GENEVA — Industrial policy (IP) is back — or rather, back in fashion. Of course, it never really went away, even in countries formally adhering to free-market principles. But the postcrisis world — in which government intervention in the economy has gained greater legitimacy — will see more of it. Likewise, China’s success, and the temptation to bandwagon on its development model, has reinvigorated IP’s appeal, as have better policy tools and greater experience of what works and what doesn’t — a point well argued by Justin Lin of the World Bank.

Indeed, a debate in The Economist last year, led by professors Josh Lerner and Dani Rodrik of Harvard University, ended with 72 percent of voters expressing faith in the merits of IP. Policymakers seem to be of the same opinion, and not just in developing countries, judging by the European Union’s launch of its 2020 flagship last year and the United States’ green energy policy.

But, for developing countries, the old dangers of IP continue to apply. First, policymakers often get it wrong, both when picking which industries to support and in implementing support mechanisms. Second, policymakers are prone to “capture” by vested interests, especially in relatively weak policy environments, leading to favoritism, inefficiency and waste.

Moreover, compared to IP’s previous Golden Age, there are several new risks nowadays. The first concerns the temptation to follow China’s lead unquestioningly. Policymakers must recognize that the Chinese model includes features peculiar to its gradual introduction of market mechanisms — which, compared to IP’s new converts, represents a move in the opposite ideological direction.

Indeed, China’s growth rate has been boosted over the last two decades by the country’s demographic and land-resource “bonuses,” which enabled it to maximize the benefits of globalization. Many other developing countries simply cannot emulate this success in all respects; they must formulate plans specific to their own natural endowments, institutions, and business environments.

The second risk stems from the globalized character of virtually every industry commonly considered a candidate for support. Whereas the original concept of IP involved shielding industries from international competition, today’s world requires integrating local productive capacity in global value chains. That implies the need for policies based on exposing industries to international competition.

The third risk is that industrial policies run afoul of international commitments and obligations. These include not only World Trade Organization rules, but also proliferating regional and bilateral trade and investment agreements — all of which have sharply reduced the scope for IP formulation by limiting options for protecting and supporting industries and businesses selectively.

Despite these additional dangers, none of the major multilateral development institutions could easily argue today that developing countries should not formulate a development strategy that envisions sector-specific sources of economic growth, priorities for industrial development, and government support of such development with fiscal, financial and regulatory measures.

So, how should developing country policymakers go about it?

Pick the right winners. Industrial policies must be based on a country’s factor endowments, and should build on concrete opportunities to integrate industries and firms in global value chains — for example, by deepening existing linkages with international production networks and export markets — while avoiding overinvestment in international growth laggards.

Moreover, policymakers should consider which industries deliver the biggest development bang for the buck: export-generating industries do not always have the greatest impact on employment and value added. Domestic industries, including services, which often account for more than half of value added even in developing economies, should not be neglected.

Be prepared to let losers go. Even the most obvious choices for IP support, seemingly sure winners, will sometimes disappoint in today’s uncertain economic environment. Governments should recognize mistakes and withdraw support before it becomes entrenched or too costly.

Investment policies should extend to both domestic and foreign direct investment. In particular, this means nurturing international competitiveness by focusing on industry productivity, not just on support for domestic players. Similarly, technology policies should encompass both domestic development and technology transfer. And enterprise development policies should address linkages with multinational firms. In today’s world, development objectives can realistically be met by supporting industrial growth, not only domestic industrial growth.

Because international trade and investment regimes ensure access to foreign markets and investors, they are a prerequisite for industrial policy. Even so, developing country policymakers can build some measure of freedom into new trade and investment agreements. At the same time, they should focus on those IP measures that run the least risk of clashing with international obligations: regulatory facilitation rather than restrictions, investment in infrastructure rather than in specific economic activities, and fiscal incentives that are accessible to all.

Finally, policymakers in developing and developed countries alike must recognize that global economic governance discussions cannot focus exclusively on monetary and exchange-rate issues. As more countries embrace IP, competition and conflict is bound to intensify. Avoiding a global race to the bottom in regulatory standards, a race to the top in incentives, or a return to protectionism, will require better global coordination.

James Zhan is director of the Investment and Enterprise Division of the United Nations Conference on Trade and Development (UNCTAD). © 2011 Project Syndicate

In a time of both misinformation and too much information, quality journalism is more crucial than ever.
By subscribing, you can help us get the story right.