The return of industrial policy has been saved
We would like to thank Jim Kilpatrick for his insightful comments and feedback.
Cover image by: Adamya Manshiva
United States
United States
Industrial policy has a long history. England used tariffs and other policies to support its wool industry in the 14th century. By the early 19th century, other countries had implemented their own policies, including France, which raised tariffs to protect its manufacturing and agriculture sectors, and the United States, which implemented the Tariff Act of 1816 to protect its infant industries.1 Japan used subsidies and trade protection to promote key industries in the 1960s, and South Korea provided loan guarantees to firms in heavy industries in the 1970s.2
Industrial policy likely peaked just before the 1980s ushered in a period of skepticism surrounding government intervention, with the Washington Consensus advocating for privatization and for government to allow the free market to determine resource allocation. However, due to recent disruptions and geopolitical tensions, global policymakers are once again looking to industrial policy to meet their objectives. Recent examples include the US Inflation Reduction Act,3 the European Battery Alliance,4 and Japan’s strategic supply chain initiatives.5
With industrial policy gaining popularity around the world, this paper explores its implications for global businesses. First, industrial policy is unlikely to supercharge economic growth and could create inefficiencies and distortions as protectionist measures have been included in many of them. Furthermore, such policies are unlikely to tame inflation. On the positive side, such policies have a reasonably strong record of pushing past the technological frontier, which suggests that they could yield improvements in semiconductors and green technology. More protectionist policies are likely to result as governments compete with one another to offer the most attractive incentives. Business leaders will need to adapt to this rapidly changing landscape.
Industrial policy is generally any government intervention that provides support for a particular sector or industry. Interventions such as tariffs, trade restrictions, and subsidies can help shield a domestic industry from import competition. Governments can provide tax credits or direct funding to encourage investment, while government procurement can be used to boost demand in the sector. Although industrial policy can be used for any sector, historically such policies have mostly focused on heavy industry, technology, energy, agriculture, and anything that could have military applications.
Industrial policy is at its best when it focuses on information and coordination problems that arise when establishing a new sector. Information problems are related to cost discovery in an economy where producing a product and making a profit from it have been unproven. Once that product is proven to be profitable, new entrants can come in and compete without having to make the initial investment for discovery. Industrial policy, therefore, can be used to either cover the cost of that discovery or protect the firms that made such an investment. The coordination problem arises when there are deficiencies in upstream or downstream industries that need to be overcome. For example, energy infrastructure may need to be improved before a steel manufacturing plant can be established.
Solving these problems requires continued information exchange between the private and public sectors. However, that information exchange needs to occur without the public sector being captured by the private sector, and without the public sector picking “winners.” When that happens, governments end up subsidizing losing ventures, wasting taxpayer money, and potentially preventing more productive entrants from gaining a foothold. An extreme example of industrial policy gone wrong occurred in Tunisia where policy was used in the late 1990s and early 2000s to protect banking, telecom, and transport industries from competition.6 At the time, the president’s family had business interests in these sectors, and their firms ultimately accounted for 21% of profits in the economy but only 1% of private sector output and 3% of employment.
Industrial policy is often used to increase economic growth in a country. Developing countries have used it to improve living standards, while developed countries have used it to bolster job growth. Although the current wave of industrial policy has been more focused on securing supply chains, reducing greenhouse gas emissions, and increasing technological capabilities, proponents of many of these policies have argued that they also support strong economic growth.
The results of using industrial policy to supercharge an economy have been mixed. The Peterson Institute for International Economics found varied economic outcomes of industrial policy in the United States. Industrial policies that involved either trade measures or subsidies to specific firms had few successes in generating new employment growth at a reasonable cost, though some seemed to prevent employment from declining. On the other hand, industrial policies that focused on public and private research and development had a much better track record of creating new jobs.
It should also be noted that researchers do not always agree on the efficacy of industrial policy. For example, some economists see a causal link between protectionist policies in Taiwan and South Korea and their historically high economic growth rates.7 However, other economists have shown that growth in these countries might have occurred without industrial policy.8 The latter group suggests that the reduction of barriers to trade is the true driver of their economic growth miracle. Indeed, Latin America’s experience with industrial policy beginning in the 1950s significantly raised barriers to trade, shielding domestic producers from international competition and making them highly inefficient.9 This led to a sharp decline in economic activity and a reversal of the initially protectionist policy.
