Innovation for all?: Legitimizing science, technology and innovation policy in unequal societies

(1)

Innovation for All?

Legitimizing Science, Technology and Innovation Policy

in Unequal Societies

A PhD Thesis

submitted to the

Department of Science, Technology, and Policy Studies (STePS), University of Twente

31.07.2011

(2)

Abstract

Most of the policy lessons concerning Science, Technology, Innovation (STI) policy promoted by international organizations derive from the experience of industrialized countries. The realities in later industrializing countries differ from these experiences as inequality rates remain high, while poverty and redistribution dominate their political agendas. As a way of improving economic and technological advancement and societal welfare, STI policy has increasingly gained importance in public policy in the developing world. This thesis seeks to understand how governments in later industrializing countries with high rates of inequality (LICHIR) legitimize and develop their STI policies. The author argues that governments in LICHIR face multiple and competing policy burdens. To overcome these burdens, political actors try to legitimize and to develop their STI policy responding to both domestic and international driving forces. These policy forces are fuelled by three sources of legitimacy: internationalization, technonationalism and social development.

Evidence derives from a comparative analysis of the STI policy processes in South Africa and Brazil from 1990- 2010. Quantitative and qualitative data come from the Global Innovation Index, policy and research documents and a sample of 99 interviews with actors in government, industry and academia.

(3)

Abstract

De meeste lessen voor Wetenshap, Technologie en Innovatie (WTI) die door Internationale organisaties bevorderd worden zijn gebaseerd op ervaringen van geïndustrialiseerde landen. De realiteit in ontwikkelings- en later industrialiseerende landen verschillen echter: ongelijkheid blijft nog steeds hoog en bestrijding van armoede en herverdeling domineren nog steeds politieke agenda’s.

Tegelijkertijd heeft WTI beleid inmiddels zijn plek in dde openbare beleidsvorming van die ontwikkelende wereld gevonden, als een steeds belangrijkere insteek ter bevordering van economische en technologische vooruitgang en sociale welstand. Onbestaanbare doelwitten? De onderzoeksvraag van deze proefschrift is: Hoe legitimiseren and ontwikkelen regeringen van Later-Industrialiserende Landen met Hoge Ongelijkheid (LILHO) hun WTI-beleid?. De auteur argumenteert dat regeringen in LILHO landen veelvuldige uidagigengen geconfronteerd zijn.

Politieke akteuren proberen om hun WTI-beleid zo te ontwikkelen en te legitimeren dat dit nationale én internationale drijfkracht krijgt. Hun beleidsinitiatieven worden aangevuurd doorr drie bronnen van legitimiteit : internationalisering, techno-nationalisme en sociale ontwikkeling. Empirische evidentie komt voort uit een vergelijkende analyse van WTI-beleidsprocessen in Zuid-Afrika en Brazilië tussen 1990 en 2010. Kwantitatieve en kwalitatieve data komen van de Global Innovation Index, van beleids- en onderzoeksdokumenten en van een steekproef van 99 interviews met akteuren in regering, nijverheid en academie.

(4)

(5)

Table of Figures

Figure 1: Brazil and South Africa compared to the OECD average in main STI Indicators . 9

Figure 2 LICHIR Countries with GINI Index < 45 and Innovation Index > 2,5 ... 18

Figure 3 Correlation Gini Index and Innovation Output ... 18

Figure 4 Correlation Innovation Output and Innovation Input ... 19

Figure 5 Correlation Gini Index and Innovation Input... 20

Figure 6 Relationships and Definition: STI Policy ... 22

Figure 8: Legitimacy in the policy process... 43

Figure 9 Interests, institutions and ideas shaping international impacts on domestic politics ... 46

Figure 10 Modern Science and Innovation Policy between three poles of legitimacy ... 48

Figure 11 Heuristic: International innovation policy arena ... 51

Figure 12 Comparing countries and levels of abstraction ... 63

Figure 13 Subject of chapter 5: Internationalization and technonationalism ... 73

Figure 14 Parallel Governance Structures in STI for the four main racial groups ... 89

Figure 15 Actors Matrix South Africa ... 94

Figure 16 Actors Matrix Brazil... 97

Figure 17 Subject of chapter 6: Social Development vs. Techno-nationalism ... 103

Figure 18 Gross R&D expenditure in Brazil and South Africa 1991-2009 ... 104

Figure 19 Brazil and South Africa’s performance in the main S&T Indicators …………..105

Figure 20 Number of Doctorates in Brazil (1987- 2008) and South Africa (2000-2007) ... 129

Figure 21 Number of Researchers in South Africa ... 133

Figure 22 MCT Budget allocation 2010 ... 141

Figure 23 Subject of chapter 6: Internationalization and Social Development ... 147

Figure 24 Overview intergovernmental organizations in STI ... 150

Figure 25 Overview STI in the UN Climate Change Negotiations ... 155

Figure 26 Overview World Bank Engagement in STI in Brazil (1976-2005) ... 160

Table 1 Concepts of developing countries and scientific -technological advance ... 16

Table 2 Overview main STI policy fields and instruments ... 25

Table 3 The four types of innovation policies and instruments (examples): ... 26

Table 4 Types of interaction between governments and intergovernmental organizations in STI ... 54

Table 5 A taxonomy - STI Policy and instruments between technonationalism, social development and internationalization ... 56

Table 6 Hypothesis deriving from the variance between the two cases ... 58

Table 7 Concepts, Indicators and sources of information ... 61

Table 8 Comparing Brazil and South Africa in a Most Similar Case Study Design ... 64

Table 9 Interviews Numbers and types of organization... 68

Table 10 Linking empirical information and conceptual framework ... 69

Table 11 Sectorial Funds in Brazil ... 117

Table 12 MCT’s plan for STI for National Development ... 125

Table 13 Science and Technology for Social Development in Detail ... 126

Table 14 Post Graduates* in South Africa and Brazil by race/ color ... 131

Table 15 Government-financed R&D in business, 2007 as a % of R&D ………..136

Table 16 South African Government Expenditure on R&D as a Percentage of GDP ... 139

