadian path of development and shows how structures of dependent technological development distinguished the Canadian economy from other advanced economies. I also argue in chapter 3 that the particular approach to accumulation represented by Canadian Fordism came under strain with the extension

26410_Smardon.indd 13

2014-09-17 10:53:22

14 Introduction

of this form of capital to other advanced economies outside of the United States. Even though the second period of Canadian Fordism maintained a high relative standing in terms of various economic indicators such as productivity and per capita income, important tensions emerged that undermined the integrity of this model of development and were related to the reliance of the Canadian manufacturing sector on a domestic market that could not support Fordist accumulation in the same way as in the past – especially in relation to the United States and to the Japanese and leading European economies, which were undergoing their own USled Fordist development after the war. I discuss the Auto Pact as a moment of transition between the earlier model of Fordist development, with its reliance on the Canadian domestic market, and a new form of accumulation that began to dominate the approach of manufacturing capital and the federal state, which was premised on expanding scales of production and productivity by moving in the direction of continental rationalization. In chapter 4, I discuss how the initial trends established with the Auto Pact were amplified in the period of post-Fordist neoliberal restructuring leading up to and after the 1987 Canada–United States Free Trade Agreement. The movement toward continental rationalization accelerated, and Canadian manufacturers abandoned their historic focus on the Canadian domestic market along with their support for tariff protection and domestically centred accumulation strategies. A pattern of accumulation was established that combined staple production and technological dependency with new forms of export-oriented manufacturing growth. I argue that, while removing the problems associated with relying on the more restricted Canadian domestic market, the new growth strategy created another set of difficulties. For various reasons, the new pattern of Canadian accumulation within a restructured regional space was no longer able to maintain the same relative levels of manufacturing development in research-intensive areas vis-à-vis other advanced economies. The processes of transfer and catching up that underlay the ability of Canadian industry to follow American developments and establish Canadian versions of production in new areas of research-intensive production in the earlier periods were attenuated. In addition, the Canadian economy was opened up to international trade without having the innovative capacities to compete in key areas of research-intensive production. The result was an overall deterioration in the condition of manufacturing in the Canadian economy, leading to declining export shares, rising import shares, and falling relative productivity.

26410_Smardon.indd 14

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 15

I also argue that, because of differences in how Fordism was integrated into the European and Japanese economies, the impact of postFordist regional restructuring did not lead to the same reduced presence of research-intensive advanced manufacturing in those economies. A different basis of Canadian accumulation, still marked by a lack of development of original technological capacities, was established in which dependent relations and advanced manufacturing development were no longer articulated in the same way, and levels of productivity growth in the Canadian economy were more limited in comparison with those in other advanced economies. Part 1 of this book thus provides an overview of the integral role of dependent technological development in three key periods of Canadian manufacturing, beginning in the first decade of the twentieth century and ending in the first decade of the twenty-first, which defined the ­approach of Canadian industry to domestic r& d and innovation. It provides an alternative view of the political economy of Canadian development, which is a complex story in itself, but one that is essential to an understanding of the evolution of federal industrial r& d and innovation policy. Part 2 builds on the foundation established in part 1 by discussing the political economic situation as it related to the formation of policy over the different eras of development, and by referencing the broader structures of Fordism and post-Fordism where that is appropriate. This part of the book explores the history of federal r& d policy over the forty-year period from 1960 to 2000. Comprising eight chapters (5–13), it starts with the approach of the Diefenbaker Conservatives in the 1960s to promoting r & d and innovation and concludes with the policies and initiatives implemented by the Chrétien Liberals in the 1990s. By organizing my analysis of r & d policy along these lines, I depart from standard innovation policy analysis, which typically discusses policy developments without examining how they are connected to ­wider political-economic structures or the broad sweep of change across historical eras. As a result, key determinants of the content and impact of policies, and of what can be accomplished by state-based initiatives, have until now been largely ignored.

t he “ t r a n s f o r m at iv e s tate,” the glassco f ram e wo r k , a n d p r ivat e capi tal Certainly, the pattern of interpretation described above has characterized the large literature that exists concerning the role of the state in

26410_Smardon.indd 15

2014-09-17 10:53:22

16 Introduction

technological development, particularly in scholarly work on national and regional systems of innovation. In chapter 1, I examine this literature, which places at the centre of its analysis the institutional capacities of states to either lead or support transformative projects that restructure the technological capabilities of domestic industries. Major emphasis is placed by writers informed by this theoretical framework on the development of a “transformative state” that is able to implement a range of policies leading to new forms of state–business linkages, intensified training and skills development, higher levels of investment by the state and business in r & d, and new forms of cooperation or alliances among different firms in the development of new technologies. All of this discussion has been conducted in a manner that neglects the crucial role of political economic structures in determining the feasibility of transformative projects. As I show in the Canadian case, the transformation of the economy involved much more than the existence of a state with a commitment to establishing appropriate forms of technological infrastructure. It also involved challenging the organization of the underlying structures of the Canadian economy that lay at the heart of low domestic technological capacities, and that dated back to the early part of the twentieth century. In my discussion of federal r & d policy, I draw out the implications of this situation for the extensive series of programs that were implemented by the federal state. I argue that the transformative intentions of the various programs came up directly against, first, the legacy of Canadian Fordism that had governed Canadian industrialization since the first decade of the twentieth century and, then, with the development of neoliberal restructuring in the 1970s, the new form of post-Fordist accumulation that was established. These various measures were not capable of shifting the underlying dynamic of accumulation in the Canadian manufacturing sector which, growing out of an historical process of technological borrowing, was systematically structured away from the kind of innovation-based growth that was being promoted by the federal state, especially when the programs were framed by the assumptions of what I term the “Glassco Framework.” Within this framework, which was first enunciated in the Glassco Commission report on government organization in the early 1960s, the basic assumption of federal programs was that a transformation in the commitment of the private sector to creating domestic technological capacities could be generated by increasing various subsidies to private capital, shifting a greater proportion of federal science expenditures to

26410_Smardon.indd 16

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 17

the private sector, or creating greater instances where federal scientific activities were coordinated with private sector r& d – while leaving untouched the specific context of dependent technological development that governed how private capital organized r& d and original innovation in the Canadian economy. A number of initiatives were followed, including increased marketization of federal research expenditures; greater subsidies to specific firms on a project-by-project basis; expanded tax incentives, which gave firms maximum control over how they spent the federal incentive money; and new programs promoting “partnerships” between private capital and federal science institutions. These various incentives, while providing considerable benefits to Canadian industry, were not sufficient to outweigh the fundamental forces structuring the commitment of private capital to r& d and original innovation in the Canadian economy. Specific projects were supported, and particular industries, such as the defence and space industries, were linked in a major way with federal assistance programs, but there was no broad process of transformation. I argue that the context of dependent technological development had another important impact: it was central in shaping the response of private capital to the various incentive programs and informing the views of the representatives of private capital concerning what should be done. Because of the historical legacy of dependent technological development, Canadian domestic capital did not, for the most part, have a substantial connection to the domestic creation of new technologies. With specific exceptions, such as Nortel and Research in Motion, it avoided entering fields of manufacturing production involving heavy new expenditures on innovative capacities. As a result, domestic capital in the manufacturing sector was not oriented toward using the various federal incentive programs as a means of building industries that rested on novel technologies. There was also little support within domestic capital for federal strategies that moved outside the Glassco Framework and altered the dominant position of American transnational capital in research-­ intensive areas, or were focused on forms of state-led restructuring of the organizational context of innovation in the Canadian economy. Instead, the various recommendations submitted to the federal state by the representatives of private capital concerning the development of the incentive programs, and the positions taken by these representatives in the advisory task forces created to assist the federal state in the formation of policy, argued for the approach identified as the Glassco Framework. The primary thrust of their position was to call for measures

26410_Smardon.indd 17

2014-09-17 10:53:22

18 Introduction

that would expand the financial resources available to firms – whether by receiving greater federal subsidies for specific r& d projects, gaining more research money by shifting a greater proportion of federal science expenditures to the private sector, or reducing the amount of taxes paid to the federal state by means of higher tax deductions and credits for a given level of r & d expenditures – without challenging the dominant ways of organizing accumulation in the Canadian manufacturing sector. The dependent framework of technological development and its associated set of class interests, particularly the role and position of American transnational capital in research-intensive industries, were systematically not confronted in this policy discourse.

f e de ra l p o l icy a n d t e c h nologi cal de p e n d e n cy The strength of the Glassco Framework was reinforced by the way in which it fit with the dominant approach of the federal state to industrial development. Strategies promoting greater domestic technological capacities that moved outside the Glassco model and targeted major aspects of the organizational context of private sector decision-making went against the grain of key federal commitments in the area of industrial development. An open-door policy toward American transnational investment was a central part of how economic growth was constructed in the Canadian economy in the two Fordist periods, and state policy relied explicitly on a model of development that did not emphasize ­domestically controlled sources of technology creation. In the period of neoliberal restructuring, these trends continued, as there was increased emphasis on entering new trade agreements to create an environment attractive to foreign investment. Core aspects of potential state-led ­industrial strategies that targeted the domestic operations of American capital and focused state support on building domestic technological capacities were prohibited under these agreements. Within this policy context at the federal level, the principles of the Glassco approach that emphasized forms of cooperation and coordination with private capital and worked within the structural parameters governing Canadian technological dependency were highly attractive. The strength of the commitment to the Glassco approach was illustrated at different moments when more targeted strategies that were concerned with the structural context of r& d and innovation were put forward in the federal state. I discuss several instances when strategies of

26410_Smardon.indd 18

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 19

this sort were raised, including Walter Gordon’s proposals as Minister of Finance in the period 1963–65 and the recommendations of the Watkins Report in the latter part of the 1960s; the proposals made in the early 1970s by both Herb Gray, as Minister of Industry, Trade and Commerce, and Maurice Lamontagne, who headed the Senate Special Committee on Science Policy; and the further set of state-led restructuring proposals that were presented by Herb Gray in the early 1980s. These various initiatives all ran up against the hostility of business organizations in the private sector and the opposition of business Liberals in various Liberal governments, who regarded them as deeply disruptive to the structures of private investment and growth in the Canadian economy. I argue that, in these instances, the various strategies could not be carried forward in the face of the determined opposition generated by senior state officials and business organizations. A key moment of opposition occurred in the latter part of the 1960s and early 1970s, when momentum was building both within the federal state and on the part of various social forces, including women’s organizations, peace groups, worker unions, and nationalist movements, for challenging the role of American capital in the Canadian social formation. This momentum was broken for a number of reasons, including the economic restructuring that took hold in Canada, as in other advanced capitalist economies in the latter part of the 1970s and early 1980s; the growing strength of neoliberal views that rejected nationalist forms of state intervention; the limitations of the “social vision” of unions and the Left in Canada, arising from their continuing commitment to the Keynesian welfare state institutions of the postwar Fordist accommodation; and growing divisions within various organizations that had supported social change at the beginning of the decade. All of these changes undermined the level of popular support for more structuralist initiatives. The moment of challenge passed away without leading to any fundamental alteration in the approach of Canadian industry to r&d and innovation. I also argue that the particular circumstances in which new programs were developed provided a further source of support for the principles of the Glassco approach. Key moments of r & d policy formation occurred when economic growth was flagging in the Canadian economy and tensions existed in the relationship between the state and the “business community.” The new incentive programs provided a means of building better relations with private-sector interests by acting upon their recommendations and, at the same time, allowed the federal ­government to argue that it was promoting growth by creating a more

