Submitted by Bruno Prior on Mon, 14/12/2020 - 12:54

1. From 2007, Summerleaze invested significantly in renewable heat (specifically, the supply of wood pellets for heating), in the belief that even the British government would not be able to ignore for much longer the obvious constraint that they would only be able to decarbonise so far by focusing on the 20% of our energy that is electricity.
2. We selected wood pellets because it was the only technology that appeared suitable to supply the quality of heat required by the insulation and plumbing in the average, draughty British building, at the scale required to make material progress in decarbonising heat. One advantage of the UK lagging so far behind was that it was not difficult to see what had worked in other countries that had made progress in this sector. Academics and interest groups promoted all sorts of magic bullets as usual, but in the real world, biomass heat dominated because it was the least challenging substitute for the existing fossil-fired heating systems.
3. When the Renewable Heat Incentive (RHI) was introduced, this reality asserted itself again, turbocharged by bad decisions in the design of the scheme (not only in Northern Ireland, but also in Great Britain to only a slightly lesser extent).
4. Biomass heat would have dominated anyway without these mistakes, because the suitable applications for the other technologies were more limited than the government’s ivory-tower advisers recognised.
5. But the government believed, as usual, that when the market did not behave in the way that their advisers had predicted, it was the market that must be wrong. The incentives must be adjusted until the “right” outcome was adequately incentivised.
6. Within a limited budget, getting more of what they wanted also meant getting less of what they didn’t want. So support for biomass heat was “degressed” (i.e. cut) rapidly, while support for the “right” technologies was increased to many times the cost of biomass heat. Biomass went from growing at 70% a year (not difficult from a minimal base that had waited two decades for an opportunity) to grinding to a halt three years later, long before achieving the critical mass to sustain an industry.
7. This was predictably not compensated by an offsetting increase in the contributions of the “right” technologies, because the reality was that the costs were higher and the opportunities were scarcer than the government’s “experts” and the industry lobbyists claimed.
8. The RHI was flawed in too many ways to count. We advised DECC of the flaws before and after the RHI’s introduction, but as usual, predictions of unintended consequences and perverse incentives were unwelcome and ignored during implementation, and then greeted with great surprise when they materialised (“20:20 hindsight…” “who could have predicted…”). Amongst the flaws were:
9. The overall budget: heat is twice the size of the electricity sector, and yet the government judged £1bn to be excessively generous to make rapid progress to catch up in this massive component of our energy, whilst happily signing up to £10bn/year to decarbonise 1/3 of our electricity.
10. Given a tight budget, it was important to get the best value possible. But that meant the technology they didn’t particularly want (biomass). So they divvied up the budget as though they would get significant contributions from multiple technologies, despite the fact that some were much more expensive than others, and with widely-varying potential. This effectively ring-fenced a fraction of an inadequate budget for the only technology that could deliver a material contribution for the limited funds, and sterilised other parts of the budget by ring-fencing them for technologies that would not materialise.
11. There are substantial economies of scale in most energy technologies. The government once again resorted to “banding” to reflect this, despite its illogicality and counterproductive history in earlier mechanisms.
    • In the RHI’s case, this meant classifying projects as “small”, “medium” or “large” and paying significantly higher tariffs for “small” than for “large”. But each band encompassed a wide range with big differences in scale economies. For example, “small biomass” covered anything from a 10kW to a 199 kW boiler, even though the latter was an order of magnitude more cost effective than the former. Support was set at the estimated level required for the average. This was exceedingly generous for the largest sizes within the band, which could consequently enjoy a payback period of 4 years on a scheme that ran for 15. Delivery was therefore heavily skewed to 199 kW boilers, not because they were the prevalent size required (heat demand is generally diffuse and smaller than that) nor the most economic if it were not for the mechanism, but because it was the size that gamed the RHI best.
    • The tariff for units over 199 kW was much lower, but the sort of application that needs that much heat is often not difficult to sub-divide. So even where the most efficient solution (barring the RHI) would have been a large boiler, this was often sub-divided into multiple smaller, more expensive boilers serving portions of the heat demand each, because this maximised the RHI. Unfortunately, the flip side was that it minimised the value achieved by the RHI.
    • The small-biomass tariff was eventually degressed to lower than the medium-biomass tariff, because of the over-deployment that the structure had stimulated. Medium biomass projects at the top end of their band (999 kW) became the new sweetspot, and took off accordingly. Ironically, many of the best opportunities (e.g. keeping chickens warm) had already been developed as multiple smaller (199 kW) boilers. So rent-seekers had to create heat loads to suit the necessary scale. They realised they could dry their own wood, and get paid to do it. In fact, the support was worth more than the fuel cost to do the drying, so the more they could dry, the more money they could make. They could dry the fuel to feed the boilers to dry the fuel to feed the boilers… The medium biomass band was quickly swamped with such applications, and equally quickly degressed to a non-viable level for new projects. Most of the RHI biomass-heat budget was used up on projects that would never have been designed that way, and many that would not have existed at all, if it were not for the stupidities of the mechanism design. Only a fraction of the money went to genuinely displacing fossil fuels. The main renewable-heat technology with the potential to deliver cost-effectively at scale was wasted and discredited thanks to the terrible design of the RHI.
    • This effect could easily have been avoided. In consultation, before and after the RHI’s introduction, we set out how to achieve that. It was eventually recognised in a revision to the RHI’s biomethane tariffs. It could have been implemented across the board from the start if policymakers had the humility to consider that their initial policy designs and knowledge might be imperfect, and that experience of earlier mechanisms might count for something.