Table 17 ODA commitments by donor ... 158

Table 18 Resources spend on relations, negotiations and consular ……….166

Table 19 Overview selection of environmental innovation in South Africa and Brazil .... 173

(9)

List of Abbreviation

AAAS American Association for the Advancement of Science AAU Association of African Universities

ABC Brazilian Agency for Cooperation

ABDI Brazilian Agency for Industrial Development

ABONG Brazilian Association of Non-governmental Organizations AMCOST African Ministers council on Science and Technology ANC African National Congress

APEX Brazilian Trade and Investment Agency APL Cluster/ Local Production Arrangement ARC Agricultural Research Council

ASA Articulation of the Semi-Arid

ASSAf Academy of Sciences of South Africa

AU African Union

AZAPO Azanian People's Organization BASIC Brazil, South Africa, China, India

BERD Business Enterprise Research and Development BNDES National Development Bank

BRICS Brazil, Russia, India, China, South Africa CAPES Committee for Higher Education and Training CGEE Council of Management and Strategic Studies CNPq National Research Council

COP Conference of Parties

CSIR Council of Scientific and Industrial Research CSTP Committee for Scientific and Technological Policy CVT Vocational Technological Centers

DAC Department of Arts and Culture DAC Development Assistance Committee

DACST Department of Arts, Culture, Science and Technology DCTC Division for Scientific and Technological Cooperation DEP Department of Economic Planning

DET Department of Education and Training

DHET Department of Higher Education and Training

DIRCO Department of International Relations and Cooperation DNE Department of National Education

DSD Department of Social Development DST Department of Science and Technology DTI Department of Trade and Industry

ECLAC Economic Commission for Latin America and the Caribbean ECOSOC UN Economic and Social Council

EEA Employment Equity Act

EMBRAER Brazilian Aeronautic Enterprise

(10)

FAP Foundation for Promotion of Research FDI Foreign Direct Investment

FINEP Financing Agency for Projects and Studies

FNDCT National Fund for Scientific and Technological Development GDP Gross domestic product

GDS Growth and Development Strategy GEAR Growth, Employment and Redistribution GERD Gross Domestic Expenditure on R&D GII Global Innovation Index

GTA Amazon Working Group

HERD Higher Education Research and Development HSRC Human Science Research Council

IADB The Interamerican Development Bank IAEA International Atomic Energy agency

IBRD International Bank for Reconstruction and Development / World Bank IBSA Trilateral Forum between India, Brazil and South Africa

ICSU International Council for Science

ICT Information and Communication Technology IDRC International Development Research Centre IFP Inkatha Freedom Party

IGO International Organizations ILO International Labor Office IMF International Monetary Fund

INPE National Institute of Space Research () and the IPEN Institute for Energetic and Nuclear Research IPEA Institute of Applied Economic Studies IPR Intellectual Property Rights

ITA Institute for Aeronautics

KZN KwaZulu-Natal

LDC Least Developed Countries LIC Late Industrializing Countries

LICHIR Late Industrializing Countries with High Inequality Rates MCST Ministers Committee for Science and Technology

MCT Ministry for Science and Technology

MDIC Ministry of Industrial Development and Commerce MDS Ministry for Social Development and Combat of Hunger MEC Ministry of Education

MIT Massachusetts Institute of Technology

MITI Japanese Ministry of International Trade and Industry MNC Multinational Companies

MP Member of Parliament

MRC Medical Research Council

MRE Department of External relations /Itamaraty MSSD Most Similar Case Study Design

(11)

NAM Non-aligned Movement

NEPAD New Partnership for African Development

NESTI National Experts on Science and Technology Indicators NGO Non-Governmental Organization

NRDS National Research and Development Strategy NRF National Research Foundations

NSF National Science Foundation NSI National Systems of Innovation NWU North West University

ODA Official Development Aid

OECD Organization for Economic Cooperation and Development PADCT Support Program for Development of Science and Technology PBDCT Basic Plans for Scientific and Technological Development PBMR Pebble Bed Modular Reactor

PDB Brazilian Democratic Party

PINTEC Research on Technological Innovation

PITCE Policy for Industry, Technology and External Commerce PND National Development Plans

PPA Multiannual Action Plan

PRONEX Support Program for excellence nucleus PSB Brazilian Socialist Party

PT Worker’s Party

R&D Research and Development RTS Social Technology Network S&T Science and Technology

SARUA Southern African Regional Universities Association SASOL Suid Afrikaanse Steenkool en Olie

SBPC Brazilian Society for the Advancement of Sciences SECIS Secretary for Social Inclusion

SET Science, Engineering, Technology SFD Science for Development

SME Small and Medium Enterprises STAF S&T Agreements Fund

STI Science, Technology and Innovation

STSI Science, Technology and Innovation for social impact TAI Technology Achievement Index

THRIP Technology and Human Research for Industry Program TIA Technology Innovation Agency

TRIPs Agreement on Trade Related Aspects of Intellectual Property Rights UCT University of Cape Town

UDF University of the Federal District UFRJ Federal University of Rio de Janeiro UJ Universities of Johannesburg, UKZN University of Kwa Zulu Natal

(12)

UNCSTD UN Committee on Science and Technology for Development UNCTAD United Nations Conference on Trade and Development UNCTC United Nations Centre on Transnational Corporations

UNESCO United Nations Educational, Scientific and Cultural Organization UNFCCC United Nations Framework Convention on Climate Change UNIDO United Nations Industrial Development Organization UNISA University of South Africa

URJ University of Rio de Janeiro

USPTO United States Patent and Trademark Office WIPO World Intellectual Property Organization WTO World Trade Organization

(13)

Chapter 1 Southern Science, Technology and Innovation policies in the

framework of ‘internationalization’

An Introduction

1. Introduction

Most of the understanding of science, technology and innovation (STI) policies derives from experiences in the industrialized world. Over decades, governments in Western European countries, the US and Japan pursued explicit strategies to enhance their performance in science and research-based innovation. Based on their national experiences, the industrialized nations delegated responsibility to International Organizations (IGO) - such as the OECD and special entities within the UN - to refine the formation of science and innovation related policies and to increase impact on economic growth and development. Today, several global governing institutions provide models and programs either based on their own research or on work done in collaboration with social scientists. Not surprisingly, the ‘recipes’ for policy making (Drori and Ramirez 2003) address mostly the developed world. Recently, nevertheless, relevant international institutions and clubs such as the OECD and the G8 have started to expand their outreach beyond the participating states and involve late industrializing countries in their programs, reviews, dialog forums and other forms of ‘soft’ governance.

