ISO-NE auction reveals increases in demand response and renewables

ISO-NE’s latest forward capacity auction saw prices decrease for the sixth-year in a row, but it also saw some interesting “firsts” in new participants and increased demand response and renewable energy participation. The market is in a state of flux as the generation supply transitions to be less baseload reliant.

As we show in this webinar, the market is looking to evolve to accommodate state mandated goals and address concerns for winter energy security. Watch now to see:

• Why ISO-NE hasn't been able to buck its downward price trend;
• How newer resources are making more noise; and
• How the market will adjust itself for secure supply in the future.

Transcript

Chris: Welcome to ICF's webinar on ISO New England's capacity auction outcomes and outlook. My name is Chris McCracken and I'll be moderating the webinar today. Before we get started, I just want to cover a few ground rules for the webinar. First of all, to prevent any issues with background noise, all attendee lines will be muted. We will be taking Q&A at the end of the webinar, but you can submit questions at any time using the question box at the bottom of your control panel.

We're also recording the webinar today. We'll send an email with a link to the recording in a few days when it's available. Please feel free to share that with other colleagues, or listen to it again if you'd like. And finally, I'll remind you of this at the end again, but please do take a few minutes to take the survey once we're done. We review the feedback very carefully and we use it to make our following webinars a better experience for all of you.

You'll be hearing today from three of ICF's top experts on the New England markets: George Katsigiannakis, Himanshu Pande, and Rohit Vadakattu. If you listened to our webinar on this subject last year, you would've heard them and be familiar with their voices. Their contact info is shown here, and you would have seen it in the lead-up slides as well, but we'll be showing that to you again at the end so you can reach out to them directly with other questions and things. And with that, I'd like to turn it over to Himanshu to go over the agenda and start the webinar.

FCA trends and FCA-13 auction results

Himanshu: Good afternoon, everyone. ISO-NE conducted the 13th Forward Capacity Auction in February 2019, and today we are going to discuss the FCA 13 auction results. The agenda that we have for the presentation is shown in Slide 4. We start the presentation with the discussion on historical FCA trends, and then giving a snapshot of FCA 13 auction results. Thereafter, we discuss the key market changes that have happened since FCA 12, and the primary driver of FCA 13 auction prices. In the third section, we discuss some of the key outcomes and observations from the FCA 13 auction, and in the last section, we are going to discuss the potential market updates that could have impact on future capacity prices and capacity auctions.

Slide 6 gives a snapshot of historical trends for the forward capacity auction. In the graph, we have shown the total amount of cleared capacity and cleared capacity prices over the past 13 auctions. As you will see from the graph, until FCA 7, there was excess supply in the system, and ISO-NE had a vertical demand curve, as a result of which the prices were actually clearing at an administratively-set floor prices. In FCA 8, ISO-NE eliminated the administratively-set floor price, because of which we observed significant amount of retirement in FCA 8, which in turn tightened the supply-demand balance, and resulted in price spike in FCA 8.

In FCA 9, ISO-NE implemented the pay-for-performance initiative, and this was the first auction when New England also implemented a linear demand curve. This was the auction when actually prices peaked in the ISO-NE capacity auction, but after this auction, we have seen a constant decline in the prices. The decline in the capacity prices over the past few auctions is mostly driven by incremental supply in the system because of new resources. And, in addition to that, because of significant increase in a passive demand response, we have also seen incremental supply clearing the auction.

Additionally, there have been significant downward pressure on the demand because New England has been increasing the forecast on distributed solar generation, which in turn has basically reduced the overall reliability requirement of the systems. In short, incremental amount of new capacity additions and declining demand has resulted in significant oversupply in the system over the past few auctions, which has been resulting in, you know, declining trend in capacity prices.

The next slide provides a snapshot of FCA 13 auction reserve. The auction closing price was $3.80 per kilowatt month, and all the capacity zones and interfaces cleared at this price except the New Brunswick interface. New Brunswick interconnection interface was actually export constraint, as a result of which it cleared at a discounted price of $2.68 per kilowatt month. Overall, the auction cleared with a total supply of approximately 1.1 gigawatt in excess of the net installed capacity requirements. As a result, the auction had a reserve margin of around 19.8 percent compared to a target reserve margin of 16 percent.

