Official Report: Minutes of Evidence

The Chairperson (Mr Brett): Professor Rooney needs no introduction. He is well known to the Committee. He is from the Centre of Advanced Sustainable Energy (CASE) at Queen's University Belfast. Thank you very much for giving up your time to come to the Committee once again. I will hand over to you if you are content.

Professor David Rooney (Queen's University Belfast): Thank you very much for the opportunity to speak today. I will give a bit of background to the paper that I presented to DFE, which, hopefully, has been shared with the Committee. I will give a bit of history to it to indicate why we developed a model in the first place.

In effect, we in CASE had been quite keen to support a resolution to the renewable heat incentive (RHI) scheme. As part of that, we were involved and engaged with the Renewable Heat Association for Northern Ireland (RHANI) and others to identify information, reports and correspondence pertaining to the matter.

What became very clear in our analysis was that there were several areas of disagreement with the Department, including tariff rates and other factors that influenced the expectations on returns of investment, including counterfactual fuels, fuel costs, boiler efficiencies and other assumptions in the various different models.

Basically, my conclusion from all that was that, in order to better understand those factors, it would be useful to develop an independent model based on the available data and then to use that to support follow-on discussions.

As DFE representatives have pointed out at previous Committee meetings, the model that was developed can be described as a bottom-up model. In essence, it estimates cash flows, particularly for the 99 kW biomass boilers, as those had the most impact on the overall cost and were the focus of previous reports. The cash flow analysis estimated the income received via the subsidy, as well as the outgoings for fuel costs and operating costs. Some of those estimates were historic, and others were predictions of future cash flows. Alongside an estimate of the initial capital outlay, those were then used to calculate the internal rate of return (IRR) over the 20-year asset lifetime.

On the model development, there are two key values that impact on the subsidy income, namely the tariff rate and the load factor. The load factor is the average percentage time over the course of the year when that boiler is used at its maximum output. It, therefore, directly translates to the total kilowatt-hours of heat generated when you have the number of hours in a year and the specific technology. That combination of the tariff per kilowatt-hour and the total kilowatt-hours of heat production in a year determine the payment that must be made and must be considered together when evaluating the overall cost of the scheme. The Committee will be very aware that the original RHI overspend issues came when there was a disconnect between those two values.

The outgoing cost relates to the fuel type, efficiencies, maintenance, hassle costs etc, and it is in those areas where most of the complexities arise, given the range of values that exist. For example, on the counterfactual fuel costs, including choice, such as biomass versus kerosene, biomass versus liquefied petroleum gas (LPG), natural gas and so on, what are the actual hassle costs for users? Should the ancillary equipment be included? How much is the annual maintenance for a given technology? Are financing costs eligible?

Some of those have been highly variable over the past decade, most notably the fuel costs. That has had a significant impact on decisions relating to the point-in-time tariff setting. For example, if favourable conditions are assumed regarding the biomass pricing, the tariff could be set quite low, only for fuel costs to rise sharply in response to things such the Ukraine-Russia war, pivoting the scheme into a highly unfavourable position.

Some of the key decisions and assumptions used to develop my model are that LPG was used as the counterfactual fuel. It was the differential cost of the heat delivered using either to give a net benefit or deficit to participants. For example, if LPG is more expensive than biomass, there is a net benefit to the participant from using biomass because it is cheaper versus the counterfactual. If LPG is cheaper than biomass, there is a net deficit to users because it is more expensive to use biomass versus LPG.

The capital cost estimates were then estimated from reports, including the core ancillaries. Hassle and maintenance costs were taken as a percentage of the overall capital expenditure. That was 5%. Financing costs were considered in early parts of the model but dismissed, effectively, for the overall tariff calculation. RPI values were used to predict how those costs would change over time historically and into the future, assuming that we are now in a steadier state, hence the 2022-23 financial crisis or cost-of-living crisis jump was dismissed going forward.

To simplify the calculations, only a tier 1 tariff was used alongside the cap of 15%, which is similar to the current state. The reason for that was because I had no specific data on the actual usage figures post-2017. Prior to that, payments to boilers were in the public domain and were able to be used to give an indication of the number of boilers that were operating as well as payments received to date.

Boilers that had historic pre-2017 load factors greater than 15% were then capped at 15%. Moving forward, those below 15% retained their current load factor value. The calculations were performed for all 1,262 boilers on that original list. Those are the 99 kW boilers, and the average rate of return was calculated for all those together. Those are the basics of the model.