26410_Smardon.indd 19

2014-09-17 10:53:22

20 Introduction

innovative and competitive economy. This dynamic was particularly appealing to state officials in the lead-up to election campaigns, when there was a concerted attempt to build bridges with business interests and an urgent need to increase public support. Several instances of r& d policy formation followed this pattern: the initial creation of r& d incentive programs by the Diefenbaker Conservatives in the early 1960s; the ending of federal grants under the Industrial Research and Development Incentives Act program and the creation of major new tax incentives by the Trudeau Liberals in the latter part of the 1970s; and the development of the Scientific Research Tax Credit by the Trudeau Liberals in 1983– 84. At these moments, federal policy was not based on an analysis of the structural sources of low r & d and original innovation in the Canadian economy, or indeed on any broad analysis of the r& d situation in Canada, but was shaped in response to a set of specific political needs that moved the federal state in a direction that supported the principles of the Glassco approach. I argue that the dominance of this framework was further sustained by the malleability of its underlying assumptions, which allowed its principles to continue to be followed even though the same themes of relatively low domestic r & d and innovation continued to resurface at different times. There was always some deficiency that could be pointed to and some modification that could be supported as a way of improving the operations of the programs – incentives could be converted from grants to tax incentives, as occurred in the latter part of the 1970s, or tax credits could be expanded, as was done in the latter part of the 1970s and early 1980s. Another frequent explanation was that more coordinated approaches involving the state and private capital should be pursued, or that federal programs were too “firm-centric” and should be changed in favour of an “innovation systems” approach, as was done by the Mulroney Conservatives with the creation of Industry, Science and Technology Canada in 1987 and the “strategic technologies” initiatives in the latter part of the 1980s, and by the Chretien Liberals in 1996 with the Technology Partnerships Canada program. Or it could be argued that federal science programs and expenditures were inefficiently organized and emphasized forms of research remote from the stages of development and commercialization, and should be systematically reviewed with an eye to either shifting resources away from the state to private-sector research organizations or making federal research organizations more closely aligned with private-sector research needs – an argument that was made repeatedly throughout the entire period from 1960 to 2000.

26410_Smardon.indd 20

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 21

With each change of government, the relatively low level of industrial r &d and original innovation was noted and another round of new initiatives in this area was launched, but the point was never reached where a sustainable transformation in the domestic technological capacities of Canadian industry could be established. In a striking way, federal programs and policies returned again and again to the same themes of cooperation and coordination with private capital, and to the need to promote greater commercially oriented innovation through expanded federal assistance based on grants or tax incentives or state-business partnerships, without challenging the underlying context of the forces that governed technological dependency in the Canadian economy. In this regard, although the strategic thrust by the federal state to increase the amount of collaboration with private capital did have important consequences, particularly in leading to a growing emphasis in federal research organizations and in the universities on attracting private money to finance research projects, the minimal focus of Canadian private capital on creating domestically centred R&D and innovation capacities limited the extent to which this project could be realized.34 I note in the concluding chapter that the level of collaboration by private capital in both the public and private areas was one of the lowest in the o e cd area in the first decade of the twenty-first century. However, the failure on the part of the federal state to realize its projects, both in the specific area of research collaboration and in promoting domestic technological capacities more generally, did not lead to any fundamental questioning of its approach. Not only was the Glassco model reasserted over time, but it became ever more dominant as more targeted strategies calling for changes in the structural sources of technological dependency became less and less politically relevant, particularly with the new environment of policy-making in the 1980s and 1990s that was reflected in and reinforced by both regional and global trade agreements. In the 1990s, arguments for a structural transformation of the technological capacities in Canadian industry were still present in the federal state, such as the studies done for Industry Canada by Lipsey and Carlaw and by Gibbons that called for targeted forms of state support for specific domestically based industries through state-led industrial strategies.35 These studies were done as part of a wider initiative, taking shape in 1993 and led by Industry Canada, that continued the long tradition of federal reviews and strategies in the area of innovation policy. This policy evolved over time in a more structuralist direction until “its maturation in the 2002 Innovation White Paper,” where it became more

26410_Smardon.indd 21

2014-09-17 10:53:22

22 Introduction

concerned with “embedding innovative behaviour within the larger fabric of society,” although it never took on the interventionist dimensions of earlier approaches, such as the two sets of proposals put forward by Herb Gray in the early 1970s and early 1980s.36

t ran s f o r m in g r & d ? f e d e r al poli cy at t he t u r n o f t h e c e n t u ry However, as was illustrated by the minimal impact of the White Paper after its announcement, transformative policy initiatives, even of the more limited sort put forward by Industry Canada, were deeply out of step with the dominant policy norms in the federal state and the prevailing context of r & d and innovation in the Canadian economy at the turn of the twenty-first century. On the basis of the new innovation strategy, the Chrétien Liberals argued for much higher levels of r& d and original innovation in Canadian industry, but the entire exercise had an unreal quality, particularly with respect to the goal it set for r& d expansion. As noted by Kinder, in that strategy “the Chrétien government … made a commitment to improve Canada’s performance in research and development (r & d) so that by 2010 it [would] rank among the top five countries in the world in that policy area. Given that r& d performance by the private sector accounts for over half of all r& d conducted in Canada, this goal, regardless of its merits, is not likely to be reached unless massive new investments are made by business and industry.”37 The figures on r & d that I presented earlier show how unrealistic that goal was. Far from moving to the top five countries in 2010, r& d in Canadian industry was in steep decline, and the gap between Canada and other o e cd countries was growing. The failure of the innovation strategy to achieve its goal for r& d expansion illustrated the impasse in federal r& d and innovation policy – one that was firmly entrenched by the turn of the twenty-first century. By 2000, any meaningful tensions between structuralist initiatives and the Glassco Framework were gone as the forces, both inside and outside the state, supporting federal policies and programs that took seriously the sources of dependent technological development were weakened over the 1980s and 1990s – a process that I discuss in chapters 12 and 13. As the orthodox position of the Glassco Framework became ever more dominant, the underlying parameters governing federal r& d and innovation policy were solidified. The period 1960–2000 was thus

26410_Smardon.indd 22

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 23

crucial, as it was during this time that key tensions in the r& d and innovation policy area were settled in favour of the Glassco approach. By the turn of the twenty-first century, the contradictions of federal r & d and innovation policy were also at their height. Based on the growth in r & d in the Canadian economy, particularly over the 1990s, and the seeming effectiveness of the innovation systems framework that shaped the major investments made by the Chrétien Liberals in that decade, there was a sense that Canadian r& d and innovation had turned a corner. As one science policy commentator stated, “The impressive progress made by the government since 1993 was not a mere happenstance or a case of serendipity. It shows strong evidence of a framework that was, over time, utilized by major departments and agencies. Although the results have not been uniform in all areas of policy, the innovation systems framework has proven to be both robust analytically and effective in action.”38 As noted earlier, this was precisely when Canadian r & d and innovation began to go into decline as key Canadian firms struggled after the bursting of the dot.com bubble in 2000. At the very point that federal support for r& d and innovation had gone through another period of expansion, and federal policies and programs seemed to be working by following the prescriptions of the innovation systems approach, a process of significant decline was getting underway. Even at the best of times, the underlying sources of relatively low r& d and innovation in the Canadian economy were not changed. Over the ensuing decade, the Martin Liberals and Harper Conservatives continued to support r & d and innovation, although with less belief on the part of the Tories concerning the potential for transforming this aspect of the Canadian economy. Optimism began to erode. By the end of the decade – reflecting the deterioration that had occurred – the same concerns were being raised about the poor innovation performance of Canadian industry. Writing in 2010, Phillips and Castle stated that “most individuals and groups involved in science, technology and innovation policy agree that Canada can and should do better in terms of innovation. Many studies critical of Canadian science and technology innovation have focused on different problems within the innovation system and its implications for productivity, global competitiveness, drivers of commercialization, and the productivity gap.”39 Discussion had moved away from how Canada’s national system of innovation had reached a point of “take-off,” to once again emphasizing the weaknesses of Canadian innovation and how that was having a

26410_Smardon.indd 23

2014-09-17 10:53:22

24 Introduction

negative impact on productivity and economic growth. But this situation was a direct outgrowth of the vulnerabilities that had remained in Canada’s national system of innovation ten years earlier and were not removed by the “Golden Age” of Canadian innovation. I thus focus my study on the period 1960 to 2000 both because it covers the crucial moments when core tensions between structuralist initiatives and the Glassco Framework continued to define the r& d and innovation policy area or, by the end of the period, were resolved by the dominance of the Glassco approach, and because it covers the underlying sources of low r & d and innovation that continued to exist in Canadian industry even in the period when they supposedly had undergone major change. The r & d situation as it proceeded into the first decade of the twenty-first century was defined by this period, and it underpinned the marked deterioration by the end of the decade. In the concluding chapter, I briefly review r& d policy developments up to 2010 as part of my discussion of the federal policy impasse in the current period. I will also discuss the issue of why low r&d and innovation is important as part of a broader consideration of the manufacturing situation in Canada and the deterioration in manufacturing value-added and employment. I present evidence concerning the serious decline in manufacturing over the first decade of the twenty-first century and discuss why that decline has important ramifications for the broader services sector. I argue that any claims that manufacturing can safely be ignored in favour of the larger services sector, in terms of a knowledgebased services sector, is misguided. I also argue that the provinces do not represent a viable alternative to the federal state in the area of promoting r & d and innovation because of their grounding in political economic contexts that support similar Glassco-based approaches, the level of resources that the provinces can devote to this purpose, and because of key policy instruments that are controlled at the federal level.

di s t i nc t iv e c o n t r ib u t io n s My explanation of both the development and the impact of federal innovation policies has several distinctive dimensions. I depart significantly from the established ways of understanding the political economy of Canadian economic development. Within the standard interpretations of this process, there has been no recognition of the status of Canada as an early adopter of the frameworks and social relations of American Fordism, or of the way in which Canadian development occurred within