In general, we have around 4 percent excess supply in the system. This was also the first auction when CASPR was implemented, and there was a substitution auction in addition to the primary auction. However, we did not see any significant participation in the substitution auction. The substitution auction clearing price was zero dollars, and there was only 56 megawatt of capacity that cleared to substitution auction.

The slide on page eight provides a summary of key drivers that impacted the capacity prices in FCA 13. We have categorized the drivers based on their impacts. The factors that put downward pressure on the capacity prices were new capacity additions, continued transition to convex demand curve, higher energy margins realized in 2018, more energy efficiency, and mystic cost of service agreement.

The factors that basically provided support to capacity prices were termination of Clear River Energy Center, and significant exit via static and dynamic de-listing in the capacity auction. Four eighty-five megawatt of Clear River Energy Center Unit 1 has been clearing in the forward capacity auction since FCA 10. However, due to permitting delays and local opposition, the plant is not expected to come online starting June 1, 2019, as a result of which New England actually disqualified this plant from participating in FCA, and removed its megawatts from the FCA supply. The next section discusses some of these key drivers in more detail, along with changes in key market parameters since FCA 12.

Recap of key market changes

Net installed capacity requirements and net CONE are two important parameters that can impact demand curve use for a specific auction. As you will see from the chart on the left side, the net installed capacity requirements have been on decline primarily due to decrease in load forecast and increase the amount of distributed solar over the past few years. However, compared to FCA 12, there was no material change in the net installed capacity requirement in FCA 13.

Similarly, there was no significant change in the net CONE as well. Net CONE had materially declined in FCA 12, and that was mostly because New England had switched the reference technology from combined cycle to a CT unit, as a result of which net CONE declined from $11.64 per kilowatt month to $8 per kilowatt month. However, between FCA 12 and FCA 13, there was no material change, and net CONE just marginally increased from $8.04 per kilowatt month to $8.156 per kilowatt month. So, given that there was no material change in net installed capacity requirement and net CONE, these were not like, you know, one of the major divers that impacted the capacity prices in the FCA 13 auction.

Although there were no changes in the demand curve due to net installed capacity requirements and net CONE, there was further transitioning of demand curve to convex shape as shown in Slide 11. In FCA 11, ISO-NE switched from linear to convex-shaped demand curve. However, they decided to actually do this change by facing it over FCA 11 to FCA 13. Since FCA 11, the demand curve has been shifting towards the left to account for the change in the shift from linear to convex demand curve. In this auction, because of this transitioning, demand curve was shifted towards left as shown in the right chart, and, you know, this would have put downward pressure on the capacity prices in FCA 13.

There were significant amount of retirement de-list bids and static de-list bids in this auction as well. Mystic 7 had submitted a retirement de-list bid in this auction, and the retirement de-list was approved by ISO-NE. Overall, there was 598 megawatts of retirement de-list bids, and major portion of that capacity was Mystic 7. There was 3.1 gigawatts of capacity that had submitted static de-list bids in FCA 13. However, out of the 3.1 gigawatts, 2.5 gigawatts of capacity withdrew that static de-list bids just before the auction, and in the end there was only 670 megawatt of static de-list bids that were left in FCA 13. Out of that, only 139 megawatt was eventually approved by IMM, and most of that 139 megawatt reflect Yarmouth 1 and Yarmouth 2 static de-list bids.

I would like to highlight that even though the 530 megawatt of Westbrook's static de-list bid was rejected by ISO-NE. This unit still did not end up clearing in the auction. The capacity supply obligation for Westbrook in FCA 13 was zero. So, although its static de-list bid was disapproved, the resource still did not clear the auction.

The other change that happened in FCA 13 was there was unmitigated supply from Mystic 8 and Mystic 9. In FCA 12, both Mystic 8 and Mystic 9 had submitted dynamic de-list bids. However, because of reliability reasons, New England rejected the dynamic de-list bid of Mystic 8 in FCA 12, and Mystic 8 was actually forced to clear in the FCA 12 auction.

