Where are we in the economics of industrial policies?

Editors’ note: This column is part of the VoxDev series on industrial policy.

“Industrial policy” refers to policies that stimulate specific economic activities and promote structural change. As such, industrial policy is not about industry per se. Policies targeted at non-traditional agriculture or services (e.g. public subsidies for agricultural products, call centres, or tourism) qualify as much as incentives on manufactures . “Productive development policies,” as the Inter-American Development Bank (2014) calls them, is probably a more appropriate term. 

Market failure

Malfunction in the markets for credit, labour, goods, and knowledge have long been at the core of what development economists study. They are widely perceived to be pervasive, even if systematic evidence is sketchy and hard to come by. For example, collateral constraints combined with asymmetric information result in credit market imperfections and incomplete insurance, learning spills over to copy cats from producers who adopt new processes, or labour can move from employer to employer, taking their on-the-job training with them. Moreover, most governments already carry out various forms of industrial policy, even if they call it by other names (“export facilitation”, “promotion of foreign investment”, “free-trade zones”, etc.).  

Economic development is fundamentally about structural change: it involves producing new goods with new technologies and transferring resources from traditional activities to these new ones. But the process of structural change is rife with market failures, which is why it does not happen automatically even when governments do a decent job of providing an adequate investment environment. Investment in new industries requires finance, but firms with no track record appear excessively risky to private lenders. It needs complementary services and inputs, which are often missing. It entails training workers and managers, who then become free to circulate to competitors and copycats. It generates learning-by-doing, which others can benefit from. Under these conditions, the deck is stacked against entrepreneurs who contemplate diversifying into non-traditional areas. 

Ambiguity creates problems

While the case for industrial policy is strong in theory, it is ambiguous in practice. Sceptics raise two practical objections: 

1. Informational objection: can governments accurately identify the relevant firms, sectors, markets subject to those market imperfections? This critique is often expressed by saying “governments cannot pick winners.” 
2. Political capture: can governments withstand lobbying from powerful firms and prevent industrial policy from becoming an instrument of rent transfer to incumbents? 

Neither of these objections is fatal. Similar counter-arguments can be made in other areas of government policy as well – education, health, infrastructure, macroeconomics. In all these areas, public policies can be undermined by lack of good information and political capture by self-interested lobbies. Nevertheless, there is widespread agreement about the need for a public role. Few doubt, for example, that a government needs to have an education policy, even though education externalities are often hard to pin down and insider lobbies (such as teacher unions) can get overly powerful. The debate, however, is rarely about whether governments should have an education policy; it revolves around how to get it right. Similarly, the relevant question for industrial policy is not whether but how. 

Existing studies

Until recently, empirical studies on industrial policies came largely in one of two types. On the one hand, we had detailed country studies of South Korea, Taiwan, India, Brazil, and other countries. Amsden (1989), Evans (1990), and Wade (1995) are classics of this genre, and they give a generally positive take on the success of industrial policies. On the other hand, we had a number of cross-industry or cross-country econometric studies that regress a measure of economic performance, such as productivity or exports, on indicators of government support (e.g. Krueger and Tuncer 1982, World Bank 1993, Lee 1996, Beason and Weinstein 1996).  The latter generally reached negative conclusions on the effectiveness of government subsidies.

Neither set of studies proved entirely convincing to the opposite camp. The country studies had the usual problem that it was difficult to trace the effects of success to specific industrial policies. In any case, perhaps East Asian countries were special after all and their experience did not carry over to others. The econometric studies, on the other hand, suffered from the problem of misspecification: when government intervention is not random and responds in second-best fashion to real market failures, theory suggests the correlation at the industry level between intervention and performance should be negative (Rodrik 2012) – exactly what the studies found, but attributed instead to the failure of the policies!

Moving forward

A new generation of work has been moving us beyond the largely ideological debates of the past to a more contextual, pragmatic understanding. The most recent strand is rooted in two developments. One of these is the indisputable economic success of China, a country that has made liberal use of a diverse array of industrial policies: cheap loans, public ownership, local-content requirements, export subsidies, and technology-transfer requirements. The other is the dissatisfaction with Washington Consensus-type policies, which in Latin America and elsewhere produced weak returns in terms of structural change and productive diversification. 

The Inter-American Development Bank has been at the forefront of the new pragmatic approach, producing a series of case studies of successful and less successful interventions in Latin America. These studies analyse in some detail the nature of public-sector engagement with the private sector in a range of tradable industries (Sabel et al. 2012, IDB 2014, Fernández-Arias et al. 2016). One important difference from the earlier tradition of case studies is that these pay much greater attention to methodological issues and the problems of causal inference. Consequently, they are duly careful about the conclusions that can be drawn. Nevertheless, they provide considerable insight about appropriate institutional frameworks. 

These studies build on existing works emphasising the role of disciplined public-private collaboration as a “search engine” for identifying the most important constraints faced by entrepreneurs, as well as the most appropriate mechanisms for alleviating such constraints (Hidalgo et al. 2007, Hausmann et al. 2005, Rodrik 2007, 2008, Sabel 2007). When designed appropriately, public–private collaboration can ameliorate both of the risks identified above: lack of information and political capture. Their work draws on the experience of successful practitioners (e.g. Ghezzi 2017), while informing them in turn. 