At the same time, STI has gained importance in public policy in developing countries. But what are these policy lessons worth in the developing world where the realities differ from these experiences in the EU, the US and Japan? Despite technological advances inequality rates remain high, and poverty and redistribution continue to dominate the political agendas in developing countries. This thesis seeks to contribute to a better understanding about how science, technology and innovation (STI) policies evolve in developing countries which are relatively advanced in technological terms but still show high levels of social inequality. The overall problem this thesis tries to understand is how late industrializing countries with high rates of inequality (LICHIR) legitimize and develop STI policy.

(14)

The author argues that governments in LICHIR face multiple and competing policy burdens. To overcome these burdens, political actors try to legitimize and to develop their STI policy responding to both domestic and international driving forces. To analyze STI policy in LICHIR legitimacy is an appropiate concept in the social sciences to analyze the motivations and priorities government set in defining their interventions and resource allocation. Democratic rule reemphasizes the question of legitimacy, because the benefits of political intervention are supposed to be enjoyed by all. Providing public goods in unequal societies is a challenging task because the needs are very diverse. This gap is particularly obvious in a specialized policy field like STI policy that targets small elites. Therefore, this thesis focuses on legitimacy and priority setting in LICHIR.

The author develops a theoretical framework for the analysis of these policy forces assuming that they fuelled by three sources of legitimacy: techno-nationalism, internationalization and social development. Techno-nationalism is historically the most common form of STI policy intervention. Throughout colonial history and up to the 20th

century, STI policy intervention was strongly driven by military and security interests and economic interests. Technological development projects were tight under state control. Governments also controlled economic development and protected their economies from external competition during the 1970s and 1980s. Given the experiences of the Manhattan Project, a kind of linear thinking dominated STI policy-making, in particular the assumpation that investment into basic sciences and infrastructure would necessarily generate the intended technological outcomes. This linear perspective change in during the 1980s and 1990s towards systemic thinking on STI. The assumption now shifted towards the idea that innovation was the result of an interactive and interlinked system of public and private actors with the private firm at the core (Lundvall, 1992, Nelson, 1993, Freeman, 1987). These ideas about the determinants of innovation and its implications for policy changed alongside the more general theoretical thinking about innovation and policy. Theory, policy and practice in innovation were increasingly considered to be harmonious partners (Mytelka and Smith 2002 Kuhlmann, 2010, 1-22). International organizations (IO), particularly the OECD, EU and agencies in the UN, joint the new policy discourse and started promoting these ideas. Internationalization is the concept for these dynamics.

Internationalization became a new source for legitimacy that governments could draw on in designing the policy intervention. International norm and practice and experiences of other countries, particularly in the industrialized countries, were increasingly important examples for STI policy-making in LICHIR. International cooperation, benchmarking and dialog, bilaterally and multilaterally facilitated policy learning and transfer (Dolowitz and Marsh 2000; Rose 2005) IOs can be determining players in mainstreaming policy frameworks, institutions and ideas (Finnemore 1993; Chwieroth 2007; Béland 2009). In a global environment, driven by an increasingly transnationally active economic system, governments sought inspiration abroad to find ways to respond. Not surprisingly, the ‘recipes’ for policy making (Drori and Ramirez 2003) targeted mostly the developing world. They did not necessarily match local interests, nor did they always result in appropriate institutions in these countries.

(15)

Social development needs and demands for reducing poverty and inequality in the LICHIR dominate domestic policy discourse. Reducing poverty and inequality is not one of the main themes in European and US American innovation policy, whereas it is the main policy challenge for LICHIR and one of the main motivations to actually invest in research. External policy models might not fit the needs that LICHIR face. The analysis shows that the overall policy design follows the international discourse on systemic innovation whereas, in practice, many the actual interventions are designed in the interest-driven techno-nationalist tradition foundational to thinking about innovation. Social development needs remain urgent and policy responds to these needs. Policy innovation emerges and sets forth new institutions following the systemic ideas, and serving the linear interests. Despite the systemic discourse on innovation and development, interests shape the policy outcome. International reference to IOs and social development have become a common source for input legitimacy, whereas the outcome is bound to state-led techno-nationalist approaches to technological development. The main argument in this thesis is that STI policy --in developing countries—results from the tension between these three sources: internationalization, technonationalism and social development.

Internationalization

Social Development Technonationalism a

b

c

This thesis develops further arguments along the three sides of the triangle: Firstly, in respect to internationalization, it argues that STI policy is increasingly part of global politics. Globalization research has reached a consensus on the fact that integration of national economies into the global world economy has led to higher pressure for competitiveness at the international stage. International competition enhances the national interest in supporting research in strategic sectors. As a result, the international economic competition comes along with political competition for industry basis, investment and researchers. Policy responses are either directed towards stronger cooperation or competition (side a). During the 1990s, scholars in globalization research theorized that policy responses converge.

(16)

Convergence was the one explanation to make sense out of the fact that the problems for political interventions converged through globalization as well. As governments might perceive policy problems in the same way, they search for the same solutions. Critics argued that despite a similar perception of problems, governments respond differently. The so called policy divergence results from different political-economic cultures as well as policy competition (Jacobs 1998). Examples of arguments for divergence are theories about the ‘varieties of capitalism’ (Scharpf 1997; Hall and Soskice 2001). Empirical research could not find clear tendencies for either convergence or divergence (Schirm 2007). The literature review in the following chapter will explore the debates on internationalization and policy responses in more detail. This thesis conceptualizes the international dimensions of STI policy as the international policy arena. The focus is to understand how international driving forces shape domestic policy responses. The main argument is that STI policy in LICHIR result in an international innovation policy arena (IIPA) as a result from domestic and international driving forces.