Based on all that, the key result is that, when a tariff in my model is set at 7p per kilowatt-hour, the average rate for those 1,262 boilers was 11·9%, which is just below the 12% state aid cut-off. That 7p per kilowatt-hour is obviously matched to the 15% cut-off in terms of load factor.

The total estimated cost for that, based on my model for the 99 kW boilers over the next 10 years, was around £126 million. That leads to key differences with the DFE model, but, in essence, the two models broadly agree.

The key difference is that DFE could clearly draw from a much more detailed dataset from Ofgem in respect of the actual usage figures. That led to the proposed use of the 2017-19 banding, which was discussed with Research and Information Service (RaISE) officials just prior to this. I think that that is fair. Based on the information that I have seen, a number of different load factors and dates were chosen, which have been presented in the DFE model, and that one seems to be the most representative of fair use.

Secondly, it allows for future payments to use a tier 1 and tier 2 tariff rating, so it is slightly different from my model, which capped everything at the tier 1 band. However, overall, it means that the DFE model is probably around 12% more expensive than my original model, coming in at around £141 million for those 99 kW units.

That can be easily explained by the fact that there is a larger estimated heat output than my model predicted and the fact that kerosene was used as a counterfactual in that model versus LPG in mine.

Basically, the DFE tier 1 and tier 2 tariffs of 6·1p per kilowatt-hour and 1·7p per kilowatt-hour, with a 15% cap, are, on average, slightly more beneficial than, but in line with, my model and, overall, are fair. On the whole, it is a fair and reasonable tariff rate, based on the data that I have and the modelling that I have done. Clearly, participants and bodies such as the Ulster Farmers' Union (UFU) and RHANI will have greater knowledge of the historical factors and are, therefore, in a better position to advise on the impacts of that history.

As has been raised, the introduction of banding has been an issue. Banding was not part of the original model, but I have looked at that issue, because I know that it is important. We have had discussions with DFE on the different banding opportunities and what those might look like. Banding has clear advantages and disadvantages. The advantages include that it offers a streamlined process for DFE and participants; lowers operational costs; provides increased operating flexibility for participants; and allows for better financial planning for participants, as well as the Department. Those need to be weighed up against the disadvantage of what would be considered an overpayment relative to metering for the same amount of heat generated. That is quite clear.

It should be kept in mind that, if you are of the view that participants are entitled to a fair return on their investment in line with the 12%, the modelling work shows that 99 kW boilers should continue to receive a payment, almost irrespective of how much fuel they use. That comes about due to the fact that, over the past 10 years of the scheme, there has been a relative insensitivity to fuel price differential costs, assuming that we will not have another energy crisis over the next period. However, if you are of the view that participants should only receive an income on the heat produced and should only be able to achieve the expected return if operating at the maximum, you will tend towards metering instead. In other words, the banding system that has been proposed blends those two views. It aims to try to encourage people to maximise boiler use, in line with their historical value, by protecting their income, even if there is some variability in the output of the scheme, and allows boilers to remain on the scheme for future use as required.

You can, of course, model the potential impact. In the unlikely event that all participants decide to maximise their income and minimise their heat, there will be a significant cost saving. About half the money could be saved if everybody decided to operate in that way. However, if only a small number of participants purposefully operate or are required to operate in that way for any reason, the saving to the taxpayer significantly reduces. Clearly, that has to be weighed against the fact that metering would involve increased costs for operating, maintaining, policing etc, including the fact that meters would need to be upgraded on all sites, as they only have a 10-year window.

Effectively, that translates as there being a cut-off point as to where compliance with the scheme is or is not of value to the taxpayer. I have not been in a position to calculate that. As Karen from RaISe mentioned in the previous session, a number of different costs need to be factored in to understand whether banding versus metering would be useful. What I can say is that, given the conversations that I have had with UFU, RHANI and other participants, the drivers are to maximise heat production. I have seen no evidence that people wish to move to a system whereby they maximise income and minimise heat. RHANI, UFU and others are very clear that they want the scheme to be transparent and have proper controls in order to ensure the integrity of the overall scheme. My understanding is that they are in favour of the banding structure because of the advantages that I have mentioned.

I have provided DFE with the model that I have developed, as well as some additional briefing information. I am happy to talk through any of that.

The Chairperson (Mr Brett): Thank you very much indeed, Professor Rooney. For the public watching, I will ask for clarification: the tariffs that are currently out to consultation are not your tariffs; they are the tariffs that the Department produced.

Professor Rooney: The Department developed those tariffs independently, but the overall cost of the scheme and what it means with regard to the subsidy that would be received by participants are broadly in line with each other.