Several of the more recent industrial policies will likely support economic and job growth. The most promising policies are those that involve overcoming information challenges. The European Union’s (EU’s) Horizon Europe policy, which provides funding for research and development across numerous categories, such as energy and digitization, and the Japan-US Leading-edge Semiconductor Technology Center, which will perform research and development for the next generation of chips, have the best shot at creating economic gains.
Other industrial policies should also create jobs. The US Inflation Reduction Act, US Chips Act, EU Chips Act, and European Battery Alliance will likely have positive effects on employment, at least in the short term. However, these policies are unlikely to have a significant effect on national economic growth. For example, the US Inflation Reduction Act amounts to US$369 billion of incentives and tax credits for businesses.10 However, this large sum is spread over 10 years, representing less than 0.2% of US GDP. Plus, the bill is expected to raise tax revenue and modestly reduce deficit, suggesting that the direct intervention from the government is modestly contractionary.
Proponents of incentives for semiconductor production claim that such investments have very large economic spillover effects.11 However, indirect and induced employment gains from investment are often not realized. Recent research shows that, at the local level, tax incentives to attract business investment have no discernible economic effect apart from the direct jobs added,12 suggesting that induced employment gains, nationally, may also be negligible. This means that the 1.85 million jobs expected from the US Chips Act could fall to a number as low as 300,000.
Protectionist measures imbedded in some of these policies, such as the US Chips Act, EU Chips Act, and similar policies coming from South Korea and China, are likely to slow economic growth over the longer term. These policies encourage production to be relocated within their national borders, rather than in the most efficient location. Although such a move may improve national security, it raises the cost of production and allocates resources to less efficient activity, slowing economic growth over the long term.
Although the title of the IRA suggests that inflation will fall under the policy, it is unlikely to do so, at least in the near term. Yes, there are provisions to keep pharmaceutical prices down, which is helpful, but that will do little to restrain headline inflation. For example, all pharmaceutical components account for less than 1.5% of total US consumer price index. Energy prices could fall over the longer term as investment in renewable energy lowers prices per megawatt hour. However, energy-related disinflation related to investment associated with the IRA will likely take many years to materialize. Researchers at the University of Pennsylvania estimate that the effect on inflation will be negligible over the next decade.13 At the same time, the associated fall in demand for fossil fuels could create price spikes. For example, the lack of expected demand for natural gas could cause a decline in drilling investment, which could squeeze supply faster than demand, thereby raising prices significantly well before the energy transition is complete.
The considerable investment in semiconductors is likely to push prices down. The United States, Japan, China, the EU, Taiwan, and South Korea have all developed policies to support the production of semiconductors. The abundance of global incentives could lead to an oversupply of semiconductors. Post the widespread shortages after the pandemic hit, chipmakers were struggling with a glut of supply by the end of 2022.14 A fall in semiconductor prices, however, will only extend the multidecade trend of declining prices that were disrupted when the pandemic led to shortages. The potential fall in semiconductor prices is also not guaranteed. The policies in the United States and the EU explicitly target domestic production. Labor costs in these regions are considerably higher than those in Taiwan, South Korea, and China, where many of today’s semiconductors are produced. Employees involved in one of the new chip factories in the United States privately complained about elevated costs related to production in the country.15
Many of the recently implemented and planned industrial policies are less focused on economic growth, job creation, and inflation reduction, and are far more focused on achieving climate goals, securing supply chains, and pushing past the technological frontier. The likelihood of achieving these goals is relatively high.
The largest piece of climate change–related industrial policy is the US Inflation Reduction Act. Under this new legislation, emissions are expected to fall by as much as 20 percentage points, though most estimates put it between 10 and 15 percentage points lower.16 If these estimates are correct, they would represent significant progress toward the United States’ climate goals. The major drawback of the US policy is that it is all carrots. The lack of sticks in the policy could lead to more government waste than necessary. For example, companies will continue to qualify for the associated tax credits regardless of their ability to reduce emissions.