Reconsidering the East Asian experience

A complementary line of analysis reconsiders the East Asian experience using newer empirical and theoretical techniques. For example, Kalouptsidi (2018) has analysed the impact of China’s industrial subsidies in shipbuilding. A special challenge here is that the Chinese system is opaque, and it is often not possible to observe directly the support provided but the government. Using a dynamic model of the global shipping industry, Kalouptsidi backs out Chinese subsidies from demand shocks, testing for a break that occurred when China first prioritised shipbuilding in 2006. The paper concludes that Chinese subsidies were of the order of 13-20% and the subsidies had a significant impact on China’s expansion in shipbuilding, while diverting production away from other, more efficient countries. 

Lane (2017) analyses the effectiveness of South Korea’s big push policies of the late 1970s. He finds that the heavy and chemical industries targeted by the government grew significantly more rapidly than non-targeted industries. The effects were long-lasting: they persisted even when the supports were removed. A novelty of this paper is that Lane also estimates the spillovers of these industrial policies by examining exposure to the policies through the input-output table. He finds evidence of pecuniary externalities: South Korea’s industrial policies ended up also promoting industries that had not been directly targeted, but had high degree of linkages to the targeted activities. 

Welfare-reducing industrial policies 

Even when they are effective in promoting certain industries, industrial policies can be welfare-reducing if the expanding industries are not those that predominantly suffer from market failures. An interesting paper by Liu (2017) tackles this question by considering the role of upstream versus downstream industries. Consider a setting where market imperfections are widely distributed throughout the economy. An implication of second-best theory is that reducing the distortion in industry A may make things worse if industry B is a general-equilibrium substitute for it and also has a large distortion. That is because expanding industry A results in a contraction of industry B. But when industry A is an input (an upstream industry), it will tend to be a general-equilibrium complement to other distorted industries. So reducing the distortion in A makes things better for B (and C, D, E, etc.) as well, since it leads to an expansion of all those industries. This is a case where the second-best interactions magnify the original gain rather than countering and reversing them. Therefore, the more upstream an industry is, the more beneficial it is to support it.

Liu (2017) formalises and generalises this intuition to show that it rationalises industrial policies that subsidise more upstream industries. He then applies the framework to South Korea during the 1970s and contemporary China. He finds in both cases that industrial policies were in conformity with the first-order implications of the theory. 

Conclusion

While it is too early to suggest that research on industrial policy has taken off, these newer studies do advance our understanding of industrial policies on several fronts. We have a better sense of the economic and institutional circumstances under which industrial policy can contribute to economic development. At the same time, globalisation and the spread of new technologies create new challenges for industrial policy that have not yet received adequate attention.

References

Amsden, A (1989), Asia's next giant: South Korea and late industrialisation, Oxford University Press.

Beason, R and D Weinstein (1996), “Growth, economies of scale, and targeting in Japan (1955–1990),” Review of Economics and Statistics 78(2): 286–95.

Evans, P (1995), Embedded autonomy: States and industrial transformation, Princeton University Press.

Fernández-Arias, E, C Sabel, E Stein and A Trejos (2016), “Two to tango: Public-Private collaboration for productive development policies”, Inter-American Development Bank, Washington, D.C. 

Ghezzi, P (2017), “Mesas Ejecutivas in Peru: Lessons for productive development policies,” Global Policy.

Hausmann, R, A Rodríguez-Clare and D Rodrik (2005), “Towards a strategy for economic growth in Uruguay”, Inter-American Development Bank, Economic and Social Study Series. 

Hidalgo, C, B Klinger, A Barabasi and R Hausmann (2007), “The product space conditions the development of nations," Science 317(5837): 482-487.

Inter-American Development Bank (2014), “Rethinking productive development: Sound policies and institutions for economic transformation”, Washington, DC.

Kalouptsidi, M (2018), “Detection and impact of industrial subsidies: The case of Chinese shipbuilding,” Review of Economic Studies, 85, 2018, 1111-1158. 

Krueger, A and B Tuncer (1982), “An empirical test of the infant industry argument,” American Economic Review, 72(5): 1142-52. 

Lane, Nathan (2017), “Manufacturing revolutions: Industrial policy and networks in South Korea,” unpublished paper.

Lee, JW (1996), “Government interventions and economic growth,” Journal of Economic Growth, 1(3).

Liu, E (2017), Industrial policy in production networks, Princeton University. 

Rodrik, D (2007), “Industrial policies for the twenty-first century,” in Rodrik (2007), One economics, many recipes: Globalisation, Institutions, and economic growth, Princeton University Press.

Rodrik, D (2008), “Normalising industrial policy,” Commission on Growth and Development Working Paper No. 3, Washington, DC.

Rodrik, D (2012), “Why we learn nothing from regressing economic growth on policies,” Seoul Journal of Economics, 25(2).

Sabel, C (2007), "Bootstrapping development: Rethinking the role of public intervention in promoting growth”, in V Lee and R Swedberg (eds), On Capitalism, Stanford University Press, pp. 305-341.

Sabel, C et al. (eds) (2012), Export pioneers in Latin America, Inter-American Development Bank, Washington, DC.

Wade, R (1990), Governing the market: Economic theory and the role of government in East Asia's industrialisation, Princeton University Press. 

World Bank (1993), The East Asian miracle, Washington, DC.