    • The way to avoid that disaster would have been to pay a high tariff for the first X MWh and then a low tariff for all subsequent MWh. X does not vary by scale. For small projects, X constitutes a high proportion of their output, and the weighted average tariff is therefore high. For large projects, X constitutes a small proportion of their output and the weighted average tariff is therefore low. Mid-sized projects get mid-sized tariffs. There are no thresholds, just a sliding scale, and therefore minimal perverse incentives to target certain sizes. It is not complicated maths, but it seemed to be beyond the comprehension of the RHI’s architects.
    • Or alternatively, just pay a flat rate (e.g. 4p/kWh) on the basis that projects should get what they are worth, not what they need. Challenge installers of small projects and expensive technologies to find ways to reduce costs, or accept that they do not represent good value for taxpayers and the environment if they cannot. Encourage the large projects and cheap technologies that offer the best bang for the buck.
    • Either approach would be more rational than what we got. That is not hindsight; it was predictable and predicted.
12. The best value in renewable heat is from large biomass projects. Because heat is usually fragmented, the opportunities are limited. They require a substantial investment and take a considerable period of time to deliver. They need (a) some value for their carbon because the capital cost is not usually justified otherwise, and (b) policy stability so that investors will commit to the long timescale to deliver the project.
13. Unfortunately, practically the day before the RHI was due to launch, the government announced that they had to revise the large biomass tariff, supposedly because they had realised at the last minute that it would fail an EU state-aid test. When the revised tariff was eventually announced, it was half the original level, and inadequate to justify almost any investment under any conditions, let alone with the heightened political risk from the abrupt policy change. Consquently, almost no large biomass heat was delivered until this policy was reversed. Deployment focused on the smaller, more-expensive scales. If a government had set out to sabotage its own policy, it could scarcely have done better.
14. Most heat is used in buildings, and most of that is in homes. But in another late announcement, the government delayed the introduction of the Domestic RHI from 2011 to 2014.
    • When it launched, the budget for the scheme to cover the largest heat sector (domestic) was a small fraction of the overall budget, which was itself inadequate by comparison to the resources devoted to electricity.
    • Successful technologies quickly hit the budget limit and were “degressed” to a tariff level that would deliver little, whilst their contribution was barely large enough to be discernible in the national energy statistics.
    • Unsuccessful technologies sterilised part of a budget that was already inadequate to achieve anything significant.
    • As fast as the RHI (domestic and non-domestic) encouraged the establishment of businesses to support the nascent demand, it shot them down again when degression pulled the plug on further deployment.
15. It quickly became apparent that the design and budget of the RHI would limit its contribution. But the government still had to achieve by 2020 a level of renewables across the economy, not just in electricity, under the Renewable Energy Directive. And inadequate delivery would expose a flank for opponents to attack the government’s green credentials. What should a government do, if it needs to claim good progress but is unwilling to fund it?
16. Conveniently, government statisticians “discovered” (twice) that they had been under-estimating the amount of domestic wood-burning that was occurring. By changing the model used to estimate this element (reasonably on the first occasion, and then ludicrously on the second), they could not only radically increase the amount of renewable heat that they could claim, but also keep increasing it every year in proportion to the estimated level of stove sales, without spending a penny.
17. There were a few problems with this “model”.
    • It required us to believe that four times as much wood-fuel was available as was estimated by the Forestry Commission and the industry.
    • It relied on a survey that asked us to believe that approximately 40% of the wood-burning occurred in the summer months, contradicting an earlier part of the survey, in which the respondents had been asked to specify the summer months when they did not much use their fires and stoves.
    • It also asked us to believe that almost none of the new stoves replaced old stoves or fires (which would otherwise have had a negative effect on the estimates because renewable heat is measured in terms of the fuel inputs not the heat output and the higher efficiency of new appliances would have implied lower fuel consumption). The source of this assumption was supposedly communication from HETAS and the REA, both of whom confirmed that they had been referring to RHI-type biomass boilers, for which it would be true, not wood stoves, for which it is palpably not true.
    • Moreover, if it were true, we would have expected continued increases in the levels of air pollution traceable to domestic wood burning, but such an increase was not occurring according to the careful (and scarcely favourable to wood burning) measurements being carried out by researchers from Kings College London.
18. Frankly, it is an obvious, convenient fiction to cover the government’s embarrassment at the inadequacies of its renewable-heat policies. It is so excessive that the hypothetical contribution from this source dwarfs the contribution of the RHI across all technologies.
19. But it was not enough. One problem with statistical rather than real delivery is that it does not provide for organic growth. The amount of growth that could be assumed by the most optimistic interpretation was not sufficient to achieve the necessary growth rates in renewable heat.
20. So government statisticians miraculously “discovered” another overlooked source of renewable heat: commercial air-conditioners. They suddenly realised that these were sometimes run in reverse to provide heat, and carried out some studies to conclude that (a) the scale was substantial (again, more than the contribution of the RHI), and (b) the efficiency of these units in heating mode was higher than previously thought and conveniently just over the threshold required to qualify as renewable heat.
21. The units were re-classified as “Renewable Air-to-Air Heat Pumps” (RAAHPs) and a generous contribution was estimated from them. In a stroke of the pen, they added another 10 TWh of renewable heat to the statistics without delivering a single extra kWh in real life.
22. Unfortunately, this statistical adjustment contains no logic to ratchet the figures upwards in subsequent years, so the contribution of heat pumps to renewable heat has actually slid backwards a touch since this adjustment was made. Delivering renewable heat through statistical adjustments is not the best way of encouraging an industry to deliver real projects.