26410_Smardon.indd 24

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 25

shifting configurations of the international organization of Fordism, or of the unique combination of dependency and advanced development in Canadian Fordist industrialization. Explanations of development have either argued for a process of limited industrial development quite unlike the American one, or have located Canadian development within a nationally contained framework similar to the path followed in other advanced economies, including the United States.40 In addition, because the specific version of Canadian Fordism in the two earlier periods has not been adequately theorized, there has been insufficient analysis of how previous patterns of economic development came apart in the neoliberal era and new patterns of post-Fordist accumulation were established that rested on a different basis. I thus provide a different framework for understanding the actions of the state and manufacturing capital in defining Canadian economic development. I also provide a different explanation for how and why technological dependency was established and sustained in the Canadian economy. There are a number of further contributions. I provide an alternative conception of the political economy of Canadian economic development and integrate that framework into an explanation of federal state policy concerning industrial r & d and original innovation over the forty-year period from 1960 to 2000. Surprisingly, given the extensive commitment of the federal state to promoting greater domestic r& d and original innovation in this period, there has been no detailed attempt to examine the history of policy development in this area, and no attempt to link that history to the broader context of the Canadian political economy. There has also been little recognition of the importance attached by the federal state to promoting innovation-based exports. The various programs created by the federal state represented a significant long-term commitment to altering the export orientation of Canadian industry by changing its original technological capacities – a commitment that has been neglected within the Canadian political economy literature, which has traditionally placed the promotion of resource staples at the centre of federal export policies.41 I also review in chapter 1 the established views of the role of the state in technological development – ranging across “national systems of innovation” theory, institutionalist theory and the “developmental state” literature – and argue that these various analytical models have neglected the important role of political economic structures in determining both the outcomes of industrial strategies and the feasibility of state-led transformative projects in the area of technological development. I make

26410_Smardon.indd 25

2014-09-17 10:53:22

26 Introduction

a similar argument about the “varieties of capitalism” literature, which, while shifting the focus to firm-level structures, also does not sufficiently consider the role of political economic structures in shaping the paths followed in different capitalist economies. These various literatures have not been addressed under one review, even though they share similar themes. Finally, my work addresses the feasibility of innovation-based industri­ al transformations in Canada – a question that has generally not been considered in the various proposals to pursue a Canadian “progressive competitiveness” strategy.42 A lot of attention has been focused on the characteristics of an economy with greater domestic technological capacities; much less attention has been paid to the specific social and political contexts in which this type of economy would be developed and sustained. Various writers have called for state action to reorient the Canadian economy toward higher levels of indigenous r& d , a greater emphasis on technological innovation, and faster rates of productivity growth.43 But they have consistently neglected crucially important questions surrounding the feasibility of their strategies within the dominant structures of the Canadian political economy. In my discussion of the ability of the state to implement high tech strategies, I depart from traditional approaches to analyzing public policy that distinguish between three sources of policy development: ideological positions, such as those deriving from class-based interests; ideas that inform the parameters of debate about policy issues; and institutions that shape both the capacities to act in particular ways and the manner in which approaches to policy are formed. As stated by one analyst of public policy: “Three mutually interacting influences shape and constrain policy formation: powerful ideas, powerful institutions, and powerful interests act as gatekeepers to the process of agenda setting.”44 In my arguments, social forces, particularly class interests, are given central importance, while ideas and institutions are seen more as outcomes of how these dominant forces shaped, in different historical periods, the hegemonic ideas or approaches to understanding the relationship of the state to the economy, and the institutional structures that were put into place to regulate and manage capitalist accumulation. This is particularly the case in how I explain the ongoing debate over the role of the state in “picking winners” and engaging in state-directed strategies for transforming economies. In part 2, I claim that this contending set of policy ideas, and the institutional forms of organization that shaped the debate, were defined by the contexts of political economic forces in which they developed.

26410_Smardon.indd 26

2014-09-17 10:53:22



The Canadian Federal State and Domestic Technological Development 27

I also argue that a core source of support in Canada for ideas that denied the efficacy of the state in “picking winners” and opposed statedirected industrial strategies in r & d and innovation derived from the class-based forces in the Canadian economy and state, which saw no purpose in using state institutions in this way as a means of developing greater domestic technological capacities, and reinforced those federal programs and policies that supported the Glassco Framework. Policy debates over the appropriate role of the state, and about institutional evolution, were thus part of one overall process connecting class-based interests with ideas and institutions. I return several times to different historical conjunctures in which policy choices, or ideas, and the development of institutions, were informed by the structure of class-based representation of forces within the economy and state. Policy debates and the framework of institutional capacities were not separate sources of policy formation that came together with “powerful interests” to determine policy, but were deeply informed themselves by the wider context of social forces. A final note is necessary before moving on to the analysis offered in this book. In my arguments I break with explanatory frameworks that emphasize the institutional prerequisites for industrial strategies and argue that in Canada “the combined institutional legacy of the Westminister model of parliamentary government and federalism … fosters a weak state tradition and discourages anticipatory policy making.”45 In my work, I found little evidence to suggest that the federal division of powers or the Westminister model led to a lack of institutional capacities or an inability to formulate forward-looking transformative proposals in the area of r&d and innovation. Concerning the impact of federalism, I found no instances where the federal division of powers was regarded as a significant impediment to transformative policies and programs. The debates were conducted in terms of the appropriate role of the state in shaping production and investment in the Canadian economy, not in terms of the institutional obstacles created by the constitutional role of the federal state. Significant capacities within the federal state and moments of transformative politics also existed that did not support the notion of a “weak state tradition.” In chapter 2, I discuss the transformation of the Canadian economy in a Fordist direction that was supported in a number of ways by federal state actions. In both parts of the book, I refer to institutional capacities that existed in the federal state – such as the research capacities created during the Second World War, the connections between the federal state and the manufacturing of jet aircraft in the 1950s, and the

26410_Smardon.indd 27

2014-09-17 10:53:22

28 Introduction

awareness in the federal bureaucracy at various moments of the structural limitations on r & d and innovation in the Canadian economy – all of which could have been mobilized to support a transformative strategy. I also discuss key moments when transformative projects were generated from within the federal state, at the cabinet level with the Gray proposals in the 1970s and 1980s, and at the parliamentary level with the detailed proposals put forward by the Senate Special Committee on Science Policy in the 1970s. The latter set of proposals called for, among other things, an Office of Industrial Reorganization in the Department of Industry, Trade and Commerce. In these instances, there was no shortage of policy proposals focused on the need for a structural transformation of the organization of domestic r & d and innovation and, within the federal state, capacities existed that could have been used to support them. What was missing was a political economic context of forces that could carry them forward in the face of the commitment of class interests, as those were organized in the state and the economy in different historical periods, to dependent technological development. Although there were no guarantees that the strategies, if implemented, would have been successful, there was also no way that a transformation in the domestic technological capacities of Canadian industry could be accomplished without them. It was this dynamic that was crucial. The failures I point to are thus not failures that derived simply from an inadequate set of institutional arrangements or an inadequate set of policy measures located at the level of the state. Rather, they derived from a set of social relations that concentrated social power in the hands of a select group of capitalist economic and political elites in a way that ensured there would be little meaningful progress in establishing greater domestic technological capacities. In the next chapter, I will elaborate on this theme by discussing how this underlying principle of capitalist power has been missed in the broad literature that has attempted to explain the course of technological development under capitalism.

26410_Smardon.indd 28

2014-09-17 10:53:22

1 Promoting Domestic Technological Capacities: State Strategies and Social Antagonisms In the introduction, I discussed the growing emphasis placed on domestic technological capacities within the advanced capitalist world in the postwar “Golden Age,” and how that trend gathered momentum with the advent of neoliberal restructuring, particularly in the 1980s and 1990s. Reflecting a growing awareness of the relationship between domestic technological capacities and export performance in competitive processes, an extensive literature developed that placed domestic technological capacities at the centre of economic growth and development.1 With the growing popularity of national systems of innovation (n s i ) theory over the 1990s, this literature rapidly increased. Fagerberg and Sapprasert observe that, “in the early 1960s scholarly publications with innovation in the title were few and far between. But from then on scholarly interest in the subject gradually increased. Moreover, there is evidence of a trend break in the early 1990s, after which scholarly works on innovation increased at an even faster pace than before.”2 A core theme of this literature is that economies differ considerably in how their domestic innovation is organized and sustained, and in the ability of their domestic industries to expand internationally on the basis of developing and incorporating new scientific knowledge. Different structures of organization of innovation both within and between industries and firms, and between the state and industries, are examined in this literature as crucial determinants of the divergent abilities of economies to generate exports and promote growth more generally through domestic innovation. In this chapter, I will review different approaches that have emphasized this theme and focused on the state as a transformative agent either in a leadership role or as a supporter and enabler of new innovation

26410_Smardon.indd 29

2014-09-17 10:53:22

30

Theorizing Technological Change and Development

networks. Over the last three decades, analysis in four different bodies of literature – n si theory, institutionalist state theory, developmental state theory, and varieties of capitalism theory – has stressed the institutional capacities of states to promote change in this area. Even though these literatures share common themes, and although the same writers have sometimes made contributions in more than one body of work, they have not been discussed as part of one review.

e xp l ain in g t e c h n o l o g ic a l development: 1. n at io n a l sys t e m s o f in novati on theory Certainly, a focus on the organization of domestic innovation as the key to export performance and economic growth can be seen in the important body of theory that emerged in the latter part of the 1980s and ­focused on the concept of a national system of innovation. Freeman, whose book on Japan (1987) has been identified as the earliest attempt to explicitly use the n s i concept,3 defines it as “the network of institutions in the public and private sectors whose activities and interactions initiate, import, modify, and diffuse new technologies.”4 Lundvall provides a similar definition when he states that “a system of innovation is constituted by elements and relationships which interact in the production, diffusion and use of new, and economically useful, knowledge and … a national system encompasses elements and relationships, either located within or rooted inside the borders of a nation-state.”5 From these definitions it can be seen that the n s i approach from its earliest beginnings is concerned with specifying the institutional relationships that promote the application and use of new technologies in different national contexts.6 The n s i approach was extended through the rapidly expanding literature on regional innovation systems and on innovation clusters that emerged in the early part of the 1990s, and was a core emphasis of scholarly work after 2000.7 A central theme in this work, taken from the earlier n si institutionalism, is the importance of “networks of learning” and the way in which innovation processes are crucially affected by an appropriate context of connections in which “firms become ‘embedded’ in close vertical and horizontal relationships with nearby firms, and within a rich, thick local-institutional matrix that supports and facilitates the production (private and socially organized), transmission and propagation of new technologies (product and process).”8 Some argue that there is a tension between national systems of innovation – deriving

26410_Smardon.indd 30

2014-09-17 10:53:22



State Strategies and Social Antagonisms 31

from the broad contexts in which they operate and the wide variety of potential institutions and factors that influence them – and the more ­locally condensed and focused nature of regional systems, while others argue that the two operate in conjunction and should not be seen as contradictory.9 But, regardless of the extent to which the two systems can be regarded as complementary in practice, the overall approach to understanding the origin and genesis of innovation as a networked phenomenon remains the same. Three different theoretical sources are crucial to the n s i perspective, and to the regionally based approaches that grew out of it; these sources concern the nature of economic growth and accumulation under capitalism. The first is based on Schumpeter’s work and his emphasis on technological change as the key to economic development. Certain writers, such as Freeman and Soete, explicitly derive their theory of economic growth from Schumpeter. They argue that the most intensive periods of capitalist growth are based on the sweeping and wide-ranging changes in investment patterns that are initiated by technological innovation – especially when a cluster of technological change forms the basis for a new “techno-economic paradigm.”10 Others writing from this perspective are not as explicit in tying their explanations to Schumpeter, but their underlying assumption remains the same: the introduction and diffusion of new technologies is taken as the starting-point in understanding economic growth and development. The second source of theoretical inspiration is drawn from a critique of neoclassical approaches to understanding international trade. Over the 1960s, 1970s, and 1980s, increasing attention was paid to the different rates of export growth achieved by technologically advanced economies.11 These findings brought into question the traditional HeckscherOhlin-Samuelson model of international trade, in which relative factor endowments (that is, the relative proportions of land, labour, and capital in economies) are seen as crucial in underpinning the trade performance of different countries. In opposition to traditional trade theory, and in response to the varying export performance of different economies, the proponents of the nsi approach argue that the general equilibrium processes of international trade adjustment are weak and that different technological capabilities can lead to long-run differences in exports, imports, and income per capita.12 A “technology gap” can emerge in which certain economies, because of their greater institutional ability to develop and use new technologies, are able to generate a higher rate of technological innovation and a higher rate of export growth.