For FCA 13, Mystic 8 and 9 submitted their retirement de-list bids, but New England rejected those bids as well because of reliability issues, and decided to provide a cost of service agreement to Mystic 8 and 9, as a result of which, Mystic 8 and 9 are receiving out-of-market payments, and New England was permitted by FERC to allow these units to participate as price taker in the auction. Compared to the last auction, since last auction already had Mystic 8 forced into the supply, the only change compared to the FCA 12 and FCA 13 was Mystic 9. In the last auction, Mystic 9 was allowed to dynamically de-list from the auction, but in FCA 13, Mystic 9 was forced into the supply. So, if you compare the net supply change for Mystic 8 and 9 between the two auctions, it was just 710 megawatt capacity of Mystic 9.

Slide 14 provides overall net supply change between FCA 12 and FCA 13. So, as you will see from the graph, the majority of the new supply came from, you know, new generators which were around 837 megawatts. Then there was new passive DR which was around 566 megawatts. There was new un-mitigated supply, which was 710 megawatts, which was mostly Mystic 9. Majority of this excess supply was offset by the incremental retirement de-list bids on reduction in the supply, so there was 598 megawatt of retirement de-list, which mostly reflected Mystic 7. There was decrease in existing passive DR of around -187 megawatts, and then there was a disqualified capacity of Clear River Energy Center, which was around 485 megawatts.

Overall, the net supply increase after accounting for all these changes was 875 megawatts, but then there were incremental static and dynamic de-list that happened in FCA 13 compared to FCA 12, and this incremental amount was around 771 megawatts. So, in the end, if you account for static and dynamic de-lists from FCA 13, the net supply change between FCA 12 and FCA 13 was only 104 megawatts. So, in short, the overall supply change between the two auctions was not significant, and which is why the overall amount of cleared capacity also between the two auctions was very similar.

Higher energy margins was actually one of the primary drivers, which ICF thinks was driving the capacity prices down in FCA 13. So, as you all know that New England has significant amount of oil capacity in its capacity mix, and all these oil resources are actually currently marginal in the capacity market, and they set the capacity prices.

In New England, these oil resources make significant amount of energy margins during extreme winter conditions, and that's due to gas supply constraints, and natural gas price spikes, and natural gas plant outages, which are mostly driven by natural gas pipeline outages. In the table on Slide 15, we are showing a comparison of winter energy prices in 2018 and 2017. As you can see, because January 2018 had a cold snap, and because of severe winter conditions in January 2018, the natural gas price and power prices spiked in New England, which would have resulted in significantly higher margins for oil units. Because these units realized very high margins in 2018, we believe that, you know, overall, their bidding in the February 2019 auction was reflecting those incremental energy margins that they made in 2018, which basically put a downward pressure on the capacity prices.

Slide 16 basically illustrates all the capacity price bridge and all the factors that we just discussed, and what their impact on the capacity prices was. So, as you will see from the graph, the demand curve change had a very marginal downward impact on the capacity prices. The incremental new supply had a material impact, but most of that impact was offset by incremental existing removals and the static and dynamic de-listing removals. In the end, the major impact basically came from higher energy margins, and that was driven by incremental oil margins or incremental energy margins that oil resources realized in 2018 mostly due to severe winter conditions in January as a result of cold snap. We do expect that, you know, if 2019 winter is mild compared to 2018, then the oil resources are going to bid higher in the next capacity auction, which might basically put pressure on the capacity prices in FCA 14.

In the next section, Rohit Vadakattu is going to discuss the key outcomes and the observations from FCA 13.

FCA-13 key outcomes and observations

Rohit: Thank you, Himanshu. Looking at Slide 8, this is the first...Slide 18, sorry, this is the first auction where the CASPR regime was put into play, and we had substitution auction take place. Because it's a new market change, it was difficult to predict the participation and behavior of resources in this. That being said, we did see in the last substitution auction just a mere 53 megawatts of capacity exchanging positions, and that was between the Pawtucket Natural Gas plant and the Vineyard Wind plant, the latter of which was prohibited from participating...clearing to the RTR exemption due to its location in federal waters. The clearing price in substitution auction of zero dollars implies that vineyard was willing to accept the CSO as a price taker at $0 per kilowatt month, and this allowed Pawtucket to realize the full $3.80 per kilowatt month as a severance payment.

Looking forward, FCA 14 is not expected to be drastically impacted due to the small number of megawatts exchange hands, but if more capacity did participate, it might have had more material impact in the later auctions as more capacity retires from the system.

