This is part of a series, “Economists Exchange”, featuring conversations between top FT commentators and leading economists
In the past decade, the number of national industrial policies globally has more than doubled. Challenges such as the pandemic, the cost of living crisis and climate transition have reignited an appetite for governments to play a more active role in shaping domestic economies. Meanwhile, geopolitical tensions have spurred a drive towards strategic autonomy in sensitive sectors such as defence. Billion-dollar packages including subsidies and investment incentives such as the US’s Inflation Reduction Act and Chips Act are already shaping business decisions and threaten a global subsidies race.
Now that government programmes are also leaning towards more protectionist elements, this wave of national industrial policy contrasts with decades of globalisation underpinned by free trade. According to the World Economic Forum’s latest Chief Economists’ survey, most experts think this paradigm shift will become the de facto approach to economic policy over the coming years. Yet, more than two-thirds think it will stifle competition and lead to problematic increases in sovereign debt levels — few think it will lead to an increase in global activity or resilience.
Adam Posen, president of the Peterson Institute for International Economics, has been a rare vocal critic of what he dubs the rise of “zero-sum economics”. In an op-ed earlier this year, he outlined four fallacies of agendas like US president Joe Biden’s IRA: “That self-dealing is smart; that self-sufficiency is attainable; that more subsidies are better; and that local production is what matters.”
We discussed the comeback of government interventionism, the flaws of programmes such as the IRA and what a more effective approach to industrial strategy and the climate transition might look like. Posen also offers his thoughts on how the UK ought to shape its own long-term economic agenda.
Tej Parikh: How did we get to this moment — when arguing for a more interventionist role for government in our economies has become mainstream?
Adam Posen: I think there was a legitimate delegitimisation of excessive reliance on market outcomes. The financial crisis of 2008 was largely the result of deregulation and supervisory failures. Then some came out very strongly against austerity, arguing that secular stagnation and low rates meant there was room for fiscal spending. I think those two things created the groundwork, but ultimately it is political self-delusion. You’ve got people throughout the west who are falsely claiming that the economic downsides of activist industrial policy are avoidable and that the main problem has been foreign trade.
TP: Recent shocks like the pandemic and cost of living crisis, all of which justifiably involved state support, also give the narrative staying power. Plus, the urgency of the climate transition needs a joint public-private effort.
AP: Yes. And once you put narrow government interventions into place, they tend to stick around, expand and get distorted by entrenched special interests. But over time, they tend not to serve what they were initially supposed to do. That to me is part of the argument against the Biden administration’s economic policy — and the UK Labour party’s current proposals, if we look at Rachel Reeves’ recent speech — they think that over time their programmes are not going to be subject to corruption or engender retaliation from other countries outweighing the benefits. That is mistaken.
TP: Industrial policy has been around for a while — what do you see as the most egregious elements of this wave?
AP: What makes today’s version worse isn’t just that they are large-scale and wasteful. The first big problem is viewing industrial competition as zero sum; the idea that you can create lasting comparative advantage so that your locally headquartered companies dominate an industry. The reason that’s bad is because a) it doesn’t usually work, and b) it just invites retaliation. Second, today’s industrial policy efforts are being combined with a bunch of antitrade measures, not just on tariffs, but also local content requirements and barriers to investment.
Third, about green technology, the most important thing is to get the best technologies out and as widely adopted as possible. This subsidies war combined with trade barriers and domestic investment incentives means we’re likely to see a repeat of what happened with vaccines, which is rich countries hoarding them, and developing countries having to go cap in hand to the big blocs. As a result, we will get far too piecemeal, far too slow a spread of the best green tech, and a lot of resentment, leading to slower take-up of it.
The final thing is cost, in the US case in particular, because much of the IRA and Chips Acts spending is open-ended producer subsidies. They’re not putting in this much money to create this factory or that infrastructure. They’re saying the more you produce, the more money you’ll get, and that’s open-ended.
TP: Is this the end of comparative advantage as an organising principle for the global economy?
AP: No, not the end. As we’ve seen when Napoleon tried with the Continental System to blockade the British Isles over 200 years ago, and other embargo attempts since, comparative advantage is real, not easily denied or created. There is a reason why the supply chains pre-Covid developed organically. In the end, cost-advantaged or desirable products will get through by some means, as with smuggled drugs or counterfeit movies. There will be less variety and availability, and prices will be higher which makes everyone worse off. But you are unlikely to create a whole new industry out of these measures that will move the needle long-term and the futile effort is going to be costly.
TP: Still, it feels like the notion of comparative advantage needs reiterating. So, what do you think are the economic fallacies at the heart of agendas like the IRA?
AP: Evidence supports the idea that government spending on R&D, worker training, infrastructure and expedited regulation of innovations is positive. Once politicians start handing out money to individual companies for specifically located production, though, and they start favouring those companies over potential rivals, it becomes a bad dynamic.
There’s nothing like having the state as your de facto guarantor. A host of bad things happen to society as a result. We’ve discussed how on the international front it leads to retaliation, shutting out of poor countries, reduced adoption of new technologies, and corruption.
