«Industrial policy and the creation of new industries: evidence from Brazil’s bioethanol industry Santiago Mingo*,y and Tarun Khanna** Downloaded ...»
Industrial and Corporate Change Advance Access published November 7, 2013
Industrial and Corporate Change, pp. 1–32
Industrial policy and the creation of
new industries: evidence from Brazil’s
Santiago Mingo*,y and Tarun Khanna**
Downloaded from http://icc.oxfordjournals.org/ at Harvard Library on July 10, 2014
Industrial policy programs are frequently used by governments to stimulate economic activity in particular sectors of the economy. This study explores how an industrial policy program can affect the creation and evolution of an industry and, ultimately, the long-term performance of firms. We examine the history of the Brazilian bioethanol industry, focusing on the industrial policy program implemented by the Brazilian government in the 1970s to develop the industry. We put together a novel data set containing detailed information about the history of bioethanol producers. Our findings show that plants founded during the industrial policy program tend to be, in the long run, more productive than those founded before the program was in place. Based on additional analyses and complementary fieldwork, we infer that the wave of acquisitions that occurred after the end of the industrial policy program had an important effect on the performance of the plants founded when the program was in place. Industrial policy, especially in conjunction with a competitive post-industrial policy business landscape, can succeed in nurturing competitive firms.
JEL classification: L25, L52, O25, Q16.
1. Introduction Governments play a major role in fostering industrial activity and can generate significant changes in the structure of an industry (Malerba and Orsenigo, 1996;
Porter et al., 2000; Ring et al., 2005; Athreye et al., 2009). State intervention in specific industries—commonly called industrial policy—can stimulate growth and
´˜ ´˜ *Escuela de Negocios, Universidad Adolfo Ibanez, Av Diagonal Las Torres 2640 – Penalolen, Santiago, Chile. e-mail: firstname.lastname@example.org **Morgan Hall, Harvard Business School, Boston, MA 02163, USA. e-mail: email@example.com y Main author for correspondence.
ß The Author 2013. Published by Oxford University Press on behalf of Associazione ICC. All rights reserved.
S. Mingo and T. Khanna 2 of 32 create new jobs (Johnson, 1984; Griffiths and Zammuto, 2005; Spencer et al., 2005).
More specifically, an industrial policy period is a distinct period during which the government makes or fosters large investments and uses multiple tools to develop a particular industry—for example, subsidies, incentives, tax reductions, new regulations, institutional reforms, and investments in infrastructure. As Pack and Saggi (2006: 267) put it, “Industrial policy is a type of selective government intervention or policy that attempts to alter the structure of production in favor of sectors that are expected to offer better prospects for economic growth in a way that would not occur in the absence of such intervention in the market equilibrium.” The idea of using industrial policy programs to push the development of an industry is attractive.
Many governments in both developed and less developed countries have used and Downloaded from http://icc.oxfordjournals.org/ at Harvard Library on July 10, 2014 are using industrial policy programs with the hope of increasing the level of economic activity and technological development in their nations (Rosenstein-Rodan, 1943; Amsden, 1989, 2001; Murphy et al., 1989; Stiglitz, 1996; Kim, 1998; Hausmann and Rodrik, 2003; Mahmood and Rufin, 2005; Lerner, 2009; Chu, 2011).
In this article, we explore how an industrial policy program can affect the creation and evolution of an industry and, ultimately, the long-term performance of firms.
Despite the importance of industrial policy as a development tool, its effects on firm strategy and profitability have not received significant attention in the literature (for some exceptions, see Branstetter and Sakakibara, 1998; Porter et al., 2000). A particular limitation of the existing literature is its focus on industry aggregates rather than leveraging and studying firm-level responses to industrial policy programs (Aw et al., 2001; Barua et al., 2012; Kolesnikova, 2010).
We aim to improve our understanding of how governments affect both the development of new industries and the strategic choices made by new and existing entrepreneurs (Johnson, 1984; Griffiths and Zammuto, 2005; Athreye et al., 2009;
Pearce et al., 2009). Firm profitability depends on many different factors, including aspects related to the industry, firm, and business (McGahan and Porter, 1997, 2002;
Adner and Helfat, 2003). In addition to these factors, we emphasize the importance of government-level issues that affect profitability. An industrial policy period is an important and eventful stage in the life of an industry—substantial changes occur in the environment surrounding new and existing firms (Thomas, 1994). Thus, industrial policy offers a unique setting that can provide multiple lessons to scholars studying the impact of the environment on organizations, entrepreneurship, and business performance (Child, 1972; Aldrich, 1979; Gartner, 1985; Donaldson, 2001).
The empirical analysis focuses on the history of the Brazilian bioethanol industry and the industrial policy program implemented in the 1970s to develop the industry.
This program led to the creation of one of the largest alternative fuel industries in the world (Hausmann and Wagner, 2009). The bioethanol program in Brazil is an interesting setting because the onset and end of Pro-alcohol (this is the name given to the subsidy period) were due mainly to exogenous shocks—high and low oil prices, respectively. Our historical analysis of the bioethanol industry in Brazil and its Industrial policy and new industries 3 of 32 industrial policy program is based on a novel data set containing detailed information about the history of bioethanol producers coupled with fieldwork.
2. Literature review and theoretical framework Government policies can play a major role in the creation and development of an industry (Malerba and Orsenigo, 1996; Dosi et al., 1997). More specifically, industrial policy programs can provide a way of generating the big push needed to expand markets and get out of no-industrialization traps (Rosenstein-Rodan, 1943; Murphy et al., 1989). These traps exist for different reasons. For example, the presence of a Downloaded from http://icc.oxfordjournals.org/ at Harvard Library on July 10, 2014 small domestic market combined with costly foreign trade can explain why new technologies are not adopted (Murphy et al., 1989). Theoretically, implementing an industrial policy program would be beneficial when, despite the nation’s potential comparative advantage in an industry, some form of failure is preventing investors from investing in the industry at a socially optimal level (Brahm, 1995).
