In terms of raw spending, the Inflation Reduction Act (IRA) is the biggest piece of climate legislation ever passed in any country. Ageconmt’s Nick Haggerty recently gave a brief overview of the climate spending provisions. The bill amounts to a firehose of tax credits and other subsidies for all kinds of carbon emission-reducing technologies. Similarly, the CHIPS act passed shortly before the IRA directs $250 billion in federal spending towards semiconductor research and manufacturing, with the aim of keeping pace with China in the semiconductor technology race.
These two bills signify that the US is embracing “industrial policy”, an idea that has generally been out of favor among economists and Western governments in recent decades. Industrial policy, broadly defined, refers to a set of government policies designed to boost domestic production in certain economic sectors. These policies can include sector-specific tax breaks or other government subsidies, favorable trade rules, or simply direct government production (e.g. highways).
The U.S. has occasionally embraced industrial policy, but typically only in response to some external threat. The starkest example is the government-led World War 2 mobilization. After that, competition with the Soviet Union compelled large government outlays towards the space race and other military technologies–this is what eventually led to the development of the internet and GPS technology.
But industrial policy is most often discussed in the context of developing countries. Japan and the other “Asian Tigers” (Hong Kong, Taiwan, Singapore, South Korea) used extensive industrial policy to become high-income manufacturing powerhouses. But other countries, notably in Latin America, were much less successful. Most Western governments and economists in recent decades have promoted an alternative paradigm called “the Washington Consensus”, which is a set of recommendations regarding macroeconomic policy for developing countries (sometimes enforced by the IMF as a condition for financial assistance). The Washington consensus recommends low government spending and borrowing, low tax rates, privatization of state enterprises, and deregulation, among other things. Broadly speaking, it amounts to “where feasible, get government out of the way and let the market rip.”
Economists tend to disfavor industrial policy because they argue that governments should not be “choosing the winners”. Even if a certain industry favored by policy successfully grows, it may have come at the expense of other sectors that would have performed even better. Capital seeks the highest return, and industrial policy may just get in the way. Further, politically powerful sectors can lobby the government for policies that favor them, and this again may come at the expense of more beneficial sectors.
On the other hand, one of the most common arguments for industrial policy is the “infant industry argument”, in which new industries should be protected from global competition until they mature. Another argument is that some industries confer benefits to society that are not captured strictly by market values (or “externalities”). Both of these arguments can be applied to the IRA: China–using extensive industrial policy of their own–has raced ahead of the US in green technologies like solar panels, which promises to be a major growth sector for many years to come. And the subsidies for green tech will have the positive externality of reducing carbon emissions and mitigating the impact of climate change. Another argument in favor of industrial policy is that some products are so important that they need to be produced domestically to ensure a steady supply. This is one of the impetuses for the CHIPS act, which aims to bring semiconductor manufacturing back to the United States so that it is less vulnerable to future supply chain disturbances.
Whatever the theoretical merits, the US’s recent legislative activity implies that, for now, industrial policy is back in a big way. Whether it represents a structural shift in economic policy or a temporary blip remains to be seen.