The Chairperson (Mr Brett): Yes. In fairness to the Department, it probably had access to more information on historical fuel use beyond 2017. You have taken into account the differences in fuel type. Were there any other issues that you think led to a slightly different calculation? Ultimately, you seem relatively content.

Professor Rooney: The bulk of the difference was down to load factor usage values.

The Chairperson (Mr Brett): Yes. I am looking through the Department's model. It does not seem to have taken into consideration the historical use when there was uncapped usage. You seem to have done that in your model, or am I picking it up wrong?

Professor Rooney: The model that I was shown by DFE is a future prediction model of the costs of the overall scheme. It did not utilise IRRs as a metric to determine the actual tariff rate. To get that IRR value in my model, I needed to use historical payments. That is why my model includes those values and future predictions.

The Chairperson (Mr Brett): Have you a view on compliance? The Department has made clear that it will not do inspections and that it will rely on proof of purchase of fuel etc. Have you a view on that?

Professor Rooney: There is a blend. The Department has indicated that it would want to undertake some level of inspections; in fact, RHANI and others have asked for increases in inspections to ensure the integrity of the scheme.

The Chairperson (Mr Brett): Yes. Would you describe this as RHI closure, continuation or, as was mentioned in the previous session, graduated closure?

Professor Rooney: That is partly semantics. This is a different scheme. It is a scheme that is proceeding to an end point. Clearly, no new boilers have entered. In my opinion, it is progressing to closure.

The Chairperson (Mr Brett): Perfect. Thank you.

Mr Honeyford: I want to pick up on a couple of things. You said that one of the advantages is a lower cost for the Department, but there is not a lower cost for the Department. It will cost £1·4 million a year to operate the scheme, which is more than 50% more than what it costs now.

Professor Rooney: If metering etc comes in, there will be an additional resource request in order to be able to measure and analyse all that. As Karen and Aidan from RaISe pointed out in the previous session, there is a question about that breakdown and what it means in either of those two scenarios. I cannot answer as to what those actual costs would be.

Mr Honeyford: Do you believe that the 50% cut-off is necessary? Does it encourage heat to be generated? We need confidence and closure, and the public need transparency. I do not understand how having those bands, as they sit, encourages heat production.

Professor Rooney: They encourage heat production by encouraging people to move into the higher bands. Yes, you could operate at 51% rather than 50%, and you would see an uplift. For example — these are broad calculations — if you were in the top band, you might receive £10,000, and, if you were in the middle band, you might receive £7,500; there would be a differential in received income of about £2,500. The question is whether increasing your fuel usage would put you in the higher band. Part of the discussions with participants has been about the fact that they would like to move into the highest band possible. Therefore, banding would encourage heat production. You can, of course, create additional bands to try to push more and more people into the top band. The alternative, which was also looked at, is that you reduce the payment in the middle band in order to encourage more people to move into the top band.

Mr Honeyford: Why were those ruled out?

Professor Rooney: They have not been ruled out. The current banding structure was considered to be the most fair, as it takes into account some of the issues around historical payments, but the option is there to change those bands and the payment delivered into those bands as required, in order to put more people into the higher bands.

Mr Honeyford: Thank you.

Mr Delargy: Thanks for your presentation. I want to pick up on two points. I am sure that you will be glad to hear that one is on banding. There was a comment in the previous presentation on the use of more bands. In your expert opinion: first, do you think that that would be practical; and, secondly, how do you think that that would work, and what should the parameters be?

Professor Rooney: It will be interesting to see how many self-declarations come back and to compare those against the meter readings or other datasets that come in to support all that. Yes, you could have more bands. The more bands that you put in, the closer that you get to a metering scenario. Part of the original conversation was about having only one band. The single payment has been discussed. This approach recognises that some participants may have machines that are on the scheme but are not being used and, therefore, should not get the full payment. The overall answer is that, yes, you could have additional bands. It would create a slightly additional burden. The more bands that you put in, the closer that you get to metering, so somebody has to draw the line somewhere.

Mr Delargy: That is interesting. It is a point for reflection for us.

My second and final point is around the fluctuations in fuel prices. You mentioned that, in some ways, those could be mitigated. Are there any definitive ways that you can mitigate those fluctuations? Should any review mechanisms be looked at as part of the scheme to ensure that there is an ongoing review throughout?