Inadequate infrastructure—for example, “hard” infrastructure such as roads or “soft” infrastructure such as education—can be the result of a coordination failure:
the industry is not developing because good infrastructure does not exist, whereas the required infrastructure is not being developed because the industry does not exist. In other words, whereas the rate of return to coordinated investments is high, the rate of return to individual investments remains low. Under these circumstances, governments can induce the coordination of different agents by directly investing in infrastructure. This intervention can be very important to get out of a “bad equilibrium” (Rodrik, 1995, 1996). For example, it is difficult to imagine the development of world-class electronic industries in some East Asian countries if government policies designed to improve the educational infrastructure in math, science, and engineering were not previously established (Stiglitz, 1996).
Coordination failure can also prevent the emergence of crucial intermediaries that facilitate different types of market transactions (Khanna and Palepu, 2000, 2010).
Intermediaries usually wait until a critical threshold of activity is surpassed before entering a market or industry. For example, financial analyst firms will not focus on an industry that is below a certain size. The shortage of information about the industry will discourage investors from investing in this market. Thus, the size of the industry will probably stay low and financial analysts’ lack of interest in the market will continue. Other types of crucial intermediaries are venture capital firms, market research and advertising firms, insurance companies, and technical educational institutions. Naturally, the absence of intermediaries will further increase the hurdles that an industry must overcome to get out of a low-equilibrium trap. In this case, the government can also step in to support the development—and sometimes take the role—of these important agents. Getting out of these low-equilibrium traps is considered critical to the achievement of higher levels of economic S. Mingo and T. Khanna 4 of 32 development. The argument is that after an economy gets out of the trap, a momentum builds between growth and productivity that drives industrialization and technological development forward (Amsden, 1989; Mahmood and Rufin, 2005).
Industrial policy might be arbitrary in its support of specific industries, but the bright side of “promiscuity” can be the activation of a broad-based growth momentum. In part, this explains the popularity and support that many politicians and business leaders give to industrial policy programs (Amsden, 2001).
A complementary perspective is proposed by Hausmann and Rodrik (2003). They argue that, in less developed countries, entrepreneurs are more reluctant to learn “what a country is good at producing.” This phenomenon occurs because the first entrepreneur who makes the “discovery” can appropriate only a small part of the Downloaded from http://icc.oxfordjournals.org/ at Harvard Library on July 10, 2014 social value that is generated, especially in emerging markets where there is a lack of property rights and weak institutions are ubiquitous. If this is the case, Hausmann and Rodrik recommend governments to use industrial policy to stimulate the discovery process.
Industrial policy also has critics. Easterly (2006) states that less developed countries have “bad governments,” inadequate institutions, and a lack of ability to generate good policies. According to him, governments in less developed countries will have a lot of difficulties implementing industrial policies successfully, no matter how appealing these might be. Hayek (1945) argues that governments lack the information needed to coordinate a large industrialization program. According to Krueger (1974), widespread government intervention amplifies returns to political rent seeking, intensifying poverty traps.
Industrial policy programs elicit strong reactions from academics, economists, and policy makers (Rodrik, 2004; Pack and Saggi, 2006). Those who believe in the efficiency of markets see industrial policy as an invitation to rent seeking behavior.
Those who believe that the presence of market failures is pervasive argue that industrial policy is a necessary tool to develop new industries and increase the level of economic development.
2.1 Theoretical framework Industrial policy programs can have a significant impact on the number of newly created firms and their characteristics (Sine et al., 2005). During an industrial policy period, many new entrepreneurs enter the industry. Naturally, the ability of these entrepreneurs plays a crucial role in the foundation and early development of their ventures (Baron et al., 1999; Shane and Venkataraman, 2000; Santarelli and Vivarelly, 2007). Additionally, the organizational environment at the time of founding can also have an important impact on the behavior of these new entrepreneurs and the success of their companies. Naturally, all these changes that occur during the industrial policy period can have long-lasting effects on the future performance of firms. To understand how industrial policy programs can affect the creation and Industrial policy and new industries 5 of 32 subsequent evolution of an industry, we divide the theoretical discussion in two parts. First we focus on the industrial policy period and then on the post-industrial policy period.
2.1.1 Period of industrial policies When studying industrial policy programs, it is important to consider the effects of entrepreneurial ability. The ability of an entrepreneur can have a significant impact on the performance of his or her firm (Gimeno et al., 1997; Shane, 2000). We argue that subsidies and incentives associated with industrial policy programs facilitate the entrance of a high number of entrepreneurs with differing ability levels (Santarelli and Vivarelli, 2002). We use the term industrial policy entrepreneur to denote those Downloaded from http://icc.oxfordjournals.org/ at Harvard Library on July 10, 2014 entrepreneurs that enter the industry during the industrial policy period. In this study, our focus is on the ability to manage companies successfully in different types of business settings. For example, industrial policy entrepreneurs with a high ability would be able to adapt to their surroundings and make their businesses succeed.
Certainly, government subsidies and incentives attract a group of high-quality entrepreneurs that in the past have been reluctant to enter the industry. However, industrial policy programs also attract low-ability entrepreneurs who would not enter the industry under normal circumstances (Evans and Jovanovic, 1989). For example, empirical studies have shown that the introduction of public policies that ease the cost of external finance attracts individuals with a wide range of entrepreneurial ability, including less gifted entrepreneurs (Nanda, 2011). Naturally, the long-term effects that low-ability entrepreneurs have on the industry depend on the capacity of the industry to select them out at a later stage—for example, after the possible winding-down of the industrial policy program.