26410_Smardon.indd 31

2014-09-17 10:53:22

32

Theorizing Technological Change and Development

A final theoretical source of influence is a body of work focusing on the role of interactive processes between industries and firms that form the basis for technological learning and accumulation.13 These interactive processes are seen as socially and institutionally embedded, and thus the process of technological change is seen as an endogenous creation. It is not exogenously given, as is often the case in neoclassical models of economic activity. Within the nsi approach, these endogenous processes of technological learning are crucial in allowing economies to make effective use of research and development (r& d ) and other forms of technological knowledge. The development of the n s i approach thus draws heavily on a Schumpeterian understanding of capitalist growth and development. Although, as Fagerberg and Sapprasert note, the work of Freeman and other nsi theorists departed from Schumpeter in examining the different abilities of national economies to organize and implement technological innovation, and despite the fact that their conclusions had a more explicit policy focus, the nsi perspective nevertheless shares a common substantive approach, as it brings together three strands of economic theory that share a focus on technological innovation as the central determinant of trade patterns, employment levels, and capital investments.14 In turn, the Schumpeterian framework of analysis that underlies this approach has led to a particular set of policy recommendations for states. A central goal of public policy with respect to R& D is to determine the most important growth areas of technological development and to encourage greater indigenous technological linkages within priority sectors of the economy. These linkages, in turn, will further the level of technological learning and make more effective use of an expanded level of r &d resources. Public policy also has to encourage greater training and skills development so that the labour force has the requisite abilities to work in the new areas of technological innovation. Finally, the state has to provide the infrastructural support that is needed by the industries being established in the new areas of production and trade.15 According to this framework, by pursuing these policies the state provides the key means of overcoming the “institutional inertia” of previous “techno-economic paradigms.” Lastres summarizes this position when she states, “The role of government in stimulating the renewal or breaking-up of those mature development blocks locked into old technological trajectories, and in supporting the formation of new ones is particularly important. This role is not circumscribed to the promotion of particular r & d projects. It has also to deal with all sorts of policies

26410_Smardon.indd 32

2014-09-17 10:53:23



State Strategies and Social Antagonisms 33

concerning education and training, diffusion, adaptation and effective exploitation of new technologies, incentives or investments in new equipment and various other aspects related to the general economic and social environment.”16 In short, the n s i approach calls for the development of a transformative state with the capacity to move the economy in a fundamental way toward technological advancement.

e xp l a in in g t e c h n o l o g ical development: 2. i n s t it u t io n a l is t t h e ory Starting in the 1980s, a parallel development identified as the institutionalist approach emerged in the area of state theory and came to the same conclusions about the role of the state in technological development. These same conclusions were reached even though they were ­developed by theorists working in different disciplinary contexts. n s i theory was developed primarily by economists and departed from neoclassical models of technological change and international trade, whereas institutionalist theory was developed by state theorists who moved away from what they saw as an excessive reliance on “society-centred” explanations for the actions of the state. In contrast to Marxist and pluralist approaches, these writers argue for a view of the state in which it has a more autonomous role to play both in determining its own actions and in shaping the development of social interests.17 The development of institutionalist theory led to a characteristic approach to understanding how the state acts as an autonomous force both on its own and in relation to society. Central to this body of theory is the notion that different organizational relationships structure the interaction of individuals within the state, as well as interactions between the state and society. Peter Hall states in his application of institutionalist theory to industrial policy that “the emphasis is on the relational character of institutions; that is to say, the way they structure the interaction of individuals. In this sense it is the organizational qualities of institutions that are being emphasized; and the term ‘organization’ will be used … as a virtual synonym for ‘institution’.”18 For the institutionalists, organizational structures are crucial in shaping and constraining the outcomes of political activity. The focus of their analysis is on the different capacities of various institutional or organizational arrangements to achieve specific outcomes. The relevant political actors in each policy area have different institutional or organizational capacities, which then form the basis for different types of state–society

26410_Smardon.indd 33

2014-09-17 10:53:23

34

Theorizing Technological Change and Development

relations – ranging from a policy network in which an autonomous state is able to impose its preferences on society, to a policy network in which a highly dependent state merely reflects a variety of externally generated demands and pressures.19 In the area of industrial policy, emphasis is placed by the institutionalists on the question of whether there is an autonomous state with the institutional capacity to draw together a plan of action for the economy and apply it in a coherent fashion over the conflicting objections and demands of organizations within society. Crucial to the existence of this autonomous “state tradition” is the development of bureaucratic centralization, along with a group of state elites with an independent source of legitimacy and an “administrative culture” in which the state “embodies a sense of the public interest that is more than the sum of private interests in society.”20 States that are characterized by these conditions are capable of implementing “anticipatory” industrial policies based on a long-run perspective of the economy and are concerned with underlying structures of industrial organization. These anticipatory industrial policies are contrasted with “reactive” policies that are based more on short-term responses to industrial problems and do not involve a focus on underlying industrial structures. Atkinson and Coleman claim that “it is hard to over-emphasize the importance of state tradition for the conduct of industrial policy. Where state tradition is weak, state institutions often reflect the interests of the strongest organizational forces in society. Much of the state’s apparatus is devoted to transmitting and responding to these demands. And because these demands are conflicting, and state structure under these ­circumstances is generally inchoate, industrial policy is typically a rather confusing amalgam of reactive policy initiatives. Policy innovation is inhibited by the absence of political and bureaucratic leadership.”21 Similarly, Hall observes in his comparison of state policies in Britain and France that “the form of state involvement in the economy that emerged in Britain was fundamentally different from that in France, and it had different consequences. In particular it was characterized more by the growth of bargaining between the government and the two sides of industry than by the growth of unilateral state control … The experiences of other nations, such as France and Japan, suggest that in order to tackle underlying productivity problems Britain would have needed an industrial policy based less on voluntarism and more on rationalization enforced directly by the state.”22 There are clear parallels between this view of the state and that in the n si approach. Both perspectives emphasize a transformative state that

26410_Smardon.indd 34

2014-09-17 10:53:23



State Strategies and Social Antagonisms 35

can autonomously implement policies leading to far-reaching, long-term changes in the economy. In the nsi conception, technocratic elites design and implement policies that are necessary to adapt the economy to the needs of the latest “techno-economic paradigm,” while in the institutionalist view the same technocratic elites implement “anticipatory” industrial policies capable of transforming the economy and moving it onto a path of higher economic growth and more intensive technological change. The nsi approach can thus be characterized as a form of institutionalism in which there is a major focus on the transformative role of interventionist states.

e xp l a in in g t e c h n o l o g ical development: 3. t h e d e v e l o p m e n ta l s tate A third theoretical framework, which emerged during the 1990s, was concerned with the ability of East Asian states, such as Taiwan, Singapore, and South Korea, to rapidly transform their economies and develop greater exports on the basis of a higher level of technological capabilities.23 At the centre of the explanation that this framework provides for the transformative capacity of the East Asian states is the notion of the “developmental state.” This is a state that is able to establish institutional connections of “embedded autonomy,” a term developed by Evans,24 or “governed interdependence,” a term developed by Weiss.25 Both terms refer to a situation in which a set of trained and knowledgeable bureaucratic elites have the capacity to transform the direction of the economy by establishing forms of “disciplined support” and a dense set of connections with relevant industry associations and firms.26 Referring to the Japanese Ministry of International Trade and Industry (miti) as the paradigmatic example of embedded autonomy, Evans makes the following claims: If miti were not an exceptionally competent, cohesive organization, it could not participate in external networks in the way that it does. If mi t i were not “autonomous” in the sense of being capable of independently formulating its own goals and able to count on those who work within it then it would have little to offer the private sector. miti’s “relative autonomy” is what allows it to ­address the “collective action” problems of private capital, helping capital as a whole to reach solutions that would be hard to attain otherwise … This “embedded autonomy”… is the key to the developmental state’s

26410_Smardon.indd 35

2014-09-17 10:53:23

36

Theorizing Technological Change and Development

effectiveness. “Embedded autonomy” combines Weberian bureaucratic insulation with intense connection to the surrounding social structure.27 These developmental linkages between the state and industry are particularly important when technology-intensive forms of production are being supported. Referring to the transformative capacity of states, Weiss argues that “in recently and maturely industrialized settings, the latter depends very largely on the nature of the state’s domestic linkages. Indeed, it would appear that the more industrialized the country and the more sophisticated the technology becomes, the more critical the policy linkages between economic bureaucracy and [the] industrial sector for transformative success.”28 The similarities of the developmental state literature with both the n si and institutionalist approaches are strong. Indeed, Evans and Weiss ­emphasize the direct connection of their work to institutionalist forms of analysis.29 As with the other perspectives, developmental state theory places great emphasis on an autonomous state that, on the basis of an elite group of trained administrators, is able to move the economy, through the creation of appropriate linkages with business, into a project of industrial and technological transformation. The state is conceived of as a set of institutions that lies above the economy and possesses the autonomous ability to issue directives concerning the future course of technological development. Because of the emphasis on putting sufficient state capacities into place, the analysis of what stops the developmental process from occurring is couched in terms of a failure to have the right kinds of state institutions for policy creation. And, for the ­developmental state theorists, one of the key sources of this failure is ­neoliberalism. In this literature, attention is focused on “market fundamentalism” and its manifestations within the institutions of the World Bank and the International Monetary Fund, as well as in various academic approaches, as the central obstacle to the application of the model.30 Neoliberalism is also pointed to as a key reason for the failure to sustain transformative state capacities once they have been established. For example, it is argued that the declining ability of the South Korean state to sustain its transformative capacity after the latter part of the 1980s stemmed from the growing role of US-trained economists within the Korean state who were “preaching state retreat from economic affairs.”31 The concentration on state-led transformation leads to a similar view of neoliberalism in the other approaches. Writers are skeptical of the