Secondly, domestic policies are highly influenced by international institutions, interests and ideas. Domestic policies are therefore not only a result of national demands. International organizations try to influence domestic policy through development programs, policy lessons and ideas (side b). Thirdly, despite the systemic rhetoric there is still a strong focus on allocating funds to basic sciences and state controlled technology projects rather then innovation in the firm and society. Side c reflects the domestic dimension that is significantly different from the industrialized world. As high rates of inequality and poverty prevail despite healthy economic growth rates, it is a fair question to ask how state intervention can contribute to a better distribution of income and wealth in a society.

In sum, governmental actors seek legitimacy in techno-nationalism, internationalization or social development. The extent to which either of these poles manifests in the policies and institutions depends on the actors’ ideas and interests. Taking into consideration the prevailing ideas in economic theory and technological advance that international organizations promote among member and other countries, the analysis traces the influences of these international actors, ideas and institutions they promote in the domestic policy making process. At the same time, in the domestic policy-making arena, it looks at how governments deal with the challenge of inequality and poverty in priority setting and agenda setting of their policies.

1.1. Relevance and motivation

The central research problem of this thesis is to find out how late industrializing countries with high rates of inequality (LICHIR) legitimize and develop STI policy. How do government actors know what they want? How do these actors perceive a problem and the need for public policy intervention? Which national and international factors influence the problem perception of the actors? Answers to these questions have both social and theoretical relevance.

(17)

SOCIAL RELEVANCE

Science, technology, innovation and their integration into foreign policy, macroeconomic policy and social policy in advanced developing countries are the central area of research. Science and technology, and particularly Innovation policies are a relatively new policy field in developing countries. The ideological perception and motivation for policy intervention in this field has changed over the last two decades. Whereas initially the focus for public policy support lay in funding basic sciences and state controlled technology, nowadays governments perceive technological innovation as one of the main motors for economic growth.

Neoliberal doctrines prevented governments in developing countries from implementing innovation- and industrial-policy measures to enhance the innovative performance of the economy during the 1980s and 1990s. These orthodox policy perceptions changed with the criticisms of the Washington Consensus and during the late 1990s and early 2000s governments in developing countries started to formulate and implement implicit innovation targeting public policy intervention. These policies show both, continuation of lessons deriving from the experiences of the industrialized world as well as policy innovation and novel models. The thesis analyzes the composition of these policies distinguishing exogenous and endogenous policy models.

New tendencies in the global governance of STI emerge, and late industrializing countries play important roles in the changing architecture. Governments of developing countries participate and shape the architecture in global STI governance, and join the discussions in the OECD, the G8 and pushed STI on the agenda of the G20. Furthermore, there are increasing initiatives in South-South cooperation in this area.

In their economies, firms have started to become generators rather than merely buyers of foreign technology. Technical cooperation programs in the South promote both innovation and research from these countries into other, less developed or neighboring countries. New lessons might be drawn and be useful for other developing countries to look at in designing their own policy interventions. The social relevance of this dissertation is particularly high for three groups: Firstly, academics, policy advisors, government officials, firms and interested individuals and groups in the social society in developing countries who are concerned about public policy to enhance science, technology and innovation in an inclusive way. Secondly, actors in the industrialized countries might find this work relevant to structure and rethink their cooperation with these countries, and re-direct their scientific-technological cooperation, as well as their official development aid (ODA) and cooperation in other sectors such as environmental, energy and health cooperation that could focus on funding research in these areas. Thirdly, actors in multilateral cooperation, organizations and forums, will want to better understand the consequences of their actions and to find useful forms for future cooperation between international organizations and national governments. The relevance is valid for both donors and governments in the industrialized world who plan their cooperation with developing countries in this policy field. And secondly, for other developing countries, who not yet have functional STI policy in place and seek to learn from the experiences of Brazil and South Africa. These two countries are members of the

(18)

BRICS, a group that stands for its economic strength, but given their high GINI Indexes that can be exclusively found in Latin America and Africa to such a degree, their experiences can be useful for other developing countries on these countries who have not yet achieved these stages of economic and social development.

THEORETICAL RELEVANCE

For the policy analysis, social scientific research of innovation has considered theoretical frameworks focusing on national dimensions. The approach on national systems of innovation shaped both theory and praxis in of innovation policy during the 1990s (Mytelka and Smith 2002). Current analysis of innovation policy in BRICS and other developing countries concentrate on these ‘national innovation systems’ as the main framework in that innovation occurs. This national approach is a tool that derived from empirical studies in Japan, the United States and Western Europe. The NSI has three short-comings for the analysis of developing countries:

Firstly, at the national level its main focus is on the productive and the scientific-technological system in a country. This focus leaves out the consumers and the poor. Inequality and poverty are the main preoccupations of policy strategies in democratic developing countries. Therefore, SI policy analysis has to take into consideration how the SI policy integrates into the overall development strategies, if, and how it tackles the main policy problems defined by the government.

Secondly, the global economic, scientific-technological and ideological framework matters for national policy making in developing countries. Policies do not only result from the national domain. They frequently react to global economic and security threats, and crisis.

Thirdly, the framework is not valid as a theory that explains a reality of innovation policy and innovation processes. The approach has normative values, in reality, the policies target the basic funding and state controlled technologies, according to the logic of the linear model of innovation.

International influences on public STI policy have therefore not yet been subject to systematic analysis. The traditional innovation systems framework does not integrate that both domestic and international driving forces matter for priority and agenda setting of public policy interventions.

Another theoretical gap that this thesis addresses is the gap between the innovation research, policy analysis and research in international relations in the political sciences. With its analysis of national and international influences on the policy perceptions and interventions of governments in late industrializing countries, this thesis contributes in a limited way towards closing the research gap between innovation research and political scientific research of international relations. Beside the empirical contribution of bringing two mutually neglected disciplines together, the theoretical issues raised in this thesis are in the field of agenda and priority setting, as well as the role of international actors in the domestic policy process. Furthermore, the thesis presents a contribution that links innovation research to approaches from the political sciences. The political sciences have stayed away

(19)

from analyzing policy questions of technological development (Kohler-Koch 1986). This critique is already 25 years old, but since then only small achievements can be found particularly in the area of policy analysis, as shown in the following chapter.