Moving on to Slide 19, we see approximately 990 megawatts of generators de-listed from the system through dynamic and static de-list bids. So, initially, approximately 3.1 gigawatts of capacity were applied for static de-list bids, but approximately 2.5 gigawatts withdrew the de-list bid. That being said, we will keep an eye out, as Himanshu mentioned, on the Yarmouth units, which d supply, give in static de-list bids and dynamic de-list bids, which indicates a degree of exit system in future auctions. And we will keep a closer eye on that. As Himanshu mentioned also, Westbrook did submit a de-list bid that was rejected by the ISO, but it did end up not clearing the auction, anyway.

Going forward to Slide 20. Clearing the auction did come as a surprise to many, including us as well. As you can see in our calculations here, we calculated the estimated net CONE based on ISO-NE and CONE information, as well as the forward data based on trade as of February 2019. Based on that, we were able to calculate the net CONE of the new CC to be approximately $6.10 per kilowatt month, which means that Killingly must either be seeing a favorable gas supply or must have secured another line of revenue which allowed it to clear at the lower clearing price of the FCA 13.

Moving on to Slide 21. This last auction also saw a healthy participation by new renewable capacity. Approximately 199 megawatts cleared the auction, of which 145 megawatts cleared through the RTR, and 54 megawatts of Vineyard Wind capacity cleared through the substitution auction.

In the next auction, when Vineyard Wind is permitted to participate through the RTR, we expect to see the amount of the new renewable participation and clearing in the auction to increase given the fact that there are still 300 megawatts of RTR exemption remaining in the system.

Moving on to Slide 22. We also see an increase in battery participation as generation capacity. Approximately 450 megawatts of battery capacity offered as capacity in the last auction, and of that, only 5 megawatts cleared. However, the fact that approximately 500 megawatts did offer its capacity shows an upward trend in battery participation, and we see this trend continuing in the market.

Moving on to the next slide, Slide 23. Continuing with the series of first [SP], in this auction we saw the most amount of new demand response to clear compared to all the previous auctions, and this reflection approximately 654 megawatts of new DR. However, despite this new record DR, it still doesn't meet the CELT forecast, and this is partly because some DR which cleared in the FCA 12, which amounts to approximately 200 megawatts, did not clear in this auction, resulting in an overall approximately 440 megawatts of more DR clearing in this last auction relative to FCA 12.

Moving onto the next slide, Slide 24. Positive demand response and EE are also...continuing with series of first, we saw...the first auction, we solar-cleared the auction as passive demand response. And even though if it was offered and it was offered in the prior auction, it did not manage to clear. Likewise, this is the first time that solar and storage participated and managed to clear approximately 20 megawatts in the auction. In the active DR phase also, battery continues to participate and clear in the last auction. So, that being said, I will pass...throw onto George who will talk about market updates and looking ahead.

Market updates and looking ahead

George: Good afternoon, everybody. As many of you know, the New England ISO is very concerned about the energy security, especially the availability of energy to meet the energy demands during the winter months. Gas pipeline constraints and system resources where power plants and nuclear power plants have been retiring have created some dynamics that the system, although it does have sufficient megawatts, may not under extreme conditions have the energy requirements to satisfy the demand. We expect the proposed solutions to...proposed measures to address this problem to have major implications to the capacity market.

And, actually we already saw that in the last auction, as Himanshu discussed before, Mystic 8 and 9 were allowed to participate in the auction as price takers. This fact by itself, ignoring all the other factors, has depressed the capacity prices, something in the range of $1.80 per KW month. If, for example, those units were allowed to leave the auction as they have a request requested, the prices would have been higher in that range, I think, including those factors.

Similarly, following this Mystic debate, FERC requested New England ISO to implement permanent solution to this problem, and ISO is in the process of doing so. They're trying to have those implemented from the FCA 16, but for the next upcoming auctions, Auction 14 and Auction 15, they have proposed what they call the Interim Winter Energy Security Proposal, which was actually debated a couple of weeks ago in the NEPOOL Participants Committee. And, although it was rejected, did not receive enough votes to be filed as a joint proposal at FERC, ISO-NE indicated that they planned to file this proposal with FERC at the end of March.

In the next slide, 27, we do provide a high level overview of what this proposal is about. The design is not different from the forward capacity market. You have a forward sale of allowable inventoried energy at the forward settlement today that is calculated from the ISO to reflect the LNG contract. And it was set actually at $82.50 per megawatt hour. So, resources...qualified resources, resources that they can...inventoried energy can sell forward. They are inventoried energy at the forward settlement rate times the allowable inventoried entity.