26410_Smardon.indd 36

2014-09-17 10:53:23



State Strategies and Social Antagonisms 37

claims made by neoliberals that the way to promote economic growth and transformation is simply through a retrenchment of the state in the area of industrial policy. Applying the n s i formulations in the Canadian context in the 1990s, Wolfe lamented the growing reliance of the Canadian state on forms of free trade and trade-based adjustment: “The gradual dismantling of the fiscal instruments that reasserted the primacy of the federal government in the postwar era has left it with little sense of purpose or rationale in the areas of fiscal or social policy. The enactment of the Free Trade Agreement in 1988 signalled a definite shift in favour of a trade-led adjustment and industrial policy, further constraining its scope for policy leadership, with little regard for the impact of this move on the country’s manufacturing core.”32 Similarly, reacting to the economic crisis in 2008–09, and the failures of neoliberalism “in various parts of the world, such as Latin America or the transition countries of central and eastern Europe,” Etzkowitz and Ranga argue for a “triplehelix” approach to organizing innovation on the basis of institutional relations between “government, industry and university actors” and claim that “a well-thought out ‘picking winners’ approach may be beneficial to fostering innovation and long-term growth.”33 There is thus a characteristic answer from the proponents of all three of the institutionalist perspectives that emerged over the 1980s and 1990s to the question of what role the state should play in technological development. In their view, the state should embark on a systematic process of “institution-building,” which will lead to improved processes of innovation in the economy.34 Further work has elaborated on this theme by proposing ever more complex ways in which innovation processes can be organized. One of the most recent studies in this area stresses forms of “path interdependence” that would lead to innovations possessing the qualities of “uncertainty,” “surprise,” and “entropy” and – in contrast to the emphasis on specialization that is often present in “cluster strategies” – would be based on forms of collaboration and learning across different sectors in a context of “transversality.”35 By this construction, the state takes on multiple roles, of which, in relation to promoting innovation, a particularly effective one is the creation of “platforms” that enable firms across different sectors to investigate possibilities that would not have occurred to them in their more isolated or divided settings. For Cooke, “firms and other complex organizations require assistance in rebalancing asymmetric information by learning, preferably in ‘living laboratory’ contexts, from each other’s experiences but aided by transversality intermediaries.”36

26410_Smardon.indd 37

2014-09-17 10:53:23

38

Theorizing Technological Change and Development

Over time, however, some disagreement has developed concerning the extent to which this process of “institution-building” involves state-­ directed processes. In relation to recent proposals to establish a centralized European Institute of Technology to coordinate various regional areas of innovation specialization, Cooke notes that “against a key insight of complexity theory, namely that complex systems have ‘no global controller’ but evolve through the interaction of adaptive systems that largely self-manage, ‘smart specialisation’ clings to the belief that a ­‘hierocracy’ can control and direct r di for a continent from a position of elevated omniscience.”37 There was also a shift after 2000, based on the work of Malerba, to a focus on “sectoral systems of innovation” with different mixtures of institutional elements not primarily located within national boundaries.38 The various differences within the approach do not, however, remove a set of common limitations, which are shared with another recent form of institutionalist analysis: varieties of capitalism theory. I will review the latter analysis before discussing the problems common to all of this institutionalist literature.

e xp l ain in g t e c h n o l o g ic a l development: 4. vari e t ie s o f c a p ita l is m In the first decade of the twenty-first century, a new perspective emerged that centres on relations, at the level of the firm, concerning banking and finance, training, and labour relations and places emphasis on the “comparative institutional advantages” within coordinated market economies (c mes) as opposed to liberal market economies (l m e s).39 State-directed processes are given less attention in this model than in the three frameworks described earlier. Indeed, as argued in the core writings of Hall and Soskice, neoliberalism, and its reliance on “sharpening” the role of markets, is promoted as a way of furthering the development of l m e s. They state that “because of the bluntness of the instruments available to states and the importance of markets to these economies, deregulation is often the most effective way to improve coordination in l m e s.”40 In c mes, it is acknowledged that the important role of non-market coordinating institutions – such as various collaborative arrangements in training and apprenticeship programs, r & d arrangements between firms, and the role of social policy in supporting the long-term commitment and loyalty of workers to specific employers – undermines the effectiveness of deregulation and the “sharpening” of the market. But these nonmarket arrangements are delimited by their specific contributions to the

26410_Smardon.indd 38

2014-09-17 10:53:23



State Strategies and Social Antagonisms 39

comparative advantages of c me s at the institutional level. Hall and Soskice are clear when they note, “We have identified some specific types of inter- and intra-firm relations and supporting institutions that we associate with effective firm performance. There are other ‘non-market’ institutions in many economies that simply generate economic rents or detract from economic efficiency.”41 Organized around a dichotomy between l m e s and cm e s, the focus in  the varieties of capitalism approach on “comparative institutional ­advantages” leads to a divergence in how technological development is  explained in comparison with the other forms of institutionalism. Drawing on the distinction between l m e s and cm e s, Hall and Soskice argue that the former is oriented to “radical innovation,” while the latter is oriented toward “incremental innovation.”42 They claim that because of the greater reliance on non-market forms of coordination and denser networks of intra- and inter-firm relations in cm e s – which create various commitments and loyalties to existing employment and company relationships – the organization of institutional advantages in these economies supports incremental changes to existing technological trajectories. In contrast, the reliance on market forms of coordination in lmes supports radical forms of innovation because firms can more easily and abruptly change course – shifting employment and investment patterns to support the new forms of technology and production. These arguments by Hall and Soskice concerning technological development have not received the same attention as other dimensions of their work in the areas of gender relations, social policy, redistributive politics, labour markets, and wage bargaining.43 However, they represent an integral part of the varieties of capitalism approach both in terms of their conclusions concerning the public policy implications of comparative institutional advantages and their broader specification of how l m e s and c me s are supposed to hold together. An evaluation of their arguments concerning technological change and innovation thus sheds light on issues that go to the heart of their analysis. There are several problems with Hall and Soskice’s conceptualization of technological development and innovation. First, fitting technological development into a duality of ideal types (cm e versus l m e ) leads to an ahistorical stretching of categories that does not capture the diversity of experiences within the different categories of countries. It also leads to a certain mechanical application of derived principles, thus distorting the actual experiences of different political economies. This inability to theorize the diversity of capitalist experience is certainly apparent in the

26410_Smardon.indd 39

2014-09-17 10:53:23

40

Theorizing Technological Change and Development

Canadian case. As I will show in this book, there was a narrow capacity for “radical innovation” in Canada, even though its political economy relied heavily on market-based forms of coordination, and the capacity for engaging in radical forms of innovation in fact decreased as reliance on market-based processes increased. The lack of correlation between market-based coordination and the type of innovation enabled by the Canadian economy points to the problems involved in trying to use the duality between c me s and l me s, and related arguments pertaining to comparative institutional advantage, as a basis for understanding the concrete histories of innovation in a diverse array of political economies. The category of market-based coordination encompasses a wide variety of situations concerning innovation and cannot be used on its own to specify what type of innovation will be pursued. In Canada, core sectors of innovation were governed by the reliance on technology transfer in the Canadian subsidiaries of foreign (primarily US) transnational capital and by licensing agreements between Canadian-controlled firms and foreign (again, primarily US) capital – a form of organization that was not challenged by the Canadian state because of its commitment to marketbased coordination through unrestricted foreign investment to develop the Canadian economy. In this instance, market-based coordination was associated with little development of radical innovation. A similar lack of fit is apparent in other aspects of Hall and Soskice’s arguments. Contrary to their assertion that cm e s are more oriented toward innovation based on non-market forms of coordination, while lmes concentrate on forms of innovation organized through separate market relations, the importance of research collaboration grew systematically in various market-oriented economies, particularly the United States, and was a core aspect of their innovation structures in the last decades of the twentieth century. In the American case, starting in the 1970s, research collaborations developed along several dimensions on the part of US capital – with various alliances among US firms, foreign transnationals, and US universities as well as international forms of coordination between US transnationals and research organizations in other countries – in response to changes in anti-trust legislation in the United States that allowed for greater joint-production ventures, the deep role of military-sponsored research, the greater interdependence of technological developments in the United States and other advanced economies, the expanding cost of engaging in r& d projects, and the increasing need to integrate technological change on a variety of fronts at the same time.44

26410_Smardon.indd 40

2014-09-17 10:53:23



State Strategies and Social Antagonisms 41

On the last point, Mowery and Teece note in relation to developments in the period from 1980 to the mid-1990s that “firms in food processing and pharmaceuticals, for example, confronted the challenges of biotechnology; telecommunications and computer technologies virtually merged; and advanced materials increased their importance in a broad range of manufacturing industries.”45 These trends were reinforced by the growth of innovation “networks of learning” – discussed in relation to nsi analy­ sis – in various advanced economies across different industrial sectors that were informed by combinations of market and non-market relationships and were centred on local and international networks of knowledge generation and communication.46 A core element of innovation in market-oriented economies was thus not based on contractual relationships that could be entered into or that ended relatively easily. Innovation commitments were embedded in long-term relationships of non-market coordination, and forms of “tacit” knowledge, which were not regulated through formalized market exchanges. Firms engaged in collaborative forms of research, and the degree of collaboration increased over time for a number of reasons that cut across the various advanced economies; these factors led to an organization of innovation that moved against the bifurcated pattern claimed by Hall and Soskice. Any attempt to understand technological development as “radical” or “incremental” based on a market / non-market dichotomy across capitalist economies runs up against this reality. It also runs up against the growing role of states in coordinating international rather than nationally focused institutions. As noted below, the role of states cannot be defined solely in terms of supporting internal forms of development and particular types of technological change in national political economies. The role of states in constituting global political economic development must be considered as well. Contrary to the varieties of capitalism arguments, states have not operated solely within national economies to promote distinct forms of technological development, but have actively supported a growing internationalization of technological change across economies by agreeing to free trade agreements that codify patent protections in various economies, and by managing broad processes of international finance that underpin global movements of capital and the creation of technological alliances across sectors and regions of the global economy. The attempt to fit processes of technological development into nationally defined containers in the varieties of capitalism arguments is undermined by the core problem that innovation processes in different economies cannot be adequately