1.2. Research Questions and Hypothesis

As shown above there have been two major challenges for the governance of STI since 1990, firstly responding to an increasing economic, political and scientific globalization and secondly to design policy in more inclusive ways to overcome inequality and make public intervention broadly beneficial. Three main research questions emerge out of this background, addressing both theoretical and empirical gaps.

1. How do governments in LIC perceive problems that require policy intervention?

2. How do domestic and international institutions, interests and ideas influence STI policy in LIC?

3. How do governments legitimize and set priorities for public policy intervention in STI under conditions of high inequality as well as moderate levels of technical advancement?

The guiding hypotheses in this thesis are:

H1 LICHIR develop their STI policy as responses to both domestic and international driving forces. These policy forces result from three sources of legitimacy: internationalization, techno-nationalism and social development.

H2 Over the last two decade legitimacy for STI, legitimacy has been gradually shifting from techno-nationalism, towards social development and internationalization. The way the international framework shapes the domestic policy responses, depends on domestic institutions, interests and ideas.

H3 Legitimate STI Policy requires legitimacy in both input and output. International sources and state led interests determine both input and output legitimacy STI policy more than social problems.

Futher hypothesis derive from the theoretical framework that is presented in chapter 3.

1.3. Methods and Research Design

The nature of the research questions requires a qualitative approach to investigate their answers. Nevertheless, quantitative work, particularly on the causal relationship between inequality and technological advance plays an important role in framing the thesis. Quantitative data on inequality and technological progress determine the case selection and set criteria and boundaries for scope and empirical focus of the analysis. These data are the

(20)

GINI Indices for inequality and the Science and Technology Indicators in the World Development Indicators.

Both South Africa and Brazil top rank the list of highly unequal and simultaneously technologically advanced developing countries.

Figure 1: Brazil and South Africa compared to the OECD average in main STI Indicators

Source: own compilation based on (OECD 2010), http://dx.doi.org/10.1787/888932334963, http://dx.doi.org/10.1787/888932333234

In a comparative approach, the thesis investigates the STI policy making in both countries in the period from 1990-2010. Both countries had inward looking economies until the 1990s. Closed political systems, the dictatorship until 1989 in Brazil and the apartheid regime in South Africa, focused on public scientific-technological missions that benefited the ruling elites and the researchers themselves. These missions concentrated mainly on energy research, medical research, agricultural and mineral research and research in defense.

Both countries carry the myth that their relative technological success is still a result of the heritage from the previous authoritarian regimes. To which extends this is true, has not been systematically investigated and this question will not be the subject of this study either. But what is known for sure is that both Brazil’s and South Africa’s new governments faced two major challenges with the political change: firstly, opening the economy and governing its globalization after decades of growth models based on government investment and import substitution; secondly, turning away from policy towards inclusive policies benefitting the population as a whole. These two turning points influenced both the design and implementation of scientific and technological policy. Therefore, comparison between

(21)

the two countries is relevant in terms of how STI policy evolved under the conditions of globalization and how policies were designed to be more inclusively.

Qualitative data that helped to answer these questions derive from policy documents, governmental strategy papers, studies and guideline paper from international organizations, as well as interviews with policy makers, academics, representatives of firms, NGOs and international organizations. These 99 interviews form the core primary data set of the thesis.

1.4 Structure, scope and boundaries

This analysis contributes to a better understanding of where the ideas come from, how policy problems are identified, how policy agendas are set and how policies are deliberated, and finally adopted. The analysis does not adress the full policy cycle (see Laswell 1957 on p. 44 in chapter 3). Some answers, if possible are presented on implementation and evaluation, but these last two stages are not part of the systematic analysis. The thesis concentrates on the problem recognition and priority setting against the concept of legitimacy.

In late industrializing countries such as Brazil and South Africa, STI policies, are increasingly part of their foreign policies. In the past, this particular policy field was rather inward oriented with the overall objective of generating new scientific knowledge and enhancing technological innovation within national borders. The shift towards international governance of SI results from the tensions that emerge from two parallel and strongly interrelated processes: whereas private actors in economy and research internationalize their activities, politics for SI typically remain in the national framework. Meanwhile foreign policies in Brazil and South Africa have become aligned to their national development strategies. Foreign threats are perceived to primarily originate from external economic instabilities rather than from military and security issues. Therefore, persistent ‘globalization’ and a nation-based international political system are not necessarily contradictory as long as governments integrate domestic-policy issues into their international agenda.

The thesis is structured in eight chapters along its conceptual core. The first chapter is this introductory chapter that provided an overview and background on the thesis, and concludes with this last section on the thesis’ structure.

Chapter 2 provides a literature review and definitions. Literature derives from three different fields: policy analysis, international relations and innovation research. Reviewing the literature, the author also evaluates the different indices that exist for assessing technological advancment, innovation performance and input and inequality. The global innovation index was the only index that was used to integrate all variables. Assessing theories on the relationship betwenn innovation and inequality, the author guides the assessment in the light of her own analytical findings on these relationships based on the GII. Futhermore this chapter presents definitions found in literature and specifies the concepts that are used in the thesis. The chapter follows the guiding question: What is different about STI policy in late industrializing countries with high inequality rates? The chapter supports the argument that there are two main differences: firstly, given that high inequality and poverty trigger policy problems for social policy and development policy, governments in

(22)

late industrializing countries face manifold policy burdens. In consequence, governments need to bridge very diverse interests of poor people who need basic services, and better off elites. Secondly, later industrialization occurs in a globalizing world economy that requires the political economy to respond and adapt to international competition and regulation. Hence, domestic STI policy is increasingly international.

Chapter 3 provides a conceptual framework that is the core tool for the analysis. The author visualizes the three main sources for legitimacy in STI policy in LICHIR as seen in the triangle above. This framework links up with the literature on internationalization and policy response and builds the hypothesis based on these findings.

Chapter 4 specifies the research design and the methodology that guides the research process. Moreover it presents the criteria for the case study selection are presented, the way and the type of data used and how it was used to answer and operationalize the research questions.

