Gordon Hughes: Response To Robert Gross et al.
Robert Gross and a group of colleagues at Imperial College have submitted Supplementary Evidence to the House of Commons Select Committee on Energy and Climate Change that comments on the evidence which I prepared for the Committee on behalf of the Global Warming Policy Foundation. Since it took them about 3 months to prepare their comments, it is not possible to prepare a full rebuttal at short notice; I will do this in due course. Nonetheless, their arguments illustrate important features of the UK government’s approach to policy analysis in the energy sector which should be highlighted.
A key problem is a recurring confusion between predictions (what will be), plans (what should be) and policy scenarios (what would have to be). The Imperial College group assert that I have overstated the amount of wind capacity in 2020 and, thus, that my estimates of the costs of current policies are too high. In doing so, they are not comparing like with like. They rely upon a comparison between a set of predictions (which change regularly) and my policy scenarios, which spell out what would be required to meet the UK government’s targets for renewable energy. The UK has a very poor record in energy forecasting over the last 30-40 years, so any evaluation of policy should be based upon something more solid than the most recent set of predictions.
Perhaps official predictions are supposed to reflect a firm plan for 2020. In that case the policy analyst has to ask whether the critical assumptions built into the plan are well founded and how the plan is to be achieved. An endemic problem for policy formulation in the UK as well as planning around the world is the tendency to over-optimism, to assess what one wants to happen rather than what realism and evidence suggests will happen. The outsider has to ask: what steps have been taken to minimise the risks and consequences of over-optimism?
The Imperial College group rely heavily upon claims based upon their models of wind output, dispatch, etc. Such models can be instructive but they do not provide concrete evidence, especially as their models explicitly exclude consideration of the stochastic nature of wind power. My approach is different. I have based my policy scenarios on data relating to the operation of the electricity system over the last 4 years (and rather longer for wind farms). I am unwilling to assume that key variables such as load factors, the pattern of wind availability and its correlations with demand, etc are going to be substantially different from what has been the case in the past.
Similarly, the Imperial College group appear to miss the point of my argument that the reductions in carbon emissions may be much smaller than conventional claims suggest. My examples are deliberately simplified to illustrate the point that the net outcome will depend upon adjustments in the whole electricity system, which – in turn – will be affected by future investment decisions. This has nothing to do with the efficient use of existing capacity, but future investment will depend upon how that capacity is used. It seems that both the Imperial College group and DECC assume there will be large investments in CCGTs in the middle or later parts of the current decade to replace plant which will be restricted or retired as a consequence of the EU’s recent Industrial Emissions Directive. The difficulty is that the load factors for the new plants may be very low if the predictions of wind penetration are realised. In equivalent situations elsewhere the response has been to invest in single cycle plants or to run older plants in single cycle mode. This is not a nonsense scenario but a reflection what actually happens.
The story of wind power in Europe is one of consistent over-optimism about performance and costs reinforced by an apparent unwillingness to define clear policy options and then construct analyses based on concrete evidence. Modelling is not a substitute for evidence. In practice, the Imperial College/DECC claims about the costs of current policies rest upon the old story of “this time everything is going to be different”. Really? How could we test this?
I would propose a simple market test that bears directly on the subject of the original ECC hearing. If the proponents of the Imperial College/DECC view of this issue believe that my calculations of the costs of existing policies are wrong, why do they not endorse a major and continuing reduction in the level of subsidies? For example, the costs claimed would be consistent with a reduction to 0.5 ROCs per MWh immediately and to zero by 2020 for onshore wind and something equivalent for offshore wind.
The current and proposed systems of subsidies for renewable energy are equivalent to a major programme of taxation and public expenditure with a probable cost that will run into the billions or even tens of billions of pounds per year over the next decade. It is incumbent on those who wish to retain these subsidies to demonstrate that taxpayers are receiving good value for their money. I have profound doubts that current and/or proposed policies meet that criterion. Clearly, the Imperial College group believe otherwise, but they have produced no evidence to support their view. No one doubts that we can live with an electricity system including large amounts of wind power. The question is whether the public is willing to bear the full cost when this is fully transparent. At the moment, the costs are neither properly assessed nor are they transparent.
Gordon Hughes, 22 October 2012