26410_Smardon.indd 41

2014-09-17 10:53:23

42

Theorizing Technological Change and Development

considered in frameworks that recognize only distinct and separate forms of territorially based capitalist institutional logic. In this respect, the other forms of institutionalism differ from varieties of capitalism analysis. Because, in these theoretical models, technological development is not fitted into the c m e –l m e duality, there is no attempt to make broad arguments linking the type of innovation (radical / incremental) to this bifurcation. Particularly with respect to the expansion of work on regional networks and innovation clusters, the different intersections of market / non-market institutions in innovation processes in various capitalist economies are examined, and this dimension is distinct from the varieties of capitalism approach.47 Although, as argued below, this work, like the varieties of capitalism perspective, is limited by a focus on institutional frameworks abstracted from both the social relations of capitalism and the geopolitical relations of states, it is nevertheless less restricted in its presentation of the concrete ways of organizing innovation in different institutional frameworks. There is also greater recognition of the role of state intervention in supporting innovation both at the lower scales of the firm, locality, and region and at the larger scales of the “techno-economic paradigm.” This dimension disappears from the varieties of capitalism approach, with its claims concerning the benefits of deregulation in enhancing the “comparative institutional advantages” of l me s.

i n s t i t u t io n a l is m a n d e q ui li bri um outcomes Although they differ in their emphasis on state-directed change and their conceptualization of technological development, the various institutionalist approaches to understanding technological change that have been reviewed here share a common way of viewing social and political processes. In all of these perspectives, change is considered, in a way that parallels neoclassical economic theory, as a movement from one equilibrium position to another within a framework that, while acknowledging different social interests, is still characterized by a harmonious unity. Hall and Soskice provide a clear illustration of this point when they state, in relation to coordinated market economies, that “the equilibrium outcomes on which actors coordinate have been unsettled by economic shocks many times in the past. In each case, new equilibria have been found through processes of negotiation and compromise. The process of adjustment may well entail a period of conflict and suboptimal outcomes, as each side tests the power and resolve of the other. But the

26410_Smardon.indd 42

2014-09-17 10:53:23



State Strategies and Social Antagonisms 43

presence of institutions that entrench the power of the actors, whether employers and trade unions, give them strong incentives to cooperate with each other, and the availability of deliberative institutions facilitates coordination.”48 This emphasis on equilibrium outcomes follows the same focus as in the other three forms of institutionalist analysis as they emphasize harmonious frameworks of technological change characterized by configurations of “networks of learning,” “techno-­economic paradigms,” or developmental state strategies. In each case, it is acknowledged that although equilibriums are not permanent and will be disrupted, new equilibrium outcomes at higher levels of economic development can be reached through appropriate state policies and the construction of institutions with the right combination of market incentives, “community” incentives supporting trust and coordination, or platforms supporting “tranversality.”49 This approach raises several issues. First, the inequalities of capitalist arrangements – the ownership of the means of production by a specific class with all of its prerogatives in making investments and determining the allocation of capital, establishing the parameters of workplace organization, shaping social relations in ever more commodified directions, and expanding profitability by exploiting labour – are emptied out of the analysis in favour of equal bargaining between “actors” or shifting configurations of cooperative relations that promote “learning.”50 The emptying-out of social relations extends to the gendered and racialized divisions in capitalist society in which class relations and the reproduction of labour power are directly shaped by a shifting set of both domestic and international forces.51 Second, as part of this neglect of unequal relations within capitalism, social arrangements are depicted as constantly informed by the construction of equilibrium paths that are capable of balancing the interests of different institutional “actors” or social interests, rather than by a contingent articulation of antagonisms that, while leading to periods of relative stability, is never able to remove the fundamental oppositions and tensions involved in social relations. Although forms of class organization and “cross-class coalitions” are acknowledged and discussed, particularly in the varieties of capitalism literature, they are based on the actions of labour and capital as they meet one another in different institutional arrangements that create alternative “incentive structures” for supporting coordination or competition.52 The role of the state within this perspective is to promote the achievement of these balanced outcomes, either through explicit forms of direction or by providing supportive or enabling contexts for these

26410_Smardon.indd 43

2014-09-17 10:53:23

44

Theorizing Technological Change and Development

arrangements. The insertion of the state into supporting domestic social relations with all of its capitalist prerogatives is missing from the analysis. Finally, the global dimensions of state relations are neglected as state policies are considered within nationally contained formations that are separate from international structures of political economic development and contestation. This separation, along with the failure to consider domestic relations of capitalist power, leads to a restricted view of the relationship of the state to globalization. Globalization appears as an external set of forces to which states can respond in different ways: by engaging in various forms of institution-building, leading in the direction of new “techno-economic paradigms” or more vibrant “networks of learning,” as in the nsi approach; by developing more effective linkages between the state and private sector firms in broad, long-term developmental processes, as in the developmental state perspective; or by implementing appropriate policies supporting the comparative institutional advantages in specific national economies, as in the varieties of capitalism framework. In this work, international and national forces are integrated, but only in a limited way. Globalization is viewed through the prism of a common external framework of competitive pressures to which various national states respond in different ways to maximize their domestic advantages.53 But this approach to viewing the role of the state is highly questionable. States are embedded in domestic and international structures of capitalist power that do not allow them to simply manage their economies to find the best solutions for their domestic populations.

p ol i t ic a l s t ru g g l e a n d t echnologi cal change The problems posed by innovation-based transformative strategies illustrate this point. The transformation of economies with low domestic technological capabilities involves major changes in economic and political structures; as new relationships are created to integrate r&d more closely into the domestic production activities of industry and new networks among firms and between the state, various research institutions and private capital must also be created. In implementing these changes, a number of challenges to existing social relationships will be necessary – particularly if, as in the Canadian case, the creation of a transformed set of domestic technological capacities means that the domestic operations of foreign capital must be altered in fundamental ways, and entrenched ways of organizing economic growth, around which constellations of

26410_Smardon.indd 44

2014-09-17 10:53:23



State Strategies and Social Antagonisms 45

state and capital interests have been constructed, must be changed. In these instances, the process of technological restructuring cannot be understood simply as a process of policy development and implementation involving new technological linkages and networks and improved forms of education and training, or a process of establishing an appropriate set of developmental state apparatuses, or a process of creating different “institutional complementarities” in the areas of finance, investment, and labour relations; it also involves challenging a range of established interests both domestically and internationally – and there is a real possibility that a political and social basis for engaging in this struggle does not exist. I argue that the Canadian case is an example of this situation. Any transformation of the domestic technological capacities of Canadian industry meant that core relationships that underpinned dependent technological development would have to be taken apart. Transformative strategies in the Canadian context had to confront this reality and could not occur without major forms of struggle as longstanding practices – such as the allocation of responsibility for r& d and innovation by US transnational capital to its Canadian subsidiaries – and the entrenched position of domestic capital in areas of production without substantial commitments to domestic technological capacities, were confronted. Moreover, the ongoing development of federal r& d and innovation policies was consistently defined, in the announcement of policy changes in the lead-up to election campaigns or through ongoing reviews of ­federal programs, in ways that were tightly connected to the market-­ oriented preferences of Canadian manufacturing capital that had evolved through successive periods of development in the Canadian economy. Shifting the orientation of the Canadian economy was thus a very tall order – one impossible to accomplish without a confrontation with established patterns in the state and in the economy. As I discuss in the book, even though there were moments when a transformative project of this type was asserted at the federal level, the social and political basis for these projects was never sufficient to make them a reality. This dimension of considering change in terms of political struggle is entirely missing from the ns i , institutionalist, developmental state, and varieties of capitalism literature. The transformation of economies is portrayed as a confrontation with institutional structures that were part of old “techno-economic paradigms;” or as a result of “shocks” that disturb the workings of a particular way of organizing firm-level arrangements; or as a struggle against outmoded policy approaches that

26410_Smardon.indd 45

2014-09-17 10:53:23

46

Theorizing Technological Change and Development

hold economies back from higher levels of innovation and growth. In all of this, there is no discussion of change in terms of political power and the way in which dominant sets of interests must be disrupted in order to effect transformations in capitalist institutional structures. In this regard, the international dimensions of state relations and political power need to be emphasized, especially as they were particularly important in structuring Canadian r & d and innovation policies. As I discuss in part 1, throughout the three key periods of manufacturing expansion in the Canadian economy – the first period of Fordist growth in the early twentieth century, the period of Fordist reconstruction after 1945, and the period of neoliberal restructuring that started in the 1970s – Canadian accumulation was premised on particular relations between Canadian domestic capital and US transnational capital and was constructed around particular distributions of international investment and trade. In considering new strategies to transform the domestic technological capacities of Canadian industry, the federal state was not simply choosing between different ways of organizing the national economy, but was considering directions that would alter its international posture concerning the regulation and treatment of transnational capital within its national boundaries, and would go against well-established international relations that structured the approach to promoting processes of economic growth and accumulation. In part 2, in both chapters 7 and 9, I examine how the consideration by the federal state of transformative projects concerning domestic technological capacities explicitly took into account the relations between the Canadian and American states, and the freedom of manoeuvre the federal state had concerning the operations of US transnational capital in Canada. In these moments, federal policy-makers were considering their policy options within an international set of relations that very much constrained what could be done. This international set of relations and its role in conditioning domestic structures of accumulation is not adequately considered if international power relations are excluded from the analysis. Furthermore, the federal state was enmeshed in supporting broader international structures of liberal capitalist arrangements that had direct ramifications for national strategies concerning industrial transformation. This was particularly the case with the intensification of neoliberal globalization over the 1980s, 1990s, and 2000s and the US- led institutions that supported this process. As I discuss in chapters 10 and 11, the new trade agreements – the Canada–United States Free Trade Agreement,

26410_Smardon.indd 46

2014-09-17 10:53:23



State Strategies and Social Antagonisms 47

the North American Free Trade Agreement, and the creation of the World Trade Organization – and the commitments made by the federal state in those agreements, directly shaped the kinds of industrial strategies that were followed in trying to change the orientation of Canadian industry to r & d and original innovation. The global movement to neoliberalism, and the role of states in supporting this movement, created a well-­ developed set of mechanisms – including “national treatment” clauses and prohibitions on “trade-distorting” practices in various trade agreements – for disciplining states to follow particular types of policies. These mechanisms were part and parcel of the broader restructuring of social relations on a global scale that was reinforcing capitalist structures of accumulation and power, and could not be understood simply as the outcome of various nationally based strategies for promoting technological innovation or “institutional complementarities.”54 As Panitch and Gindin note in relation to the role of the state in the current era of neoliberal globalization and its architecture of globally integrated production, finance, and trade networks, “the varieties of capitalism approach argues that one of the ways in which states can successfully cope is by paying special attention to adapting or restoring, under the new conditions created by globalization, the national institutional structures making for economic coordination. What is obscured here is the extent to which globalization is a development not external to states, but internal to them.”55 The view of the state in nsi , institutionalist, developmental state, and varieties of capitalism theory rests on the notion that it is the idealized “repository of community values and societal needs.”56 The state presides over national formations and moves them in appropriate directions to new equilibrium positions, whether through state-led transformative strategies or by supporting different configurations of institutional relations among the state, research institutions, and firms. But this view of the state neglects the way in which it exists as part of a contradictory social ensemble, both domestically and internationally, that defines what is possible at a given moment in time and shapes the ongoing actions of the state in relation to the organization of the economy. As I discuss in the book, this condition certainly characterized the attempt by the federal state to increase the level of domestic technological capacities in the Canadian economy. The possibility of engaging in transformative projects was constantly subject to the prevailing organization and articulation of opposed dominant social interests, both inside and outside the state, and how that related to the strength of social forces in favour of

26410_Smardon.indd 47

2014-09-17 10:53:23

48

Theorizing Technological Change and Development

these projects in different periods. A theoretical framework that avoids any discussion of this dynamic by viewing capitalism as the result of equilibrium processes generated by institutional frameworks isolated from unequal social relations cannot adequately comprehend this process. However, before embarking on a discussion of the failure of the federal state to change the structures of dependent technological development in the Canadian economy, it is first necessary to examine why Canadian industry has continued to rely so heavily on the technological capabilities of industries in other countries, and to examine the consequences of this dependence for the performance of the Canadian economy. The reliance of Canadian manufacturing on technology developed in other countries, particularly the United States, and the relatively low level of investment in domestic technological capabilities represent deeply embedded practices within the Canadian economy – practices that have only been reinforced over time as the initial form of industrial development in Canada shaped the strategies followed by Canadian manufacturers in subsequent periods. The various structural sources of low domestic r & d and original innovation, and the way in which they have led to a self-reinforcing dynamic, now becomes the focus of our attention.