Pushing past the frontier of semiconductor technology via industrial policy is particularly likely. Taiwan has the most advanced capabilities regarding the fabrication of semiconductors, thanks in large part to its policies aimed at protecting the industry. The Japan-US initiative to provide research and development for the industry through the creation of Rapidus Corporation could make further advancements. As already mentioned, industrial policy focused on information challenges typically has the best track record. For example, the United States has produced multiple cutting-edge technologies, such as the internet and global positioning system,17 through the Defense Advanced Research Projects Agency (DARPA), a research agency of the Department of Defense. The DARPA was created in response to the Soviet Union in 1958, which has clear parallels to today’s US-led policies aimed at competing with or restraining China. The subsidized research provided via Rapidus Corporation could yield similar results. However, the process of how the organization functions will be critical in determining how successful it is. Pushing past the technological frontier is understandably difficult and likely requires some particular features, such as limited tenure of program management, a sense of mission, and tolerance for failure.18
Efforts to secure supply chains of critical materials will likely provide mixed results. On the upside, the US CHIPS Act has generated plenty of interest in producing semiconductors in the United States. An increase in US capacity should reduce its vulnerability to political shocks. The United States, as well as the rest of the world, is heavily reliant on Taiwanese production for advanced chips, raising concern about reliable access to crucial technology. However, the United States has recently stipulated that chipmakers taking US funds under the CHIPS Act cannot expand capacity in China over the next decade and has banned US companies from exporting the technology or machinery used to create advanced chips to China.19 These stipulations may create new challenges to attract the most advanced chipmakers, which already have plants in China.
It is too soon to determine exactly how successful Japan’s incentives to move supply chains of medical goods and semiconductors out of China have been or will be. We know that the share of Japan’s imports coming from China fell from 23.5% in 2019 to 21.0% in 2022, a potential sign of the policy’s success. However, Japan’s incentives aimed at relocating supply to Association of Southeast Asian Nations countries, where the share of goods imports in 2022 was unchanged from the share in 2019. Meanwhile, goods imports as a share of GDP have increased markedly in 2021 and 2022, indicating that domestic production has not met domestic demand. Although none of this is conclusive evidence that supply chain incentives have been ineffective, it does suggest that the effects are so far limited.
Efforts to build battery capacity in the United States and Europe should reduce risks to the supply of this critical product. However, eliminating risks to these supply chains altogether will be virtually impossible. For example, batteries often require critical materials that are not available domestically. For example, the United States has no known economically viable reserves of manganese, which is used in some batteries. For Europe, the only major producer of manganese is Ukraine, which presents obvious risks to the supply of the metal given its ongoing war with Russia. In addition, the production of manganese is overwhelmingly concentrated in South Africa, followed by Gabon and Australia. Ukraine produces less than 10% of the manganese that South Africa produces.20 The materials required to make batteries continue to change, but the minerals needed will challenge the supply chain resilience of batteries.
Building cutting-edge semiconductor production in Europe faces demand-side challenges. Although the EU plans to develop advanced chip production domestically, it is not clear who would purchase those chips. The EU lacks domestic chip designers, and US designers are likely to tap either US-based producers or cheaper ones in Taiwan and South Korea. This presents a challenge to Europe’s efforts in developing domestic production of advanced semiconductors.
The resurgence of industrial policy will almost certainly create a wave of knock-on regulatory effects. After the United States passed the Inflation Reduction Act, the EU mobilized quickly to respond with its own climate-related industrial policy. The EU feared that American subsidies would divert investment away from Europe. Similarly, the EU produced its own Chips Act in response to the US policy of the same name. China and South Korea have responded similarly. This creates an opportunity for companies in sectors related to emissions reduction or semiconductors to benefit from competing subsidies across the globe.
One downside is that some countries may not respond with their own subsidies, but rather with challenges at the World Trade Organization (WTO). Cases brought before the WTO often take years to adjudicate. In the meantime, plaintiff countries can respond with countervailing duties to correct the imbalance that such subsidies have allegedly caused.21 This could quickly raise the cost of international trade between the plaintiff countries and those that have implemented industrial policy. Other countries may respond with lower barriers to trade. For example, the United States has embedded content requirements in some of its recently passed legislation, meaning that inputs must either be sourced from the United States or from trade partners with which it has a free trade agreement.22 Neither the EU nor the United Kingdom has a free trade agreement with the United States. Pursuing such an agreement would ensure that their domestic companies can benefit from US subsidies.
