In my previous post, I argued that the critiques of "neoliberalism" that have become widespread over the past decade (including by yours truly) needed a fresh look in light of contemporary economic conditions. The part of the neoliberal consensus that I want to give special attention in this post is industrial policy, that is, the targeting of specific sectors for government help, through some combination of subsidies, trade protection, low-interest rate loans, and the like.
Economists never liked industrial policy in the heyday of the Washington Consensus, and a large majority still don’t. Industrial policy distorts resource allocation by making it subject to political considerations. Yet industrial policy has become a key element of the Biden administration’s economic plan, embodied in last year’s "Creating Helpful Incentives to Produce Semiconductors" (CHIPS) Act and the amusingly named Inflation Reduction Act. The former provided hefty subsidies to re-shore or friendshore semiconductor production, while the latter promoted transition to alternative energy through a complex series of subsidies and trade restrictions. In a speech at the Brookings Institution in April, National Security Advisor Jake Sullivan outlined a New Washington Consensus that put industrial policy at its core.___STEADY_PAYWALL___
Let’s begin our reconsideration of industrial policy with some general observations about the role of industrial policy in economic history. Many critics of the Washington Consensus pointed out that some of the most spectacular growth stories of the 20th century came from countries like Japan, South Korea, and Taiwan that were heavy users of industrial policy. Their growth models depended on some version of infant industry protection, whereby the government shielded new industries from global competition up to the point where they had come down the learning curve far enough to compete on their own in a global marketplace. At that point the protections were removed, and you had world-beating industries like Toyota, Hitachi, Samsung, and Hyundai. China used similar protections to build up its alternative energy industry, which has now achieved a scale and efficiency that will make future foreign competition very difficult.
It is also not the case that the United States has never used industrial policy historically. The rising industrialists of late 19th century America were all in on protective tariffs, which allowed them to compete against Europe’s better-established manufacturers. They turned to free trade only when they were able to compete on their own and faced countervailing protections as they sought to enter foreign markets. The biggest U.S. practitioner of industrial policy was the U.S. Department of Defense, which funded research, development, and the scaling up of work on semiconductors, radar, computers, the internet, and a host of foundational technologies.
There were of course many counterexamples of failed industrial policy. Argentina wasted a lot of money trying to create a national car, the de Tella, while Indonesia has been struggling to promote a domestic aerospace industry. India has been pouring money into the creation of a domestic semiconductor industry to little effect. State-owned industries often inject politics into the running of commercial enterprises, leading to bloated payrolls, inflexibility, and unsustainable deficits. Industrial policy in the form of infant industry protection can work if the country in question chooses the right sector, employs the right incentives, and is disciplined in its willingness to remove the protections when the infant grows up to be an adult. This is more a political than an economic matter, and the countries of East Asia were in general better able to pull this off than developing countries in Africa, South Asia, or Latin America.
So what about the Biden administration’s industrial policies? It is hard to argue that they are aiming at the wrong targets. Given the rising competition with China, it was not prudent to let advanced semiconductor become so concentrated in Taiwan which may be vulnerable to Chinese intimidation or takeover in the future. And speeding up the energy transition is also an obvious priority: market forces by themselves will not do this nearly fast enough.
The devil is in the details, however, as to how these policies are conceived and implemented. The IRA in particular has become the source of particular controversy because it mixes the promotion of alternatives with a traditional form of economic nationalism, protecting American jobs at the expense not just of ideological competitors like China, but democratic allies in Europe and Asia. Many European and Asian companies are considering moving investments from their own countries to the United States due to the hefty subsidies involved, which has stimulated the EU to consider its own set of Europe-wide subsidies to counteract them. This could lead to a destructive non-zero sum competition that leaves both the United States and its allies worse off.
The IRA has already created a blizzard of rules and exceptions to rules that create perverse incentives for companies. It provides a tax credit for electric vehicles and a removal of the previous cap on subsidies of 200,000 cars per manufacturer. But those tax credits apply to only the lower price range of vehicles, and require that their final assembly be done in North America. However, these rules don’t apply to leased vehicles, and so European and Asian carmakers are seeking to sell through this channel. There are also complicated rules about the dates by which final assembly must take place, with the IRS specifying a list of individual vehicles that qualify for the credit. Battery sourcing is also subject to complex regulations: 50% have to be manufactured in the United States and 40% of the minerals used have to come from American sources. The IRA also specifies a large number of new labor rules that will apply to a series of specific subsectors like green hydrogen, carbon sequestration, refueling stations, and the like.
This policy has been successful insofar as a significant number of new investments in U.S. manufacturing capacity have been announced. The problem with industrial policy in the past, however, is that the complex rules underlying it have grown more complex and rigid over time. There is a danger that subsidies will continue past the point they are needed, because it is too difficult politically to remove them.
In this respect, the American track record has not been so bad historically. The U.S. government invested in several foundational technologies in the 1990s, beginning with the internet, that it then privatized and opened up to commercial development. This of course occured at the height of the neoliberal consensus; whether industries created by government subsidies will be allowed to stand on their own is an open question for the future.
Francis Fukuyama is chairman of the editorial board of American Purpose and Olivier Nomellini Senior Fellow and director of the Ford Dorsey Master’s in International Policy program at Stanford University’s Freeman Spogli Institute for International Studies.
Image: A close up of a circuit board. (Unsplash: Bermix Studio)
American Purpose newsletters
Sign up to get our essays and updates—you pick which ones—right in your inbox.Subscribe