But the domestic effect on any country that goes down this path is worse. You end up with entrenched incumbent companies becoming a political sacred cow, as we have seen with state-owned enterprises in China. This spirals because when you don’t have enough competition in key industries, not only do consumers overpay, you slow innovation. You have unfairness from incumbents’ political weight being thrown around. You crush new entrants and dynamism.
TP: Yes, and trying to replicate entire supply chains is enormously inefficient. Duplicating green or chip technology supply chains runs into the trillions.
AP: Trying to achieve self-sufficiency in any major industry that isn’t a simple extractive one like a mine is self-defeating. The reason is the value of diversification. Yes, attenuated supply chains, dependent upon potentially hostile hosts, are a vulnerability. But so is having all or most of one’s production at home, subject to natural disasters, climate shifts, unstable politics, domestic terrorism, and undependable or deficient production due to the corruption of too-big-to-fail local producers. Russia and North Korea have worked very hard to be self-sufficient, with limited supply chains, and it has not worked out well for them.
TP: So, hypothetically, let’s assume that this type of industrial policy will remain the new status quo across the world. What will things look like in 10 years’ time?
AP: The real damage from decoupling and conflict between the US, China and other economic blocs is reduced productivity growth. We would see less diversification both financially and in inputs, including of ideas and business practices, along with less competition, which directly diminishes productivity. We would also see further restrictions of migration, foreign direct investment, flows of information and technology once economic nationalism is entrenched.
So, if we continue down this path, we’re looking at a meaningfully bleaker outlook for average growth in the world. It’s going to be harder for the developing world to break through except through political pandering to China, EU or US, which they cannot depend on. There’ll be the occasional country that has a temporarily critical mineral supply or whatever which will try to play off the three in a bidding war, but that never lasts as an advantage. The lasting large magnitude decline in average global productivity growth will hamper our response to climate change.
TP: It will also exacerbate existing problems with limited fiscal space.
AP: Yes, public spending needs to increase over the next 10 years for defence and for dealing with ageing populations too. If we are in a subsidies war between the EU, US and China, and some others try to play as well, then the fiscal crunch becomes even worse. It is far more constructive for our societies to spend public monies on these priorities rather than chasing manufacturing white elephants.
TP: A big debate right now is how to help developing countries meet the demands of climate change — could agendas like the IRA add to that burden?
AP: For all the talk about how ashamed so-called neoliberal economists should be about trade and inequality, the fact is, the people pushing for manufacturing jobs in specific places in the UK or the US are immorally slighting just how important trade, cross-border investment, migration, and technology transfer has been to billions of people in the developing world. This isn’t just about China. This is hundreds of millions of people in India. This is people in Poland, Turkey, Indonesia and Vietnam, and southern Africa and large parts of South America.
This has not been at the expense of average western middle-class people — in the US, domestically-driven tax and public spending cuts did that regressive harm, as austerity did in the UK, not trade with developing economies. Many in developing economies are feeling once again disappointed, if not betrayed, by the shift in western views. When US or UK officials say not just we need growth, but that their government’s priority is making sure formerly imported goods are produced in specific electoral districts, whether it’s the north of England or the west of Pennsylvania, they are adding to that burden.
TP: Do you think the expectations around domestic job creation for reshoring are realistic?
AP: As has become very evident in the creation of the semiconductor fabrication plants in the US, we do not currently have the right workers for a lot of these jobs. Over time we may accumulate them through retraining and shifting workers. If this is a national security vulnerability, though, why not allow legal guest workers or migration or offshore to allies to make it happen more quickly?
Even if we retrain the required workers, medium term we are talking about potentially increasing manufacturing employment in the US by about 1 per cent of the total workforce. That is not trivial, maybe 1.5mn new jobs, but this is not some fundamental transformation of the economy or blue-collar workers’ prospects.
TP: On that point, some point to high levels of investment in postwar West Germany, South Korea and Japan as examples of how governments can, so to speak, create comparative advantage. What are your thoughts on that argument?
AP: Well, South Korea’s industrial policy only really kicked in after the country was well developed, and the same was true in Japan. The actual role of the Ministry of Economy, Trade and Industry in Japanese postwar development was helping workers exit declining industries, which is good, and wasting money, which is bad. Germany offered some subsidised financing to industry, but, like Japan and South Korea, mostly benefited from an undervalued exchange rate for an extended period combined with US defence-driven booms requiring imports.
In all three of these countries as well as Taiwan, what you see is that over-dependence on a few favoured sectors and a small number of protected large companies has not been an unalloyed good. It caused entrenchment and corruption. We should not understate how much concern for the German auto industry, or BASF (the world’s largest chemicals group), led to Germany’s extreme dependence on Russia for cheap energy too. One should not understate how much the chaebol (business conglomerates) in South Korea have been ripping off Korean consumers, and overtly corrupting national politics.
Now, South Korea has a recent different kind of industrial policy success: the booming exports of cultural products, from K-pop to soap operas to film to cuisine, like Cool Britannia under Blair and Brown, but bigger. This did have a government investment aspect, but it was about investing in training, talent and marketing. It did not involve restrictions saying Korean teens couldn’t import J-pop recordings, or Korean Americans couldn’t produce gochujang in the US.