Widespread use of government subsidies could also encourage currency intervention. Smaller economies are unable to provide the sizable incentives that a country like the United States has recently implemented. For their businesses to remain internationally competitive in the face of these subsidies, currency manipulation can be used. Indeed, several countries have likely benefitted from artificially low exchange rates. Taiwan’s semiconductor industry has benefitted from a very weak Taiwanese dollar, which has yet to recover the value lost during the Asian Financial Crisis in the late 1990s.
Despite its shortcomings, industrial policy is clearly back in fashion, and the governments’ hands-off approach to the economy over the last 40 years has come to an end, at least for now. As industrial policies proliferate, businesses should consider taking the following actions:
Mehdi Shafaeddin, How did developed countries industrialize? The history of trade and industrial policy: The cases of Great Britain and the USA, United Nations Conference on Trade and Development, December 1998.
View in ArticleAndrei Levchenko and Jaedo Choi, “When industrial policy worked: The case of South Korea,” Centre for Economic Policy Research (CEPR), November 9, 2021.
View in ArticleCongress.gov, H.R.5376 - Inflation Reduction Act of 2022, August 16, 2022.
View in ArticleEuropean Battery Alliance, “Building a European battery industry,” accessed May 24, 2023.
View in ArticleJapan External Trade Organization, “Manufacturing government initiatives,” accessed May 24, 2023.
View in ArticleBob Rijkers, Caroline Freund, and Antonio Nucifora, “All in the family: State capture in Tunisia,” Journal of Development Economics 124 (2017): pp. 41–59.
View in ArticleDani Rodrik, “Getting interventions right: How South Korea and Taiwan grew rich,” Economic Policy 10, no. 20 (1995): pp. 53–107.
View in ArticleArvind Panagariya, Debunking protectionist myths: Free trade, the developing world, and prosperity, Cato Institute, July 18, 2019.
View in ArticleWerner Baer, “Import substitution and industrialization in Latin America: Experiences and interpretations,” Latin American Research Review 7, no. 1, pp. 95–122.
View in ArticleU.S. Department of the Treasury, “Treasury announces guidance on Inflation Reduction Act’s strong labor protections,” press release, November 29, 2022.
View in ArticleSemiconductor Industry Association, Chipping in: The positive impact of the semiconductor industry on the American workforce and how federal industry incentives will increase domestic jobs, May 2021.
View in ArticleCailin Slattery and Owen Zidar, “Evaluating state and local business incentives,” Journal of Economic Perspectives 34, no. 2 (2020): pp. 90–118.
View in ArticleAlex Arnon and Kent Smetters, “Inflation Reduction Act: Comparing CBO and PWBM estimates,” University of Pennsylvania, August 5, 2022.
View in ArticleAsa Fitch, “For chip makers, the flip from shortage to glut intensifies,” Wall Street Journal, November 4, 2022.
View in ArticleJohn Liu and Paul Mozur, “Inside Taiwanese chip giant, a U.S. expansion stokes tensions,” New York Times, February 22, 2023.
View in ArticleMelissa Barbanell, “A brief summary of the climate and energy provisions of the Inflation Reduction Act of 2022,” World Resources Institute, October 28, 2022.
View in ArticleGary C. Hufbauer and Euijin Jung, Scoring 50 Years of US industrial policy, 1970–2020, Peterson Institute for International Economics, November 2021.
View in ArticleCongressional Research Service, Defense Advanced Research Projects Agency: Overview and issues for Congress, August 19, 2021.
View in ArticleNikkei Asia, “U.S. CHIPS Act fund bars chipmakers from China expansion for 10 years,” February 28, 2023.
View in ArticleMelissa Pistilli, “Top 10 manganese-producing countries,” Investing News, April 25, 2023.
View in ArticleJames Bacchus and Simon Lester, “Trade justice delayed is trade justice denied: How to make WTO dispute settlement faster and more effective,” Cato Institute, November 20, 2019.
View in ArticleWhite House, Building a clean energy economy: A guidebook to the Inflation Reduction Act’s investments in clean energy and climate action, January 2023.
View in ArticleWe would like to thank Jim Kilpatrick for his insightful comments and feedback.
Cover image by: Adamya Manshiva