I'm still puzzling over the Appellate Body's reasoning on "benefit" under SCM Agreement Article 1.1(b) in the Renewable Energy/Feed-In Tariff case, mentioned in the last post (Luca Rubini is doing the same thing in his comment to the post). Very briefly, the AB seems to say that if the government "creates a market," rather than just intervenes in an existing market, you can't simply assume that a financial contribution confers a benefit. Rather, you have to find an appropriate benchmark that comes from -- and here I'm a little unsure about things -- a similar situation: "an appropriate benefit benchmark for windpower and solar PV electricity generation in Ontario should be one that, within the parameters of the Government of Ontario's definition of the energy supply-mix, reflects what a market benchmark would yield for wind- and solar PV-generated electricity." (Para. 5.227) In other words, find a similarly distorted market, and compare your situation to that situation. If your situation is more distorted than other distorted markets, than there is a benefit. If it's not, there is no benefit.
I'm not sure that's exactly right, but it may be something along those lines.
Part of the reason this case is so difficult to follow is that electricity markets are absurdly complicated. I thought it might help to put forward two examples of similar issues using more familiar products.
First, let's say that governments are worried about the polluting effects of coal, and they hope to develop natural gas as an alternative. They also want to ensure a long-term supply of energy. They recognize that there is lots of shale gas out there, which would help with these goals, but unfortunately it's hard to get to it. It's too expensive and thus there is no market for it. So, a government subsidizes research on techniques to extract it. The research pays off, and now there's a booming market for shale gas. Would it be said that the government created this market? In that situation, how would the AB's test be applied to determine whether a benefit has been conferred? What is the benchmark? If other governments have offered subsidies, do you compare this government's subsidies to other governments' subsidies? If no other government has offered subsidies, what is the benchmark for comparison?
For the second example, let's say a government wants to end reliance of fossil fuels, so it decides to promote battery-powered cars. But there's no market for them; they are too expensive. So the government offers tax credits to consumers who purchase these cars, thus creating a market. How would the AB's test apply here? What is the benchmark? Do you compare this government's tax credits to those of other governments? What if no other government offered tax credits?
I have no answers at this point -- only questions.