Innovation in Brazil

Advancing Development in the 21st Century

Elisabeth B. Reynolds, Ben Ross Schneider, Ezequiel Zylberberg

• Publisher: Routledge
• Year: 2019
• Print ISBN:
  9780367146894
• eBook ISBN:
  9780429626883
• Topic: Business
• Subtopic: Business General
• Index: Business

Since the early 2000s, state-led and innovation-focused strategies have characterized the approach to development pursued in countries around the world, such as China, India, and South Korea. Brazil, the largest and most industrialized economy in Latin America, demonstrates both the opportunities and challenges of this approach. Over the course of nearly 20 years, the Brazilian government enacted various policies and programs designed to strengthen the country's capacity to innovate. It increased spending on science and technology, encouraged greater collaboration between industry and universities, and fostered the creation of new institutions whose primary aim was to facilitate greater private research and development (R&D) spending.

In this book, the editors unite a diverse array of empirical contributions around a few key themes, including public policies, institutions and innovation ecosystems, and firms and industries, that collectively make the case for a new, forward-looking innovation agenda aimed at addressing persistent challenges and exploiting emerging opportunities in Brazil. Its conclusions offer valuable lessons for other developing and emerging economies seeking to accelerate innovation and growth in the modern age.

With its interdisciplinary and wide-ranging contribution to the study of innovation, as well as attention to broader policy implications, this book will appeal to scholars and professionals alike.

Part I

The political economy of innovation

1 Institutions, politics, and state-led innovation

Danilo Limoeiro and Ben Ross Schneider

1. Introduction¹

Innovation in Brazil has long been state-driven. More than half of R&D in Brazil is public (and much of private R&D is subsidized by government); in contrast, in most developed countries the private-sector accounts for two-thirds or more of R&D investment. The Brazilian government employs a wide range of policy instruments to promote innovation and industrial policy generally, including direct subsidies, tax exemptions, government procurement, mandated spending and contributions by firms, local-content requirements, trade protection, and research grants.² Total spending on business support increased from 3% of GDP in 2006 to 4.5% of GDP in 2015. The bulk of this spending comes through tax expenditures (more than half of total spending) and subsidized credit; direct expenditures average only 0.5% of GDP (Dutz, 2018). The loans of the national development bank, BNDES, alone rose to 4% of GDP (which entailed a public subsidy on the order of 1% of GDP). To put these numbers in perspective, the popular and effective anti-poverty program Bolsa Familia reaches 50 million people and costs only 0.5% of GDP (Holland & Schneider, 2017). And some public institution and/or policy was behind the success cases noted in the introductory chapter including aerospace, flex fuel, and deep-water oil extraction.

Over the past six decades, the government has also created a full panoply of institutions and organizations to support innovation including ministries, state-owned enterprises (SOEs), development banks, regulatory agencies, university incubators, social organizations (publicly funded but privately operated), business/government councils, venture capital funds, research institutes, and other public agencies. Some of the earlier entities from the mid-20th century included Capes (university funding), CTA (center for aeronautics), BNDES (national development bank), and Petrobras (SOE in oil).³ Thereafter, institutional change was continuous, incremental, and mostly by accretion as new, more specialized entities were added to the institutional ecosystem such as FINEP (Financiadora de Estudos e Projetos) in the 1960s, Embrapa (Empresa Brasileira de Pesquisa Agropecuária) in the 1970s, sectoral funds of the 1990s, and more recently EMBRAPII (Empresa Brasileira de Pesquisa e Inovação Industrial) in 2010.

Decades of accretion added up over the 20th century to sustained institutional fragmentation (see Pacheco in this volume). Most attempts at central coordination have been short-lived and ineffective, in part because many entities were created with substantial formal autonomy in order to shield agencies from clientelist politicians (Schneider, 1991).⁴ Autonomy in many cases did offer significant insulation, especially in smaller entities or “pockets of efficiency” (Evans, 1995). However, after 1985, democratization subjected innovation agencies and policies to the pressures of maintaining government coalitions in Brazil’s evolving system of “coalitional presidentialism” (to be discussed in section 4), which in turn reinforced fragmentation and impeded policy coordination. Many analysts criticize this fragmentation and urge greater coordination (Zuniga et al., 2016).⁵ However, fragmentation and decentralization may at times also be one of the strengths of Brazil’s innovation system (as in the United States), because politicians are less interested in smaller, peripheral agencies.