26410_Smardon.indd 48

2014-09-17 10:53:23

pa rt o n e Permutations of Dependent Technological Development: From Early Fordism to Neoliberal Restructuring

26410_Smardon.indd 49

2014-09-17 10:53:23

26410_Smardon.indd 50

2014-09-17 10:53:23

2 Entrenching Dependent Technological Development: Canadian Fordism in the Early Twentieth Century In this part of the book, I examine the long history of dependent technological development in Canada. In this chapter, I discuss its beginnings in the rapid process of Fordist industrialization in Canadian manufacturing in the early twentieth century. The ability of Canadian industry to incorporate the practices of American Fordism and expand on the basis of extensive technological borrowing established a pattern that has continued to this day. This was not foreordained. The reliance on technological borrowing had to be reconstructed in successive periods and was the product of the specific strategies followed by Canadian manufacturing capital as it responded to the shifting environments of trade, production, and investment that were established both by state decision-making and by the patterns of capital accumulation in other countries – particularly the United States. However, the pattern established in the early twentieth century provided, crucially, the initial basis of low investment in domestic capacities for innovation – a basis that informed the subsequent strategies of manufacturing capital and was reinforced over time as the gap widened between capacities in Canada and those in other economies.

f ord is m in t h e u n it e d states and canada Starting in the 1880s, intensive development occurred in the organization of capitalist industry in the United States. A new form of Fordist organization of production developed that “spread outwards from the U SA and inspired large-scale production everywhere in the twentieth century.”1 This new form of production involved a range of changes in the organization, scale, and management of industrial production. In

26410_Smardon.indd 51

2014-09-17 10:53:23

52

Early Fordism to Neoliberal Restructuring

addition, new state structures were developed to regulate and coordinate the new form of capital.2 The crucial period of change was between the 1880s and the First World War. As noted by Chandler in relation to the new forms of managerial organization integral to Fordist production, “By the time the United States entered World War I, the revolutionary transformation of American industry that had taken off in the 1880s had stabilized.”3 Canadian capital also developed its own version of Fordist industry during the same period. As Acheson observes, “The period between 1879 and World War I witnessed a revolution in the structure and nature of Canadian industrial entrepreneurship.”4 The key moment of transformation lay between 1900 and 1914. Core sectors of Canadian industry were organized through transnational investments by companies that emerged from the development of Fordist industry in the United States, and Canadian-controlled capital developed forms of business, workplace, and technological organization that borrowed heavily on the new US structures. Particularly in the first decade of the twentieth century, Canadian industrial firms transformed themselves from small-scale operations run by families or partnerships into the large-scale operations characteristic of Fordist industry. As a result, a number of changes also occurred with respect to the social background, education, social mobility, and urban location of the Canadian business elite.5 As I will discuss, underpinning the rapid change were various state actions, ranging from the implementation of the National Policy to create favourable conditions for the growth of large-scale industry, to the efforts of Mackenzie King as Minister of Labour in developing a bureaucratic apparatus to manage Fordist industrial relations, to the support provided through tariffs and subsidies for the creation of large-scale production. Within the dominant strands of the Left political economy literature, the central role played by Fordism in Canadian industrial development during this period has not been recognized. Three key perspectives have emerged that yield one of two conclusions concerning the outcome of Canadian industrialization. Both the approach identified as the new Canadian political economy (ncpe), which argues for a commercially defined economy, and the group of approaches that reject this characterization but still accept that Canadian industrial development was limited in various ways, argue that Canadian and American industrialization cannot be seen in parallel terms. Canadian industrialization is differentiated fundamentally by its reliance on the export of staples and on the import of capital and technologies from the United States. In contrast to the approaches that

26410_Smardon.indd 52

2014-09-17 10:53:23



Canadian Fordism in the Early Twentieth Century 53

stress Canada’s limited development, a third perspective argues that the changes in this period represent a form of advanced development that, as in other advanced economies, including the United States, was following a nationally contained path. I will argue that all of these approaches miss central aspects of Canada’s development. Contrary to the first view, Canadian development, by building on the new form of capital that emerged in the United States, led to an industrialization process that at its core shared similar structures with, and indeed was modelled after, the American experience. Contrary to the second view, Canadian Fordism was derived from earlier changes in the United States and was integrated closely with US-based structures through the organization of key sectors of manufacturing by US transnational capital, the development of managerial practices based on American models of production, and a deep reliance on technological borrowing from the heartland of Fordist development in the United States. A final set of arguments that I make in this chapter concerns the location and timing of the development of Fordism. Although various advanced economies outside of the United States were marked immediately by the new form of capital through the spread of American transnational investment and attempts at imitation, the initial transnational growth of Fordist capital, as a systematic means of structuring economies, was concentrated in Canada, especially in southern Ontario, between 1880 and 1914.6 American-style industries grew in importance, American technologies were integrated into production, and American workplace practices were emulated widely. This early example of the spread of American Fordist structures has not been examined. Attention has focused on the major transfers of American technologies and forms of organization into Europe under “the Marshall Plan, the European Recovery Agency, and its successor the oecc (later oecd)” and on the response of European and Japanese domestic capital to American Fordist production structures.7 A similar focus exists in discussions of “catching up” and convergence in the post–Second World War period that do not recognize the earlier example of Canadian industrialization as an instance of the spread of American industrial practices.8 The Canadian case is also neglected in the work of the French regulation school, which, because of its identification of Fordism with Keynesian welfare state practices and collective bargaining, argues that Fordism as a way of structuring economies existed only after the Second World War.9 This absence extends to the analysis of Canadian Fordism, which follows the approach of the regulationists and places its development only in the period after 1945.10

26410_Smardon.indd 53

2014-09-17 10:53:23

54

Early Fordism to Neoliberal Restructuring

In this chapter, I present an alternative view of Canadian economic development that stresses the central role of Fordist practices in the ­organization of Canadian industry at an early point in the twentieth ­century. First, I review the dominant approaches within the Canadian political economy literature to understanding Canadian economic development before the First World War and outline how my approach differs from them. I then briefly review the changes that occurred in the United States and discuss how they constituted a new form of Fordist capital. Finally, I examine the development of Fordism in Canada and discuss the political and economic conditions that supported its creation.

vi e ws o f c a n a d ia n e c o n o mi c development Three influential views have been put forward by Left political economists with regard to Canadian economic development in the late nineteenth and early twentieth centuries. The first of these, the ncpe approach, argues that the changes that occurred in this period confirmed and re­ inforced a commercially oriented staples economy in which “the Canadian economy never fully made the vital transition from commercialism to industrialism.”11 Indigenous Canadian entrepreneurship in the area of industrial development remained limited, and there was a heavy reliance on staple exports and American sources of both capital and technology within the manufacturing sector. These factors led to an economy that did not reach the kind of mature development achieved in other advanced capitalist states.12 The role of the Canadian state within this perspective was to support and encourage this commercial orientation, and the impact of the National Policy was “to increase the benefits from being a staples exporting economy while lining the pockets of the British bond houses that financed the railways and facilitating the entry of American branch plants into Canada.”13 The second view, in a manner similar to the n cp e approach, acknowledges that Canadian industrialization was limited by reliance on staple exports, imported technologies, and weaknesses in indigenous manufacturing development, but argues that this situation did not arise from a separation between commercial and industrial capital or from a failure of the Canadian economy to make the transition from commercialism to industrialism. In this interpretation, a variety of reasons are given for the more limited form of Canadian industrialization that developed during this period – ranging from divided agrarian movements, to a “highwage” proletariat, to business and political strategies that gave primary

26410_Smardon.indd 54

2014-09-17 10:53:23



Canadian Fordism in the Early Twentieth Century 55

emphasis to staple exports and supported a narrower form of manufacturing development through import substitution industrialization (expanding domestic production by reducing imports into the Canadian market) behind tariff walls.14 Although the state is presented in these narratives as mediating between various social forces internal to the Canadian social formation, rather than as acting simply as an agent for the elaboration of dependent relations with more dominant economies, it is still viewed as supporting, in one way or another, a process of industrialization quite unlike that occurring in the US. The third view differs from first two by denying the existence of a limited industrialization process in Canada. The period from the 1870s to 1910 is seen as an important moment of internal development of ­finance capital by indigenous capitalists which, “by the second decade” of the twentieth century, culminated in “an advanced form of capitalist production, circulation and finance.”15 The centrality of resource production and the existence of weaknesses in domestic technological capacities are recognized, but these are seen as resulting from the decisions of Canadian capitalists, operating within a distinct context of nationally based finance capital, as they “struggled to realize higher than average profits, whether in the home market or in the world market.”16 A different view of the Canadian state emerges in this perspective. Rather than reinforcing and supporting a context of dependency or of limited indigenous industrial capacities, the actions of the state through the national policy provided the basis for an integrated domestic market and for the development of a more concentrated, nationally based, “monopoly capital” that in the ensuing decades would engage in transnational expansion “in both financial and industrial forms.”17 Despite their differences, these views of economic development and the role of the state in the late nineteenth and early twentieth centuries share an important similarity. The role of Canadian Fordism – as a specific version of the new form of capital developed in the US and transferred in important ways to the Canadian social formation – in underpinning both Canadian economic development and the actions of the state is neglected. Panitch and Gindin provide a recent interpretation that draws parallels between Canadian and American development and argues that Canada as a “rich dependency” represented an important early example of American “informal imperialism,” in which branch plants were established “to jump over the Canadian tariff barrier and get access to a rich domestic market which, in terms of social structure and political regime, resembled the American Midwest.”18 But, as I discuss