Chapter 5, 6 and 7 presents empirical evidence from the case studies. These chapters are sturctured along the three sides of the triangle. Chapter 5 presents an historical overview on the historical evolution of STI policy in both countries. The main argument of this chapter is that the STI policy has developed under strong international influences that laid foundations for technonationalism. These influences start with their early beginnings during colonialism. The interests and ideas of colonial rulers have influenced the innovation systems dramatically. The institutions structured around the economic and security needs of the colonizers and were formed as copies from the prevailing institutional model in the home country. The chapter builds up a historical background throughout the world wars and the cold war up the present.

Chapter 6 concentrates on the conflicting sources for legitimacy technonationalism and social development. In this chapter the author takes up the argument built up in chapter 5 that foundations for interest driven technonationalism were laid throughout the evolutions of the scientific institutions and the STI system. Therefore, technonationalism is the default choice of policy makers, because the system has evolved in this way. Under democratic rule, legitimacy in public policy making cannot be ignored as it has been before, and because of the pressure to make ‘policy for all,’ social development aspects have also found their way into innovation policy. Analyzing the instruments for social development impacts of STI, the author argues concludes that these efforts have been rather ‘cosmetic’ than systematic.

Chapter 7 focuses on the linkage between internationalization and social development. The aspects analyzed in this chapter focus on the international influences on STI policy making, like lessons and advice coming from external actors on the one hand, and the changes in international cooperation in STI policy on the other. This chapter affirms the hypothesis that the policies in STI are the joint result of both international and national forces.

Chapter 8 presents and discusses the results and concludes the theses with a presentation of the findings and their relevance for further research.

(23)

Chapter 2 Internationalization, inequality and policy response

in developing countries

2. Introduction

This chapter presents the definitions of the core concepts and available answers to the research questions reviewing the research literature. Structured in two parts, the review helps to identify and specify the intellectual problem tackled in this thesis. The first part presents the empirical literature on technological advancement and inequality in developing countries. The main argument in this first section is that previous groupings fall short, because they are selected only on one-dimensional indices. Drawing on the Global Innovation Index (GII 2011) the author suggests a new selection of indicators preparing for the choice of cases for an in-depth qualitative analysis. Furthermore, this part explores STI policy aspects and the differences across LICHIR. The argument here is that governments face a multiple policy burden, dealing with classical problems of underdevelopment and sophisticated industrialization at the same time.

The second part presents a literature review disentangling research literature on questions of globalization, policy response and innovation. Two key strands of literature present findings from two disciplines which have often suffered from reciprocal neglect: globalization research in international relations and international political economy in the context of political sciences, and economic research on innovation. The purpose of the literature review is to locate the theoretical problem of policy motivation, legitimacy and priority setting under conditions of globalization in LICHIR in the field of STI. After a summary statement, the next chapter presents a theoretical framework for the analysis of STI policy in these countries.

2.1. Definitions: Late industrialization, inequality and STI policy

General wisdom in economic theory suggests that technological innovation triggers economic growth, and therefore societal welfare. Successful economic development trajectories come along with technological advance. This proposed causality does not necessarily hold for all cases, because economic development paths also depend on other factors, historical and geographical factors as well as a country’s skills and policies, and never evolve in quite the same way in different contexts. Although the relationship between innovation and economic growth is widely accepted among policymakers and academics, positive societal benefits do not help all social sectors. Otherwise, countries with highly unequally distributed income and simultaneous technological advance would not exist. The

(24)

following section explores the answers in literatures on the relationship between technological advance and inequality.

LINKING LATE INDUSTRIALIZATION, INEQUALITY AND TECHNOLOGICAL ADVANCE

Starting with assumptions of neoclassical economic theory on technological advance and its impact on inequality, this section provides an overview on how scholars from the developing world challenged this theory. Furthermore, this section shows the existing attempts to group and analyze developing countries according to their technological advance. The author suggests new concepts for the analysis of Later Industrializing Countries with High Inequality Rates (LICHIR) on the basis of the Global Innovation Index. Neoclassical vs evolutionary economic theory

Neoclassical economic theory suggests a convergence in the world’s income, because international trade leads to equal factor prices (Samuelson Effect). In this logic, less developed countries reach the same prices for labor and capital as soon as they open their economies. In the 1980s, growth theory (Romer 1986; Grossman and Helpman 1990) started to recognize that technical change and increasing returns to scale are at the core of the process of economic growth (Reinert 1996; Reinert 1999), cit in (Freeman 2004). Neoclassical ideas also suggest that lesser developed countries have a longer way to catch up technologically, but will eventually reach the technological frontier (Reinert 2004).

Singer and Prebisch’s thesis on underdevelopment states that the benefits of innovation and technical change are transferred very differently according to each country’s level of economic development. Whereas innovation leads to higher wages, higher income, higher tax income and higher welfare in the industrialized countries, in the developing world these positive effects cannot be demonstrated; indeed the opposite occurs. According to Singer, in the developing countries the benefits of technological innovation manifest in form of lower consumer prices in the industrialized world and lower wages in the developing countries. Therefore, technological change is not necessarily beneficial for generating wealth in the developing world (Singer 1981; Reinert 2004). This theory is a Schumpeterian explanation for the effects of technological change. This school of thought gave foundations to evolutionary economics, an understanding of economic development in the Darwinian sense of economic cycles, competition and historical trajectory.

Evolutionary economics challenge the neoclassical economics, because they do not take historical, geographical factors and assume that, as in the case of international trade, technological change benefits economic growth in all countries. International trade, according to the Samuelson effect, will lead to ‘factor-price equalization’, which means that the prices paid to the factors of production – capital and labor – will be the similar throughout the world. In contrast Myrdal, representing evolutionary economics, states that world trade would tend to increase already existing differences in incomes between rich and poor nations (Samuelson, Myrdal cit in Reinert 2004). These two economic theories predict the effects of globalization in trade, capital and labor differently. Neoclassical economics assume positive effects for all countries; whereas evolutionary theory recognizes regional, geographic and historical trajectories suggesting that, depending on these, there will be overall winners and losers from globalization.

