The transatlantic subsidies race we need
European leaders should avoid a green industrial policy that emulates the US Inflation Reduction Act
BRUSSELS – The US Inflation Reduction Act (IRA) has America’s trading partners in a tizzy. The legislation is not only gargantuan, dedicating some US$369 billion to climate and clean-energy programmes; it also has a “buy American” component, delivering cash benefits only to buyers of North American automakers and subsidies to renewable-energy producers that satisfy domestic-content rules. Many countries, particularly in Europe, are now weighing the possibility of implementing their own green industrial policies. This is the wrong response.
The IRA’s subsidies for American-made products are undoubtedly contentious, particularly among leading US trading partners such as Japan, South Korea, and the European Union. US President Joe Biden is now in damage-control mode, as he attempts both to reassure partners and to find ways to soften the impact on allies by bending the IRA’s buy-American provisions.
European policymakers are unconvinced. They fear that, unless they introduce subsidies of their own, the IRA will effectively guarantee US leadership in green industries. But the logic underpinning this conclusion is dubious, at best.
True, some European firms have threatened to leave for the US if they do not receive similar subsidies at home. But European producers of, say, wind turbines are unlikely to stop producing in Europe just because they expect business to boom in the US.
“Buy American” does not mean 100 per cent American. To secure the 10 per cent domestic-content “bonus” – which the IRA offers on top of generous subsidies that do not depend on meeting the threshold – a company, such as a wind farm, must use only American steel, but only 40 per cent US-manufactured components. In other words, even with the added incentive to use American-made components, the US government expects that firms will rely mostly on imports.
Moreover, according to trade lawyers, “US-made” components could consist largely of imported sub-components, rendering the domestic-content provisions even less stringent in practice. So, while the IRA does give firms an incentive to produce certain components in the US, there will still be plenty of demand for European products, not only in the US, but also, increasingly, in Europe.
In any case, the IRA is hardly an exemplar of good industrial policy. On the contrary, some of its key provisions encourage waste. In fact, Europe has already rejected the use of inefficient and costly fixed-rate renewable-energy subsidies – which account for US$250 billion of the IRA’s total funding. Instead, most European countries use auctions to find out what subsidy rate is needed to spur emissions reductions. Some are also employing “contracts for difference”, under which the government pays investors the difference between the market price and a guaranteed minimum.
Another questionable feature of the IRA is the investment tax credit, which subsidises 30 per cent of the cost of a renewable-power plant (or 40 per cent, if the domestic-content rules are fulfilled). This is not economically efficient, as the subsidy is not linked to the amount of power that is ultimately produced. There is also a kind of moral-hazard risk: with investors covering only 60-70% of the total cost, they might not put as much effort into reducing costs, or just increase their profits. And waste in the construction of renewable plants would also lead to more emissions, especially if more energy-intensive goods like steel or concrete are used.
Finally, the IRA does little to spur green technological progress. Any project that produces zero-emissions electric power qualifies for IRA support. This means that wind farms and photovoltaic installations – for which costs continue to fall – are likely to be key recipients of IRA subsidies. But while these technologies will be vital to reaching near-term emissions-reduction goals – say, for 2030 or 2035 – a net-zero future will require new technological and engineering breakthroughs.
That is what governments should be subsidising. Rather than offering inefficient subsidies to existing technologies, governments should be funding research and development for new ones. Yet, for the EU, this presents a formidable challenge.
As it stands, most R&D spending in Europe occurs at the country – not EU – level, with funds being directed largely toward national universities and research labs. Moreover, EU budget allocations are decided every seven years, and can be changed only by unanimous agreement. As a result, it would be practically impossible for the EU to allocate more funding to R&D until 2027.
But European Commission president Ursula von der Leyen has an alternative proposal: transform the EU’s temporary pandemic borrowing programme, NextGenerationEU, into a permanent resource. Unfortunately, this scheme has little chance of being implemented, so she has also proposed loosening EU rules on state aid, in order to give member states more leeway to field national subsidies – an inferior option.
The poorly conceived, though much-vaunted, European Chips Act follows much the same pattern, giving EU member states more room for national subsidies, without providing much in the way of additional R&D funding. The US CHIPS and Science Act, which allocates US$13 billion to the development of the R&D workforce, is a much better model, both than the European version and the IRA.
The EU should not enter a subsidy race with the US, certainly not based on the IRA model. A “green-tech” race – in which both sides support R&D for new green technologies – is a much better approach. The question is whether Europe has the political will and vision needed to get into competition shape. PROJECT SYNDICATE
Daniel Gros is a member of the board and a distinguished fellow at the Centre for European Policy Studies