Beyond statism, another core feature of Brazil’s innovation system is that it is relatively, though unevenly, closed to the global economy. This closure is especially notable in trade and continuing protection against imports, but is also true of the scientific and university system (see Frischtak in this volume). However, compared to many Asian countries, Brazil is quite open to foreign direct investment (FDI), and MNCs account for half of private R&D (Do Couto e Silva Neto et al., 2013, p. 2), encouraged in large part by public policies.

The rest of this chapter is laid out as follows. Section 2 provides a basic institutional map of Brazil’s innovation ecosystem. Section 3 discusses challenges to closer collaboration among universities, public agencies, and business. Section 4 situates the innovation ecosystem in the broader political context, focusing especially on the appointive bureaucracy and coalitional presidentialism. Section 5 adds dynamics and specificity to the institutional map by briefly summarizing some well-known success cases including flex fuel, soybeans, and aerospace. Section 6 concludes with a brief discussion of the implications for EMBRAPII, a recent institutional innovation.

2. Institutions for innovation: proliferation and fragmentation

As in the 20th century, the recent evolution of innovation and industrial policy in Brazil has been fragmented across many ministries, SOEs, and agencies.⁶ No centralized agency comparable to the Economic Planning Board in Korea (or later, the Korean Ministry for the Knowledge Economy) or the Commissariat du Plan in France has existed to coordinate dispersed policies and their implementing agencies. Thus, mapping out government support for innovation in Brazil requires covering a wide range of dispersed ministries, agencies, departments, SOEs, and foundations. In this fragmentation and lack of centralized control, Brazil resembles the United States (Block, 2008; Mazzucato, 2015). The following brief overview considers several dimensions of the main organizations involved in innovation: functions, type of organization, and resources. The main functions are: (1) planning, (2) coordinating, (3) funding, and (4) actually engaging in R&D (see Table 1.1). Evaluation could be a fifth main function, but very little evaluation has been conducted (with the exception of the public research institute, IPEA: Instituto de Pesquisa Econômica Aplicada).

Ministries

At this level, innovation policy and implementation was scattered across several ministries, from the ministries of the military services to energy to education. Although the name would suggest a dominant role, the Ministry of Science, Technology, Innovation, and Communications (MCTIC) was in fact fairly small in terms of personnel and budget compared to other ministries, and its minister has often been a politician without much background in innovation (discussed further in Pacheco this volume). Table 1.2 gives the total budgets of key innovation-related agencies. Table 1.2 also includes the largest state-level institution, Fapesp.⁷

Table 1.1 Innovation institutions by type of function

Planning	Coordinating	Funding	Engaging in R&D
MCTI(C)	MEI	BNDES	Embrapa
Regulatory agencies	Interministerial councils	FINEP	CNPEM, IPT (and other public research institutes)
MEC	ABDI	Sectoral funds (FNDTC)	Fiocruz
Anvisa	CNDI	Fapesp	Cenpes/Petrobras
MDIC		EMBRAPII	MNCs (ICT & autos)
		Tax exemptions (ICT)	Embraer
		Lei do Bem	SENAI ISIs

Note: The table is illustrative rather than exhaustive.

Table 1.2 Innovation agency budgets in Brazil, 2014

Institution	Amount (US$ millions)	Percentage
Fiocruz (research institute)	1,446	25
MCTIC (ministry)	1,200	21
Embrapa (research institute)	923	16
Cenpes – Petrobras (R&D center)	769	13
CNPq (funding agency)	646	11
MDIC (ministry)	369	6
Fapesp (funding agency)	365	6
EMBRAPII (funding agency)	6	0.1

Note: Data for MCTI, FINEP, Fiocruz, Embrapa, CNPq, MDIC from Portal da Transparência; Fapesp from www.fapesp.br; Cenpes: Relatorio Tecnologia Petrobras; EMBRAPII: 2015 Management Report. Cenpes budget refers to 2014. The Temer government merged the former ministry of communications into MCTI in 2016. Percentages may not add up to 100% due to rounding. Original figures in reals were converted at the exchange rate of US$1 = R$3.25 (median rate for 2015).