Professor Rooney: It would be wise to keep an eye on fuel price differentials to make sure that there are no upsets like there were several years ago, which led to some of the difficulties that the Committee is still dealing with. I do not think that there is any way to mitigate those fluctuations. Some of the issues are geopolitical. We hope that we will be in a more stable period for the next number of years, but there are no guarantees about that. Being able to support and grow the biomass market would help to shore up local supplies. Certain things could be done, but some of it is outside our control.

Mr Delargy: Thanks very much.

Ms McLaughlin: Thank you, Professor Rooney, for your presentation. It is terribly complicated, but, from what I gather, your estimate on the rate of return takes account of the historical payments, but the Department did not do that. Do you have a view on the consequences of not taking account of those payments?

Perhaps you will give me your observations on this second point. I am really concerned about how the Department will carry out the relevant inspections programme. I do not get a sense that that has been worked through yet, and, if that is not done, I think that it will have considerable implications. As others have said, we really need transparency and good governance in this programme. The two go hand in hand.

Professor Rooney: Absolutely. On your first point about the internal rate of return, my understanding is that DFE developed two worksheets. One was the tariff-setting worksheet, which included some of that analysis. There were previous models, which I have seen, in which it was quite simple to understand the internal rate of return calculation. That was different from the DFE model that I had a look at, which was simply future prediction. It looked at the cost of the scheme based on tariff rates that were imported from other work. Hopefully, that helps.

Ms McLaughlin: Yes. What about the inspections?

Professor Rooney: Again, I will point out that UFU and RHANI have indicated that they welcome those inspections. They have even suggested to DFE, in various meetings, that spot inspections could be brought in to encourage good behaviours. You will most likely see that come through in the consultation responses. There will be a desire among DFE and participants to make sure that there is integrity, including through inspections. There is no detail yet as to exactly how that will be operated. People are waiting to get feedback from the consultations before they co-design what it will look like.

Ms McLaughlin: I have to ask. You have given the preferred modelling. However, there is real concern out there about the ongoing cost and when that cost is compared with the one-off payment, which, we have learnt, would provide much more value for money for the taxpayer. Can you speak to that one-off payment? I think that it would cost around £67 million compared with £196 million-plus. There would be £17 million for administration costs over the 10 years. The one-off payment seems to provide much more value for money, but you talked at length about the fairness of it. Can you talk about the one-off payment, please?

Professor Rooney: The one-off payment is a simple closure. It is saying, "We're going to pay off your boiler", and everybody just walks away. The reality is that a lot of people entered into the scheme in good faith on the basis that they were making an investment and that they would then be able to make a return on that investment. Simply closing down the scheme, calling it a day and asking them to walk away without realising the full benefit of that investment was always an option — and it was always going to be the cheapest option. The approach of moving towards closure with the full payment, as we have described, encourages the generation of renewable heat — i.e. it is in line with the original intent of the scheme — and allows people who made those investments to continue to receive on that basis. That is where the difference between those two numbers comes from.

Ms McLaughlin: OK. Yes, it is not just about value for money but about the whole context of renewables and outputs, and the commitment that participants in the scheme made previously. Thank you very much.

Ms D Armstrong: Thank you, Professor Rooney, for your presentation. I am sorry that I missed part of it. With regard to public perception and transparency, which other members have mentioned, how convinced are you that the banding, with the wide bands that have been given, will satisfy the public?

Professor Rooney: That is a very interesting point. It is potentially controversial in the sense that, if you take the view that people will try to maximise income and minimise heat, yes, the public would benefit. However, if you take the view that most participants will choose to maximise the heat output, that disappears. Banding tries only to encourage more and more people to move into the higher tier. There is a public-perception interest story here. There is a need to monitor it. There is a need to try to track it as much as possible and work with the sector to ensure that people are actually moving towards the higher usage figures, as they have indicated that they would, rather than trying to maximise income and minimise heat. It is something to keep an eye on. However, you can really only keep an eye on that once it is operating.

Ms D Armstrong: Have you recommendations for the monitoring of future support schemes for renewable energy, based on the modelling that you have done in this role?

Professor Rooney: We have opinions on how we might be able to use additional funding to encourage uptake of renewable heat across Northern Ireland and benefit the overall economy. Bits of work are still ongoing on how we would do that. However, the governance story behind RHI, as well as the other costs that would be associated with it, have clearly influenced what would be required. It is a very expensive, but very interesting, learning journey.

The Chairperson (Mr Brett): Those are all the questions that we have at this stage, colleagues. No doubt, we will try to indulge your kindness and be in touch during our consideration of the Bill, Professor Rooney. Thank you so much for your time and expertise on the issue. It is much appreciated.

Professor Rooney: Thank you, Chair.