(25)

Evolutionary economic theory makes generalization very difficult given the endogenous development paths. The evidence for this approach is in literature on economic history. History has already shown that economic development follows different paths in different countries. In Germany and the UK, for example, development resulted from the sheer competition between the two countries. These European experiences, considered to be the first wave of industrialization, inspired the famous academic writing on the role of the state in economic development. Adam Smith’s ‘wealth of nations’ arguing for free trade in the logic of the rationally acting ‘homo oeconomicus’ (Smith 1910) led to the German response by Friedrich List (List 1841, 1989; Senghaas 2007). List conceptualized the national economy as a system of political economy that is interwoven with other political economic systems of other countries. List agreed with Smith’s ‘division of labor’ in the first place, but argued that the linkages and networks in different sectors matter and that therefore workers, although divided in different sectors, actually need to work together. He also advocated the protection of infant industry, further developed by Joseph Schumpeter (Schumpeter 1949). These two opposing schools of thought about economic development laid the foundations for neoclassical and evolutionary theories. Both still influence economic policies in our century.

The assumptions of evolutionary history require careful case study analysis considering the historical, geographical and political economic factors on economic development. Given that the development paths are different, generalization is difficult. Still, some attempts have been made to group developing countries according to their technological and economic advances.

New Industrialization -- first in Japan, and later in Singapore, Taiwan, South Korea and Hong Kong/China -- inspired writers to conceptualize a new development path characterized by heavy state intervention and strategic choices focusing on industrial production, engineering and training of one or two high technology products (Chang 2003). Literature refers to these countries development as the second wave of industrialization. Governments in Latin American Newly Industrializing countries -- the ones most mentioned in the literature are Brazil and Mexico—never made these choices. New industrialization in the Asia group produced socio-economic development and societal welfare. In these societies in Asia, inequality rates are lower than in Latin America. The Latin American newly industrializing countries, in turn, show high indices in income inequality rates and poverty rates despite similar average growth rates.

Given these differences, other factors must play a role to convert technological advance into societal welfare successfully. In a comparative study between the development paths in South Korea and Brazil, Viotti identified the policy as the determining variable for successful and less successful industrialization (Viotti 2002). Although not entirely explicit on the matter, Viotti suggests correctly that trade, industry, and education policy matter in ‘overcoming the shortcomings of a passive to an active learning economy and successful industrialization.’ These experiences on development in ‘new industrializing countries’ suggest the crucial role of the state in public policy supporting innovation, economic development and in distributing its benefits.

(26)

Country groupings according to technological and economic development

Late industrialization is another concept trying to grasp the differences in development paths in Latin America and in other parts of the world. Literature refers to late industrializing countries (LIC) as countries which developed technological capabilities to generate mid-technology industries first and moved into high technology sectors later1

(Viotti 2001; Amsden 2003). The industrialization paths of these countries diverge as well as their policy efforts, the significance of international cooperation and dependence on other countries. Different authors conceptualize LIC in different ways. So far, there is no clear-cut definition helping to isolate a defined group of LIC (see table 1). Mani identified a group of eleven technology generating countries as those who have US registered patents (Mani 2004). Da Motta e Albuquerque takes regional inequality into account and conceptualizes South Africa, India, Brazil and Mexico as ‘immature innovation systems’ (da Motta e Albuquerque 2003). This logic comes closer to the ‘in between status’ that later industrializing countries find themselves in due to high inequality in income and regional concentration of technological capability clusters between developing countries without a scientific basis and industrialized countries with full-fledged innovation systems. Albuquerque’s study suffers from the same shortcoming as Mani’s criteria. In particular, taking US registered patents as an output indicator may be inaccurate because such patents represent only the peak of the iceberg of innovative activity in developing countries. Measurement of innovative performance is generally a problem in developing countries, because the indicators were initially designed to measure performances in OECD countries. These indicators have a narrow focus on R&D performance, patents being one of them. This narrow focus put suggests a very limited concept of innovation. In response to this concept, governments limit their innovation policy to pure R&D policy, too (Godin 2005). Other efforts in industrial development and education do not enter in the OECD set of performance indicators, although they are more important in less developed countries.

The overview in table 1 shows multiple attempts to group developing countries according to their economic technological advance. These groups are not the political groups of developing countries (G5, G77 or G20 etc.). These groupings are fairly different, and the same countries hardly ever appear in the same group.

1_{Amsden refers to China, India, Indonesia, South Korea, Malaysia, Taiwan, Thailand, Argentina,}

(27)

Table 1 Concepts of developing countries and scientific -technological advance Concept and Author Defining criteria / Indicators Countries Immature Systems of

Innvation (da Motta e Albuquerque 2003)

Developing economies with highest patent data from the USPTO, scientific publications data from ISI data base

South Africa, India, Brazil and Mexico

The Rise of the ‘Rest’ (Amsden 2003)

Robust manufacturing expertise allows to move into middle and later high technology sectors

China, India, Indonesia, South Korea, Malaysia, Taiwan, Thailand, Argentina, Brazil, Chile, Mexico and Turkey

Type 1 countries: potential to generate own technology Type 2 countries: mere assemblers of imported technology (Mani 2004)

The number of US registered patents in developing countries

Type 1: Taiwan, Korea, Singapore, India, South Africa, Brazil, China, Mexico, Argentina, Malaysia

Type 2: Paraguay, Uruguay, Vene-zuela, Algeria, Egypt, Morocco, Lebanon, Tunisia, Syria, Pakistan, Brunei, Philippines, Thailand, Vietnam Less Successful Developing

Countries

(Intarakumnerd, Chairatana et al. 2002)

Degree of learning intensity in the economy, degree of technological intensity in manufactured exports

Thailand

Newly Industrializing countries (Balassa 1981; Haggard 1986; Dahlman, Ross-Larson et al. 1987)

Economic and technological catching up in the 1980s and early 1990s, focus on economic and technological performance and policy reform. No clear indicators.

Asian NIC: South Korea, Taiwan, Hong Kong/China, Singapore

Latin American NIC: Brazil, Mexico BRICS

e.g. (Cassiolato and Vitorino 2009)

Regional focus, technological advance, foreign policy alliance between the five states since 2010

Brazil, Russia, India, China, South Africa

Dynamic adopters

(Technology Achievement Index TAI 5 0.20–0.34). (Desai, Fukuda-Parr et al. 2002)

• output: number of patents granted toresidents per capita, receipts of royalties and license fees from abroad per capita.

