In debates about industrial policy, there is often a moment when someone makes the following claim: “Every country has an industrial policy. Even if there is no industrial policy, it is a type of industrial policy. The only relevant question is what kind of industrial policy we should choose.” In my experience, people who make this argument immediately jump into why certain kinds of industrial policy, including tools like subsidies and import restrictions to support specific domestic industries or companies, should actually be very aggressive.
Of course, it is true that every country has some kind of industrial policy, if that term is understood very broadly. But I think it’s useful to think hierarchically about economic policy and its impact on industry.
The most basic layer is an economy with a legal system that enforces contracts, a functioning financial system, functional bankruptcy laws, low inflation, adequate government borrowing, good transportation and communications infrastructure, and a robust education system from K-12 to colleges and universities. Universities, workforce training for adults, etc. These characteristics certainly support the stronger development of the industry, but they do not take sides on which side the industry will emerge.
As a next step, we can imagine the insight that the long-term growth in living standards over the past two or three centuries has been closely linked to advances in science and technology. It is a standard belief among economists that unfettered markets lack investment in innovation. This is because innovations can be replicated and many of their benefits go to the users rather than the inventors. High-income countries therefore subsidize innovation in a variety of ways, including patent and intellectual property protection to increase rewards for successful innovators, tax breaks for companies’ research and development, and direct funding for science and innovation. In research institutes. These kinds of measures are intended to shape the direction of the economy with greater emphasis on technology-based growth. I argued that there is a plausible case for increasing these incentives with the goal of doubling U.S. R&D spending, despite recent modest increases in U.S. R&D spending.
However, a conceptual line can be drawn between general support for R&D and support for industry-specific goals. For example, a society may identify specific technology priorities, such as carbon-free energy production, cancer drugs, strengthening domestic semiconductor production, or artificial intelligence. A certain amount of government support for R&D can be targeted to the desired field. The government may also take other measures. We could reward certain kinds of inventions (think Operation Warp Speed for the development of a COVID vaccine), or allow companies to fund collaborative research or establish co-foundations without fear of antitrust repercussions. We form ventures with the best-performing companies in other countries. However, all of these steps focus on supporting knowledge research and development.
The next step is direct support for an industry or specific specific company. This support may take the form of direct government subsidies or tax breaks for specific businesses and/or industries. It may also involve government involvement in transportation infrastructure or workforce training that very specifically targets industrial development in a particular region.
The final stage of “industrial policy” is to foster domestic enterprises and industries through subsidies, infrastructure, and human resource development, as well as providing support for basic technologies and scientific expertise, as well as to hinder international competition with tariffs and import quotas. It will.
There may be other reasonable ways to classify these categories, but the point I want to emphasize is that using the term “industrial policy” to refer to all of these stages seems to stretch the term too far, making it obsolete. . It’s useful. In my opinion, most economists who consider themselves opposed to “industrial policy” are also supporters of at least the first two or three levels of policy above, namely the basic foundations of a strong economy, including support for research and development. . Instead, I would like to use the term “industrial policy” to focus on subsidies or trade barriers aimed at specific companies or industries.
Sometimes this kind of industrial policy has worked. Whether it’s specialized training for workers, infrastructure investments, land availability, regional research centers, or local tax breaks, there are plenty of local examples where large businesses need government support (or at least active opposition) to grow. (“Tax Increment Financing”), etc. Of course, there are many cases where local governments tried to roll out a red carpet for companies, but lost a huge amount of money without any results. Some will remember the 2000s as one of many examples.In 2018, when President Trump loudly announced that Foxconn would build a massive manufacturing facility in central Wisconsin, It never happened.
Likewise, there are several examples globally where countries have used tariffs and import quotas along with all the other technology, manpower and infrastructure steps mentioned here to help build domestic industries that have become global leaders over time. . However, in cases where it seemed to work, such as certain industries in Korea, government support for these industries was tied to the industries achieving specific export goals that would allow them to be cost-competitive in global markets. If the industry did not meet the target, the subsidy was stopped. And there are many examples of countries blocking imports just to support domestic producers.
But since these types of industrial policies are all driven by politics, they are more likely to react to a combination of powerful incumbent special interest groups and wishful thinking (after all, politicians aren’t putting their money on the line). Many prominent industrial policy efforts have yielded bad results. I wrote the following about my dissatisfaction with industrial policy a few years ago:
For example, in 1991 Linda Cohen and Roger Noll published the book T.Hee Technology Fork Barrelwas based on case studies of U.S. attempts to build infant industries in supersonic airplanes, communications satellites, space shuttles, breeder reactors, solar power, and synthetic fuels. I remember in the 1980s Japan loudly announcing its “fifth generation” computer project, but then quietly fading away. I remember Japan being a shining example of how industrial policy worked in the 1970s and 1980s. But when Japan’s economy decided to turn to computers as it entered a three-decade recession, starting with Brazil, Japan suddenly stopped being a shining example. In the 1970s and 1980s, electricity was produced and Argentina was determined to become a global electronics powerhouse. I remember the economic disaster that was the industrial policy of the Soviet Union. I remember places around the world trying to be the next “Silicon XXXX” but generally not succeeding.
Ultimately, any proposal for industrial policy must grapple with issues of political discipline. As industrial policy moves beyond the basics, such as supporting health organizations and research and development, and begins to focus on specific industries and companies, what are the chances that the policies will be effective? What intermediate goals are used to determine whether the policy is impacting the industry as desired? Will the policy be discontinued if intermediate targets are not met? The more industrial policy is captured by companies in a specific company or industry, the greater the political tension.
There is often a great irony in industrial policy. In the 1950s, the chairman of General Motors (GM) was nominated as Secretary of Defense. According to an anecdote, when asked whether he could separate General Motors’ interests from those of the broader country, he responded, “What’s good for General Motors is good for the country.” This line has been cited for decades to exemplify excessively pro-business attitudes. (As explained here, the story is inaccurate.) But during the Great Depression, when General Motors needed government subsidies to survive, many argued that General Motors was good for the country. Likewise, current U.S. industrial policy favors direct subsidies of billions of dollars to Intel and TSMC companies on the grounds that “the interests of domestic semiconductor manufacturers are good for the country.”
There is an old saying, “Government should steer the direction, not row the oars.” A useful role is to set policies such as appropriate institutions and incentives for innovation both in general and for specific industries. However, as the government entered directly into the business of subsidies and tariffs, it began to coordinate rather than coordinate, and the risk that political incentives began to override rational economic policy began to become a greater risk.
These and other issues have been of greatest interest to me recently, as the most recent issue of the Journal of Economic Perspectives, of which I am editor-in-chief, published an “Industrial Policy Symposium” in its Fall 2024 issue. Like all JEP content and archives, the papers are freely available online.
Want more? most recent Annual Review of Economics It also includes several articles on industrial policy.
