We are all interventionists now. In the US, not long ago the bastion of free market thinking, fear of China, worries over the security of supply chains, aspirations for re-industrialisation and hopes of a green transformation are combining to reshape trade and industrial policies. The EU shares US worries over China, mostly in terms of the technological threat. But it is also concerned by the “America First” character of US policymaking, notably the $369bn Inflation Reduction Act. This growing belief in the ability of governments to reshape their economies for the better may have been inevitable, given economic disappointments and geopolitical tensions. But what does it imply?
A big question is what these shifts towards economic nationalism and interventionism will do to the world economy. As things stand today, deep disintegration seems unlikely, though it is, alas, imaginable. It would also be very costly, as Geoeconomic Fragmentation and the Future of Multilateralism, a recent discussion note from the IMF, points out. Moreover, the deeper the disintegration the bigger such costs will be. Technological decoupling would be the most costly of all, especially for emerging and low-income countries. Beyond this are the inevitable geopolitical costs. As James Bacchus, former head of the World Trade Organization appellate body, has rightly noted, containing these costs in today’s world poses huge challenges.
A narrower question is how well the new interventionism will work in its own terms. Will the US federal government, which is the most active and potent player, obtain the results it wants from the policies it is now committed to employ? There are good reasons for doubt. Successful intervention is hard.
It is not that theoretical arguments for intervention are lacking. On the contrary, ever since Alexander Hamilton, arguments for infant industry protection (and other such interventions) have been well known. The core argument is that markets on their own will fail to exploit available opportunities. Harvard’s Ricardo Hausmann has recently restated these arguments. To such infant industry arguments we can add those for protecting economic, technological or military security.
Yet in practice it is quite difficult to make such interventionism work. Too often, for example, it is assumed that the successes of Japan, South Korea and more recently China are due to far-sighted government interventionism. This is exaggerated: the main engine was market competition. Moreover, government intervention becomes more difficult the closer an economy is to the technological frontier: innovation is usually harder than copying. Not least, there is a political economy of intervention, with losers picking governments rather than governments picking winners. The more open to lobbying a state is, the greater the chances of such capture. This is particularly applicable to the US.
Helpfully, in 2021, the Peterson Institute for International Economics published a briefing entitled Scoring 50 Years of US Industrial Policy. It details some grotesquely expensive policies of industrial protection, noting that “US consumers and taxpayers are currently paying more than $900,000 a year for every job saved by Trump’s steel tariffs, extended by Biden”. Sometimes, alas, bipartisanship can be foolish.
What did work? As expected, the star has been Darpa, perhaps the most successful innovation programme in world history. Another success was Operation Warp Speed, the vaccination programme of the Trump administration — a triumph many Republicans have wished to disown. Another was the North Carolina Research Triangle Park. Encouragement of foreign automobile assembly worked quite well, as did tax credits for solar panels.
Yet what is striking is how often such programmes failed to make industries competitive, save jobs at reasonable cost or advance the technological frontier. This was notably true for trade measures and firm-specific subsidies. The big successes were in combining public and private outlays on research and development, as one might expect. Given this, one must wonder whether today’s subsidy programmes will work.
Against this, there are legitimate security reasons for promoting production of computer chips, whatever the costs. Again, in the absence of better policies, subsidies for the green transition should push the economy in the right direction. Moreover, subsidies have the advantage of being transparent, while protection is a hidden tax on consumers transferred to producers. Tariffs also bias production towards the home market, while subsidies are neutral between domestic and foreign markets. Yet subsidies are not neutral across countries: those with the deepest pockets will win. Moreover, subsidies, especially subsidies limited to domestic producers, will cause friction, including with allies. The outcome will be a subsidy war. This may lower emissions of high-income countries. But it will not solve global climate change, which depends on successful co-operation towards a global transformation.
The new interventionism has many causes and many goals. In theory, it might lead to better outcomes, especially where the case for government intervention is strong, as with climate change or national security. But there are also large potential risks, not least that many of these programmes will turn out to be a huge waste of money, as so many interventionist programmes have been in the past. Moreover, these interventions will worsen the trade wars now under way. Fragmentation is very easy to start. But it will be hard to control and even harder to reverse.
Follow Martin Wolf with myFT and on Twitter