The Return of Industrial Policy

Apr 12, 2010 Dani Rodrik

Industrial policy is back, openly embraced by governments in countries like Britain and France, and by prestigious organizations like the Word Bank and McKinsey. The standard rap against industrial policy – that governments cannot pick winners – is largely irrelevant: what determines success in industrial policy is the much less demanding capacity to let losers go.

CAMBRIDGE – British Prime Minister Gordon Brown promotes it as a vehicle for creating high-skill jobs. French President Nicolas Sarkozy talks about using it to keep industrial jobs in France. The World Bank’s chief economist, Justin Lin, openly supports it to speed up structural change in developing nations. McKinsey is advising governments on how to do it right.

Industrial policy is back.

In fact, industrial policy never went out of fashion. Economists enamored of the neo-liberal Washington Consensus may have written it off, but successful economies have always relied on government policies that promote growth by accelerating structural transformation.

We hope you're enjoying Project Syndicate.

To continue reading, subscribe now.

Already have an account? Log in

Support High-Quality Commentary

For more than 25 years, Project Syndicate has been guided by a simple credo: All people deserve access to a broad range of views by the world's foremost leaders and thinkers on the issues, events, and forces shaping their lives. At a time of unprecedented uncertainty, that mission is more important than ever – and we remain committed to fulfilling it.

But there is no doubt that we, like so many other media organizations nowadays, are under growing strain. If you are in a position to support us, please subscribe now.

As a subscriber, you will enjoy unlimited access to our On Point suite of long reads, book reviews, and insider interviews; Big Picture topical collections; Say More contributor interviews; Opinion Has It podcast features; The Year Ahead magazine, the full PS archive, and much more. You will also directly support our mission of delivering the highest-quality commentary on the world's most pressing issues to as wide an audience as possible.

By helping us to build a truly open world of ideas, every PS subscriber makes a real difference. Thank you.

Subscribe Now

new comment has been posted. new comments have been posted.

0 Comments on this paragraph, 3 Comments on this article

Stan Sorscher Jan 3, 2016

Industrial policies operate best when they address market failure. Industrial policies intervene in markets to prevent market failure.

R&D is a case in point. Markets fail to pursue research.

Markets also fail through market domination, information asymmetries, equity, and corruption.

Industrial policies require strong institutions of civil society to "serve society at large." We may have gone through a critical point - too far toward plutocracy - for industrial policies to have a chance.
A new reply to this comment has been posted.
Parrain Boursorama Jan 12, 2014

The best model for expanding Alternative Energies and Environmental Protection globally is through using market equilibrium, whereas governmental subsidies and fiscal stimulus to be just supplementary
A new reply to this comment has been posted.
Michael Heller Mar 19, 2012

There is a naive quality about Rodrik's belief in government’s capacity to shape efficient economic designs. Since his writing is very influential as well as wrong-headed, it needs refutation.

First, he calls for various forms of organized government-business collaboration. In the advanced societies this was called corporatism. Despite having the advantage of modern state institutional capacity Germany and the UK largely gave up corporatism when it became obvious that corporatism was inflexible and non-evolutionary. In developing countries without formal divisions of powers and democratic accountability, government-business 'embeddedness' is more likely to simply take the form of collusion and rent seeking. Rather than engineering “financial incentives” or utopian “states of mind”, governments can more usefully focus on the economic incentives that nature provides by improving regulatory framework to ensure formally equal playing fields for competitive markets in which the incentive for cost discovery is the possibility of extinction.

Second, Rodrik calls for temporary and performance-based government incentives. This is pie in the sky. No political system has ever been devised that could withstand the interest-group pressures implied by any state-activist form of promoting business. It may reasonably be said in line with Schumpeterian theory that government’s job is to ensure that monopolistic privileges generated by market actors can only ever be temporary. It is not, however, government’s role to generate those privileges in the shape of subsidies or protection given in return for state-defined performance targets.

Third, Rodrik calls for transparent and accountable decision making in industrial policy, leaving the field open to new entrants. The welfare states in advanced democracies are already overloaded. It is absurd to suggest that governments should actually expand their field of microlevel discretionary decision making about individual enterprises. Authoritarian industrial policy is bad enough. ‘Democratic’ industrial policy subject to public choice lobbying dynamics would be worse!

Fourth, Rodrik claims there is a difference between picking winners and identifying losers. The distinction is spurious. It would not be necessary to identify a loser (and withdraw a public subsidy) unless there had already been a wasteful effort to pick a winner. Whilst it is true that learning is by doing, i.e. by trial and error, the prior task of policymaking can reasonably be to predict the unintended consequences of stupid policies. There is no point in making errors deliberately for the sake of learning. The onus of cost discovery should lie with business (when it fails) rather than the government that foots the bill. In any case, monitoring of business errors by government requires an unlikely army of politically detached and superbly disciplined bureaucrats. Even if bureaucrats had enough expert knowledge of economic life to pick winners and identify losers (which they don’t) it would still be a costly, risky and unnecessary task for government to undertake.

Fifth, there is of course evidence that industrial policy can succeed, for a while. But there are just as many examples of its catastrophic failure. If we leave aside island or city states such as Singapore, there is not a single country in the world in which the types of large-scale and prolonged neoactivism Rodrik proposes in his most recent book have not eventually led to political decay and economic crisis. Just look at Japan. (China cannot possibly sustain its present path.)

I describe the Weberian counter-logic to Rodrik’s influential neoactivism more fully in chapter six of my book Capitalism, Institutions, and Economic Development (2009/2011).

Michael G. Heller
A new reply to this comment has been posted.

https://prosyn.org/HLwbo9O

The Return of Industrial Policy

To continue reading, subscribe now.

Support High-Quality Commentary

New Comment

Email this piece to a friend

Contact us

Log in/Register

The Return of Industrial Policy

To continue reading, subscribe now.

Support High-Quality Commentary

Edit Newsletter Preferences

Set up Notification