The UK’s Electricity Market Reform White Paper claims to have quantified how offering low-C generators a hedge creates value:
In our central scenario, the FiT CfD reduces the cost of decarbonisation to 2030 by £2.5 billion compared to using the Premium Feed-in Tariff (PFiT) to deliver the same investment. ( § 2.3, p. 37)
Given that my previous post says there can be no net value at all, I find this claim startling and it should be informative to investigate exactly how this number is arrived at.
Bottom line: (i) they calculate the benefit of a hedge captured by the low-C generator, but, (ii) they completely ignore the cost to the taxpayer or ratepayer from providing that hedge.
Ignoring (ii), the cost to taxpayer or ratepayer, is obviously the big problem, and we could stop there. But let’s take the bait and go ahead to examine (i), how they calculate the benefit of hedge captured by the low-C generator.
The White Paper purports to have assessed how providing the hedge (in the form of a Contracts for Differences, or CfD) lowers the cost of capital for different types of projects. Here is Figure 7 from the White Paper with the results:
Lowering the cost of capital by this amount, and then recalculating the LCOE, generates a lower LCOE.
What is the statistical or analytic foundation for the lowered cost of capital?
One source is a study by Redpoint Energy, commissioned by the UK’s Department for Energy and Climate Change (DECC) and published in December 2010. The study simulates the earnings on a project and calculates the earnings-at-risk, i.e., a measure of the downside risk on earnings. A hedge decreases the earnings-at-risk:
Redpoint then assumes a specific linear relationship between earnings-at-risk and the amount of leverage the project can support:
…if the impact of a policy option to is to reduce the earnings at risk by 10 percentage points we assume that it is possible to increase gearing in the project likewise by 10 percentage points. (p. 131)
To the untrained financial economist, this might sounds like a possibly sensible line of reasoning. And, the basic idea that a cash flow of lower risk has a lower cost of capital is certainly sound. But these steps for calculating the lower cost of capital are entirely unorthodox as far as I can see. The Redpoint report doesn’t cite any research to justify this assumed relationship between the earnings-at-risk and debt capacity, and I can’t think of any myself. The relationship doesn’t make any sense based on standard corporate finance models: the relationship between earnings volatility and debt capacity and the cost of capital is a different one than is implied by this linear relationship. And the numbers selected have absolutely no foundation whatsoever. They just give an appearance of fact where none is warranted. The simulations Redpoint reports are inaccessible, so the results cannot be independently evaluated or tested.
A second source is a corroborating study referenced in the EMR’s White Paper, conducted by Cambridge Economic Policy Associates (CEPA). CEPA simply polled a self-selected set of potential investors (members of the Low Carbon Finance Group) and asked them how much they thought the hedge would decrease the cost of capital. The CEPA report has the appearance of doing analytic calculations, but all of this effort is entirely devoted to translating the results of the poll which are quoted in one metric and re-presenting them quoted in the metric used in the White Paper. It”s just a poll. There is no empirical data to corroborate the results of this poll. In fact, early in the report CEPA acknowledges that it isn’t using one of the familiar risk models because of the lack of available data. I have no idea why they consider this poll an acceptable substitute.
Here’s a final troublesome pair of questions. Why do these studies try to use completely novel and untested methods to estimate the value of the hedge? If the question is the value of hedge, why not use the traditional financial tools for valuing and pricing a hedge?
Remember, what we’re trying to value here is a plain vanilla derivative, or something close to it. We have familiar models for doing that. But these models, or something even vaguely similar, are not employed here. No bank or hedge fund, if it were thinking of offering the same derivative or considering investing in the same derivative, would accept a valuation done using the Redpoint or CEPA methodologies. The UK taxpayer/ratepayer should not accept it either.