Jennifer Hillman and Inu Manak have a new paper out called "Rethinking International Rules on Subsidies." In a nutshell, they argue that all the new subsidies being thrown around are likely to create difficulties in international economic relations, and some revised rules might help with that. One key aspect of their paper is figuring out how to distinguish "good" and "bad" subsidies, and making clear that certain "good" subsidies are permitted under the rules. For example, subsidies that "genuinely contribute to fighting climate change," such as "subsidies for renewable energy, carbon capture and sequestration, electric vehicles, and clean cooking stoves," would be permitted.
In terms of how to carve out these subsidies from the rules, they say:
Increased transparency should begin with making the notification process as easy and understandable as possible, with clear incentives for fulsome and timely notices. The WTO secretariat has begun that process through its shift to online tools, but more needs to be done, in conjunction with the SCM Committee, to ensure that all countries understand what they are required to notify and into which boxes their subsidies fall. A further incentive to notify immediately those subsidies that a country believes fall into the green box as non-trade-distorting (i.e., funds for basic research and development or disaster relief or the like) would be an agreement that subsidies notified as within the green box cannot be challenged at all or after a certain period of time, or until concerns over such subsidies have been raised and discussed at a meeting of the SCM Committee.81 That safe harbor from challenge could be agreed upon as part of the notification process even without an agreement on overall caps on industrial subsidies.
---------------------
81. This could be structured similarly to the now expired Article 8 provisions of the SCM agreement, which were considered nonactionable.
Footnote 81 refers to Article 8 of the SCM Agreement, which carved out several kinds of subsidies, including certain environmental ones:
8.2 Notwithstanding the provisions of Parts III and V, the following subsidies shall be non-actionable:
...
(c) assistance to promote adaptation of existing facilities to new environmental requirements imposed by law and/or regulations which result in greater constraints and financial burden on firms, provided that the assistance:
(i) is a one-time non-recurring measure; and
(ii) is limited to 20 per cent of the cost of adaptation; and
(iii) does not cover the cost of replacing and operating the assisted investment, which must be fully borne by firms; and
(iv) is directly linked to and proportionate to a firm's planned reduction of nuisances and pollution, and does not cover any manufacturing cost savings which may be achieved; and
(v) is available to all firms which can adopt the new equipment and/or production processes.
If we want to apply the Article 8 approach to carbon reduction subsidies, a key aspect is how to identify the subsidies for which legal challenges are to be limited. Here's a drafting suggestion from me for a basic Article 8-style environmental provision that identifies certain subsidies as non-actionable:
"Notwithstanding the provisions of Parts III and V, the following subsidies shall be non-actionable:
Subsidies that are designed, structured, and expected to operate with a primary purpose and impact of achieving a reduction in carbon emissions, and which do not have elements that are applied so as to afford protection to domestic production."
It's difficult to get the balance right here. I think you need some degree of scrutiny of the purpose and impact of a measure, so that it is not used as the basis for disguised protectionism, but exactly how much scrutiny is needed? I took a bit of language from the case law and a bit from GATT Article III to achieve the balance in a way that I think is reasonable, but I'm open to alternatives. One alternative is to try to get more specific about some of the details of the subsidies, but I thought relying on general principles might work better.