26410_Smardon.indd 55

2014-09-17 10:53:23

56

Early Fordism to Neoliberal Restructuring

below, although similarities with the American Midwest were an important part of the similar development paths followed in the Canadian and US economies, and the branch-plant dimension of Canadian capitalism was a crucial aspect of dependent technological development, this perspective does not capture the full process of Canada’s industrial expansion or its similarities with that in the United States. Branch-plant expansion, and “jumping over the Canadian tariff barrier,” only became important as a way of structuring the Canadian economy in the early twentieth century when the subsidiaries of US capital combined with indigenous Canadian firms, along with the Canadian state, to construct a similar framework of Fordist expansion. The coming together of the various conditions underpinning this development in the early twentieth century are not considered in their analysis. Because a systematic discussion of Fordist development is not provided in all of these perspectives, the impact of key dimensions of the Canadian economy, particularly staple exports and US capital and technologies, and the orientation of the state in supporting them, is also not adequately examined in the Left Canadian political economy literature. This literature does not sufficiently consider how the growth of staple exports and the increasing reliance on American capital and technologies were integral to parallel processes of Fordist industrialization in both the Canadian and the US economies. In supporting this parallel process of development, the Canadian federal state encouraged both commercial and industrial capital and was engaged simultaneously in responding to internal forces while assisting with the development of various connections external to the Canadian economy.

u s f or d is t s t ru c t u r e s To understand the role of Fordism in the Canadian economy, it is first necessary to specify the changes that occurred in the United States and provided the first instance of Fordist capital. Central to the transformations of the US economy in the period from the 1880s to the early part of the twentieth century was a new way of organizing industrial production. Production was concentrated in fewer, larger firms as American capital engaged in “the largest and certainly the most significant merger movement in American history” between 1897 and 1903, and new forms of industrial organization developed that inserted in a more extensive way the “visible hand” of managerial direction and control into the setting of prices and the long-term evolution of industrial capital.19 As part

26410_Smardon.indd 56

2014-09-17 10:53:24



Canadian Fordism in the Early Twentieth Century 57

of this assertion of managerial control, the internal organization of production relations (the relations between labour and capital at the workplace) were altered as all aspects of production, including workers, were subjected to new efficiency norms through “systematic and scientific management.”20 Although the new production relations were not applied uniformly, and elements of more skilled and labour-intensive processes involving characteristics of batch production continued to exist within American industrial operations, an identifiably new model of production, defined in a major way by the efforts of Frederick Taylor and other mechanical engineers, had emerged in the United States.21 As noted by Nelson, “the shop floor management movement profoundly changed the early twentieth-century manufacturing plant.”22 The new strategies were combined with different energy sources, particularly electricity, more intensive use of resources, and a series of technological innovations, centred on the US machine-tool sector, to generate much larger production volumes.23 The decisive moment in this transformation was the period between 1880 and the first decade of the twentieth century, when, on the basis of an “agro-industrial revolution” in the American Midwest, there was major expansion in industries linked to flour milling, distilling, meat-packing, and agricultural implements.24 Interrelated changes were introduced with respect to workplace relations, scale, physical organization, resource intensity, and technological aspects of production that “made the typical manufacturing plant of 1900 or 1910 a far different place from what it had been in 1880.”25 It is these new production structures – particularly as they were manifested in Henry Ford’s automobile plants – that formed the core of what came to be identified as American Fordism.26 Even though Ford was not an admirer of Taylor’s notions of scientific management, his plants took to new heights the emphasis on continuous flow processes, sequencing of production tasks, and monitoring of worker movements to ensure maximum efficiency. The rapid development of new production structures led to major increases in productivity and lower prices which, in combination with the higher wages arising from the improved position of labour in a high-growth, low-unemployment economy, supported higher levels of mass consumption. From 1880 to 1914, both real wages and labour productivity increased considerably (by 75% and 83% respectively).27 This provided the basis for classic Fordist growth as higher real wages were paid for out of the productivity gains derived from systematic and scientific management and a revamped organization of production.28

26410_Smardon.indd 57

2014-09-17 10:53:24

58

Early Fordism to Neoliberal Restructuring

Intertwined with this transformation of production and consumption was a new role for the state. The role and importance of the executive branch shifted as administrative capacities to manage the new form of Fordist capital were created and demands generated by the progressive movements concerning the need to regulate big business were mediated.29 The changes made by the Roosevelt presidency in the first decade of the twentieth century were decisive in this regard, as they marked the point at which the power of the executive to make policy and regulate economic structures relative to Congress and the courts was firmly established. As Flanagan observes, “By the end of Roosevelt’s second term in 1908, the questions facing future administrations were how much further the presidency would evolve in this direction and how this would be justified.”30 The period from 1880 to the first decade of the twentieth century was thus a momentous one in the United States. A new Fordist articulation of production and consumption developed, and important shifts occurred in state structures in support of the new form of industrial capitalism. As noted by several writers (without, however, making the same claims concerning systemic Fordism), this transformation of US capitalism is completely at odds with the characterization of the US political economy provided by the French regulation school.31 According to the regulationists, it was only after the Second World War, when “unprecedented … growth took place by taming and regulating the market,” that the Fordist capital–labour relation – involving Taylorist organization of production, the sharing of productivity gains through collective agreements, and the creation of the Keynesian welfare state – was established in various advanced capitalist economies.32 This construction of events neglects the Fordist relations that developed earlier in the United States and which, despite the absence of extensive collective agreements or a Keynesian welfare state apparatus, involved all of the central features of Fordism, including varieties of Taylorism and mechanization that supported mass production, increased wages, and rising mass consumption paid for out of productivity gains, and a new role for the state in managing the new structures of mass production and consumption.33

t he ca n a d ia n c as e The systematic development of Fordist capital was not restricted to the United States, but extended to Canada as well. The regulationist emphasis

26410_Smardon.indd 58

2014-09-17 10:53:24



Canadian Fordism in the Early Twentieth Century 59

on Fordist expansion after the Second World War also misses this. In this respect, the regulationists share a common tendency with all of the wider literature on the spread of US industrial practices after the Second World War. But Canada represented an important instance of Fordism by the early part of the twentieth century. In a series of changes that started in the 1880s and culminated in a process of growth concentrated in the first decade of the twentieth century, new forms of Fordist production and consumption were created that followed the US changes closely. At the heart of the parallel development of Canadian industry was an ability to feed off of industrial development in the United States and create a political economic context that absorbed the new form of Fordist capital in an intensive way. Initial movement in this direction occurred in the last two decades of the nineteenth century as, under the impact of the Patent Act of 1872 and the National Policy tariffs of 1879, a variety of Canadian industries expanded to occupy a greater proportion of the Canadian domestic market using technologies that, to an increasing ­extent, were patented by US residents.34 As part of this expansion, a consistent upward trend was established in the size of production establishments. Initial, but slower, growth occurred in the 1880s as establishments with an annual product of $25,000 or more increased from 60.6% of all production in 1881 to 63.6% in 1891. More rapid growth was visible in the 1890s as the proportion of manufacturing establishments in this category increased to 81.9% in 1901.35 However, the growth in the size of manufacturing plants in the last part of the nineteenth century did not represent a full movement toward Fordist production and its associated changes in the shape and form of manufacturing plants. A key indicator of the lack of change in this area was the low proportion of establishments both in the mid-range and higher scales of production relative to the US economy. Table 2.1 shows that, in 1901, establishments with mid-range scales of output (between $200,000 and under $1 million) and higher scales of output ($1 million and over) accounted for 19.7% and 14.8% of manufacturing production respectively. Table 2.2 shows that these proportions were much lower than the levels in the US, which in 1904 equalled 41.3% for scales between $100,000 and under $1 million of output, and 38.0% for scales of $1 million and over. The broad changes in the organization of manufacturing plants that had occurred in the US between 1880 and the early part of the twentieth century and dramatically altered the scale-intensity of production had not yet happened in Canada.

26410_Smardon.indd 59

2014-09-17 10:53:24

60

Early Fordism to Neoliberal Restructuring

can ad ia n m as s p ro du c t ion All of this changed with the intensive development of larger scales of production in the Canadian manufacturing sector in the first decade of the twentieth century. Table 2.1 illustrates the major shift that occurred between 1901 and 1911 as the proportion of production in establishments with mid-range scales of output between $200,000 and under $1  million increased from 19.7% to 32.2%, while the proportion of production in establishments with outputs of $1 million and over increased from 14.8% to 30.8%. The proportion of production in establishments with scales of $1 million and over had the highest rate of expansion, growing by 108.1% between 1901 and 1911. By 1911, establishments with outputs of $1 million and over existed in forty-six different industries, as compared with eighteen in 1901. Key contributors to the expansion in this category were fourteen establishments with production scales of $5 million and over, which together accounted for 8.4% of all manufacturing production. These establishments ranged across several sectors of mass production involving, among other things, smelting, iron and steel, flour and grist mill products, slaughtering and meat-packing, cigars and cigarettes, and sugar.36 These new establishments were created as part of a general shift in the structure of manufacturing as the industries at the centre of the new forms of Fordist production – transportation equipment, electrical appliances, iron and steel, non-ferrous metals, and rubber – increased their share of manufacturing output.37 The magnitude of the transformation is further illustrated by comparing average scales and the proportion of industry output in establishments with outputs of $1 million and over in the two economies. While Table 2.3 shows that Canadian scales in the industries with the highest levels of output per establishment – identified as Tier 1 industries – were smaller than those in the US (although with significant relative scales beyond the top six US industries), Table 2.4 shows that average scales in Canadian establishments in a group of Tier 2 industries were comparable to and, in some instances, greater than those in the US industries.38 In the case of boots and shoes, flour and grist mill products, and cigars and cigarettes, average Canadian scales were over 20% larger than the US industries. These results are striking when we consider that Canadian output levels in 1911 are being compared to the levels of US output in the later period of 1914, and that the higher levels of output in certain establishments with over $5 million in production are not included

26410_Smardon.indd 60

2014-09-17 10:53:24

26410_Smardon.indd 61

2014-09-17 10:53:24

≥$5,000,000 481,053

_

_

14.8

19.7

% total production

19,218

14

1,165,975

97,939

261,081

_

_ 136

375,618

947

Establishments (n)

8.4

22.4

_

32.2

% total production

* In constructing this table for Canada, I combined the categories for $200,000–$500,000 and $500,000–$1,00,000 million into one category for $200,000– $1,000,000. The US census did not provide a separate category for production establishments with outputs between $500,000 and $1,000,000. In 1911, 13.4% of Canadian production was in this category.

Source:  Government of Canada, Census of Canada, vol. III, 1901 and 1911, tables 37 and 38, lxii–lxiv (1901) and table XV, 380–384 (1911)

14,650

_

_

Total

_

_

$1,000,000–