• new technology: number of Internet hosts per capita, the share of high-technology and medium-technology exports in total exports • old technology: telephones and electricity consumption per capita • human skills: mean years of schooling in the population aged 15 and older, the gross tertiary science enrolment ratio

Uruguay, South Africa, Thailand Trinidad and Tobago, Panama, Brazil, Philippines, China, Bolivia, Colombia, Peru, Jamaica, Iran, Tunisia, Paraguay, Ecuador, El Salvador, Dominican Republic, Syria, Egypt, Algeria, Zimbabwe, Indonesia, Honduras, Sri Lanka, India

Leading, middle and third tier and emerging innovator countries (Furman and Hayes 2004)

• estimated production function for innovations at the world’s technical frontier

• Output: OECD S&T Indicators, World Bank, the USPTO and the Penn World

• Input: country-level investments in R&D and human capital (HA), and national policies (XINF). GDP PER CAPITA, R&D human capital and country-level investments in R&D (FTE R&D PERS and R&D$) • openness to international trade

Leading innovator: Germany, Japan, Sweden, Switzerland, USA

Middle tier innovators: Australia, France, Austria, UK, Belgium, Netherlands, Canada, Norway

Third tier innovator countries: Mexico, Hungary Italy, New Zealand, Spain Emerging innovator countries, Denmark, Finland, Iceland, Ireland, South Korea

(28)

Most of the authors listed above distinguish the countries with technological and economic advance only based on one or two criteria, although the factors that facilitate and constrain technological development are multi-dimensional. The Technology Achievement Index (TAI) tries to give a more comprehensive ranking of 172 countries, distinguishing four groups, leaders, potential leaders, dynamic adopters and marginalized countries (Desai, Fukuda-Parr et al. 2002). Although the index claims to cover four dimensions of technological adoption and diffusion, it limits itself to technological performance and human resource indicators leaving out indicators for governance and social development. The adoption of so-called old technologies (electricity and phones) reflects underdevelopment in the group of developing countries of my concern. This index provides a good start, but it does not consider the constraints to innovative performance and technological advance. High levels in income inequality, unemployment, poverty and state fragility constrain technological advance.

Technological advance according to the Global Innovation Index

The Global Innovation Index (GII 2011), in turn, has multiple dimensions and compiles indicators on governance as well as social benefits. The GII ranks 132 countries, but it does not create groups like the TAI and is for that reason not listed in the table above. Recognizing the importance of governance and institutions, the Global Innovation Index comes to a quite different ranking. For example, fragile states like Zimbabwe are found at the bottom of the GII ranking. In contrast, according to TAI, Zimbabwe improbably ranked higher than Brazil in its capacity to adopt and create technology.

This thesis draws on the GII, which is the most comprehensive multidimensional innovation index available today. Based on the GII, the author distinguishes those countries with an average Gini inequality value of more than 50, and simultaneously medium values of technological advance (measured in GII innovation output and input indicators details in footnote). This index helps to construct an overview on the relationship between innovation an inequality. The index ranks 132 countries. As shown in the graph below, all countries with Gini income inequality rates higher than 45 are located in Latin America and Africa. The Asian group of newly industrializing countries does not show such unequal societies which is a positive condition for a better distribution of economic growth benefits.

The group that ranks over 2,5 in the overall GII with Gini inequality rates over 45 are Nicaragua, Honduras, Guatemala, Colombia, Botswana, Zambia, Lesotho, Peru, El Salvador, Dominican Republic, Argentina, Mexico, South Africa, Kenya, Cote D’Ivoire, Mozambique, Gambia, Senegal, Suriname, Mexico, Uruguay, Brazil, Panama, Costa Rica and Chile. 1

In terms of innovation output, Brazil, Chile, South Africa and Suriname are among the most unequal and most productive countries. In terms of innovation input, South Africa is the leading country although showing a proportionally low innovation output.

(29)

Figure 2 LICHIR Countries with GINI Index < 45 and Innovation Index > 2,5

Source: Own compilation based on (GII 2011)

Knowing that the most unequal countries in the world are in Latin America and Africa, the author would like to see if there is a connection between inequality and innovation performance. The GII constructs an index for innovation output that consists of indicators for two pillars: science output and creative output.2_{Scientific and technological}

production, patents, scientific papers etc. compiles all together as innovation output. The GII allows correlating variables and its sub-indices such as the output index. The correlation between innovation output and inequality shows a quite clear negative relationship.

Figure 3 Correlation Gini Index and Innovation Output

Although this correlation is not perfect, it suggests that more equal societies have a higher innovation output. While Namibia, Botswana and Zimbabwe sit on the highest ranks

(30)

for inequality and the lowest innovation rates, more egalitarian societies like Sweden, Norway, Finland and Switzerland perform well in science and innovation as well. Brazil and South Africa cluster up in the group of highly unequal countries with GINI index rates over 50. Nevertheless, they stand a little out of the line showing the relatively most innovative output, despite the high innovation rates.

Two possible hypotheses derive from this relationship: The more equal societies are, the higher the innovative output. And reversely, the more innovative output, the more equal a society. A common problem in quantitative studies is specifying the direction of causal relationships. The data presented in the graph suggest that countries with the lowest GINI levels therefore rank higher in innovation. Does more innovative output lead to less inequality in a society? This also seems to be the case. To avoid conclusions on spurious relationships, however, the role of innovation inputs requires further examination.

Innovation output depends on innovation input. The higher the innovation input the more innovation output a country produces. This positive relationship comes out clearly in the graph below.

Figure 4 Correlation Innovation Output and Innovation Input

The GII operationalizes these values through five pillars: institutions, human capacity, Information and Communication Technology (Juma, Fang et al. 2001) and uptake of infrastructure, market and business sophistication.2_{The relationship between innovation}

input and income inequality measured in Gini Index is less obvious, but we can still see a tendency the correlation suggests as indicated through the black line.

Innovation for all?: Legitimizing science, technology and innovation policy in unequal societies