Agencies

Although many are connected to a ministry, agencies usually have some autonomy from the ministerial hierarchy and are devoted to a specific industry or sector. Regulatory agencies in areas like oil (ANP, Agência Nacional do Petróleo, Gás Natural e Biocombustíveis), electricity (Aneel, Agência Nacional de Energia Elétrica), health (Anvisa, Agência Nacional de Vigilância Sanitária), and telecommunications (Anatel, Agência Nacional de Telecomunicações) have independent authority and usually some political independence through fixed mandates for directors.⁸ In Brazil, regulatory agencies often have developmental functions and participate in committees with oversight on sectoral funds devoted to innovation (see later discussion of FINEP). In a more prominent example, the ANP formally oversees the local-content policy that Petrobras implements with a huge impact on industrial and technology policy (Lima-de-Oliveira this volume). In health, Anvisa promoted innovation in pharmaceuticals by using its regulatory power (combined with purchasing power of public health agencies) to help Brazilian producers get into generic drugs (Shadlen & Fonseca, 2013; del Campo in this volume).

State-owned enterprises

In Brazil, SOEs operate under more flexible procurement, labor, and other rules than other public agencies, so for decades, policymakers relied on them or created new SOEs to promote specific sectors (Schneider, 1991). Even though Brazil was a world leader in privatization in the 1990s in terms of the value of state assets sold, the government held on to some of the largest SOEs like Petrobras, BNDES, and smaller strategic SOEs like Embrapa and FINEP, and kept golden shares in former SOEs Embraer (airplanes) and Vale (mining), which kept them from moving abroad or from being sold to foreign competitors.

SOEs occupy a pivotal practical, and theoretically uncertain, position. From one perspective, they are merely flexible agents that facilitate and execute a range of state interventions and industrial policies decided in other ministries and agencies. However, these SOE implementors or agents have significant power resources or grow over time to have them, especially the larger ones, so that they are also participants in formulating overall policies, and independent actors in designing their own policies, as well as eventually implementers of policies decided elsewhere. This is especially the case in Brazil, where SOEs have been larger and more proactive in industrial policy than SOEs were, for example, in Japan and Korea. In the comparative scheme of things, Brazil’s developmental state was, and is, SOE heavy (Schneider, 2015b).

Within the decentralized constellation of agencies involved in innovation policy, BNDES and Petrobras emerged as primi inter pares. Within total spending on business support, the primary source of subsidized credit and minority shareholding was the BNDES, and the largest source of direct government procurement was Petrobras. For example, in 2015 BNDES loans accounted for 21% of all outstanding loans to firms and individuals; in 2010 BNDES loans rose to 4.3% of GDP, and by 2015 all BNDES assets (loans, shares, and other investments) totaled 16% of GDP (Armijo, 2017, p. 3).⁹ As such, BNDES and Petrobras engage in industrial policy in distinct ways, the BNDES through credit and share ownership and Petrobras largely through its own R&D and local-content requirements in its massive supply chain (Lima-de-Oliveira in this volume). Petrobras holds the most patents of all Brazilian firms (Brito & Mello, 2006, p. 12). In a survey in 2012 of more than 1,700 labs in Brazil, 23% of all their funding came from Petrobras – more than from any other source, public or private (de Negri & Squeff, 2016). Beyond their practical importance in terms of the large share or resources they mobilized, BNDES and Petrobras are also revealing because their technical staffs were among the most professionalized and independent within the state. That is, they were two of the long standing and historically evolved “pockets of efficiency” (Evans, 1995) (recent corruption scandals are considered later).

In the 2000s, the BNDES lending portfolio grew dramatically, doubling from 2% of GDP in 2000 to more than 4% in 2010 and came to outpace lending by the World Bank and IDB (Almeida, 2011, p. 9). The BNDES is still, as it has been for decades, the principal source of long term and export credit for large private firms. As a percentage of all financing for industry and infrastructure (including retained earnings, international loans, bonds, and equity), BNDES credit ranged from 20% to 30% over the 2000s (though it spiked in 2009 to 50%) ...

Table of contents

• Cover
• Half Title
• Series
• Title
• Copyright
• Contents
• List of contributors
• Acknowledgments
• List of abbreviations
• Foreword
• Introduction: innovation in Brazil: advancing development in the 21st century
• Part I The political economy of innovation
• Part II Towards a knowledge economy: knowledge creation and diffusion
• Part III Institutional innovations
• Part IV Firms, industries and innovation in the global economy
• Index