TP: So how do we set the boundaries of when to use subsidy-driven industrial policy?
AP: I think there should be boundaries around both the nature of the subsidies and where to use them. In terms of the nature, they should enhance supply of useful factors of production, meaning human capital, access to financial capital, infrastructure, stable availability of key inputs, and the creation of incentives to allow a market to grow, which includes competition policy. It also should spend on helping workers out of declining industries. It should not be open-ended subsidies tied to the amount a selected company produces of a specific product. It should not be open-ended in duration, either. It should not exclude competition either or move production from abroad — except on a very narrow definition of what are critical national security needs, tied equally to export controls.
TP: Your own work also suggests that the focus for subsidisation should be around new technologies and on adoption over production.
AP: That to me is the biggest economic lesson of technology policy. What matters is how well an economy adopts and encourages change as the result of innovation, not the production of the innovative product itself. This is what we saw with the last round of large-scale subsidies for semiconductors in the 80s and 90s. It didn’t matter much which of Japan, Korea, US or Taiwan-based companies produced the chips over time. What mattered was that when the internet, fibre-optic cable and highly effective dispersed computing came along, enabled by semiconductors, it was the US that adapted its industries and its behaviours really quickly to take advantage.
When this was mistakenly thought about with respect to vaccines, what matters was not that a US or Chinese-produced product was better, but that most of the world’s people did not get the most effective vaccines in a rapid manner. Similarly, on green technology going forward, it should not matter whether it’s an American or a Chinese or a European innovation that leads to the most energy-efficient housing or the best retention of charge in an electric battery or the cleanest way to create hydrogen for fuel. What matters is that as many people in as many places as possible get access to and adopt that technology.
TP: Can a global subsidies race be a good thing?
AP: The EU has been leading the world in responding to climate change. This is because it has emphasised things like its carbon pricing scheme; the shift of solar panel and some wind turbine components production from Europe to China enabled their rapid growth in renewables, which if blocked would have also blocked that progress. So, the world would be better if the US, the UK and China were to emulate European practices and possibly conform over time to them. Sadly, that is unlikely to happen anytime soon.
Therefore, I have sympathy when Biden administration officials argue that they could not let more years go by without the US doing anything on climate. But its current approach will probably undercut some of the effectiveness of the European position. When you have subsidies competing with carbon pricing, the outcome is pretty suboptimal, and the net impact on decarbonisation is unclear at a longer horizon.
TP: Some argue that if the US can produce climate technologies at scale, then the developing world can benefit from them, potentially at a lower cost.
AP: When we think about developing countries and their access to the best green technologies, the IRA is unlikely to be helpful. The US’s priority is claiming credit for jobs in specific electoral districts, and appearing to be tough against foreigners. There is no excuse for the hoarding and slow dissemination during Covid of quality medical equipment and then vaccines to the developing world. There is no reason to think, barring significant changes in policy with that as a priority, that it will be any different with green technology. In fact, having competing subsidised blocs will probably drive up the prices of diffusing green tech, and American elected officials will want to claim big surpluses in comparison to China, EU and others.
TP: So what is the alternative to programmes like the IRA?
AP: If you have to go with subsidies instead of carbon pricing, what you want to do is subsidise three things. First, productive factors like human capital, R&D and infrastructure. That includes creating sufficient public-sector demand through purchases and regulation for a market at scale for technological green innovation.
The second thing you want to subsidise is uptake of green technologies as they become available. So instead of incentivising the producers of green tech, subsidise the consumers, which means both household and other businesses. The less carbon they use, the more money they get back.
Then the third priority is to forge an international agreement that for every dollar, euro, yuan you put in subsidies for domestic production, you also put several cents in a common kitty, pre-committed to spread green technology and needed adaptation to the developing world.
TP: How do economists reassert the case for comparative advantage?
AP: Talk about evidence. But we cannot lie or overpromise. That is part of why there has been an excessive repudiation of evidence-based positions, because there was some overpromising, and now the advocates of alternatives are lying and overpromising.
TP: Regarding the UK, which you know very well from your time on the Monetary Policy Committee, if it did have an industrial strategy, what should it look like?
AP: The UK is a large, important economy, but it cannot compete in a manufacturing subsidies war with China, US and EU, nor should it. The UK should lean into being the best place to benefit from business services, higher education, cultural exports and some forms of R&D. That would change what migration policy should be and increases the importance of alignment with the EU on services regulation. That focus might initially worsen some of the UK’s regional divergence issues, although since business services work and some education can be done remotely, that should be soluble. Making remote work disperse income across the UK would cost a lot less than pointless manufacturing subsidies.
TP: Right. It is OK for countries to specialise in what they do well, and import otherwise.
AP: Of course the UK should not be solely dependent on the Square Mile. But it is logical that the UK should lean into being the English-language, rule-of-law, stable place that is conducive to work in fields that engage with higher education, that do not require large fixed capital investment, and that benefit from the ongoing globalisation that will continue, whatever happens between China and the US.
The above transcript has been edited for brevity and clarity