And the allowable inventoried energy reflects the energy that can be provided by the resource over a maximum duration period, which was set by ISO to reflect three days, or 72 hours, times the maximum output of the research, that's the maximum that the resource can provide, of course, resources that they cannot...hydro, for example, they cannot have so much energy stored. They can provide a lower allowable inventoried energy in the forward auction. So, you sell this forward commitment and you receive a payment for that.

And then, similar to the capacity market, there's the spot sale. Spot sales are occurring during the trigger conditions. Trigger conditions have been defined by ISO to reflect the days where the average daily temperature is at 17 degrees Fahrenheit, or below, and during those conditions, the actual inventoried energy of the committed resources is measured. Any deviations from the committed inventoried energy are penalized or receive bonuses. And so, with this framework, resources that they can provide energy during constrained conditions, they receive a forward payment, and then they can settle deviations from that.

Obviously, this measure is going to provide incremental payments as required to resources that they can provide energy during the energy-limited hours. In the top of that, since generation of electricity during those scarcity hours will have an impact on the inventoried energy and decrease payments, these opportunity costs are reflected in the energy margin. Overall, we do expect that the energy or prices also are going to be increased.

In the next slide, we provide what resources are allowed to participate in this Interim Energy Security measure. As you expect, those are resources that they can store fuel on-site, like nuclear, and coal, and oil. Batteries also can participate, but renewable resources cannot. As it's expected, the resources that they receive incremental payments through this measure, their de-list bids with higher revenues will be lower.

As I said before, energy prices during the trigger hours are going to be higher, so all the resources are going to be receiving higher energy prices. Overall, we do expect the Interim Energy Security Proposal to have a negative impact on the capacity prices for FCA 14 and FCA 15. Of course, that doesn't mean that the revenue of all the resources is going to be lower. Resources that they have fuel storage capability will be receiving overall higher payments, but that would create the dynamics that resources that they don't have those fuel storage capabilities with lower capacity payments will leave the auction sooner.

And that can create the dynamics to interchange with the state-sponsored resources that, as outlined here, New England states have very ambitious plans to bring online a significant amount of renewable resources and battery resources in the system, which, as you may know, through CASPR, to receive capacity obligations, they have to exchange their megawatts from the existing megawatts in the system.

Many of you have seen a lot of rumors about the residential solar resources clearing the capacity auction. What's happening is the solar facilities, that they meet some certain requirements in terms of size and in the connections rating, can participate in the forward capacity auction as passive demand resources...passive demand resource is a different name for energy efficiency...with a trigger price of zero. That's what has happened with those resources as those resources leave the auction as passive demand resources.

A question that we have been asked from our clients is, what is our views on this trend to be continuing the future? New England ISO has...actually the states of New England ISO have plans for continued subsidization of residential solar, and the question is if we do see the residential solar play a major impact in the forward auctions in the future. We do believe that the amount...the subject [SP] provided for those type of resources are significant, and the impact in the capacity market is not going to be so significant that this extra revenue will provide additional motivation for large amounts of residential solar to participate in the auction.

In terms of other parameters that they will affect capacity prices, the next two auctions, the market rules don't allow any increase in the performance payment rate, and that's not going to be a parameter that...we'll continue to increase the capacity prices. But, starting from FCA 15, when the payment rate is almost doubling, we'll see a significant impact on that.

Other smaller impacts by this upcoming capacity prices, the linear portion of the market will be eliminated in FCA 14, and as Himanshu mentioned before, the energy margins of the resources could also impact the capacity price in the future.

Questions and answers

Chris: Great. Thank you very much, George, Himanshu, and Rohit. We've got several questions that have come in over the course of the webinar, and so we'll start getting into those now. If you still have questions, please submit them, as we discussed, in your question box in the toolbar.

To start, first question is, can you describe the difference between static and dynamic de-listing?

Himanshu: Yes. Basically, New England has four types of de-listing. The first one is called retirement de-list, which basically simply means that you want to retire completely from the market, like, all the markets: energy, ancillary, and capacity. Then there is a de-list called permanent de-list which is basically similar to retirement, but it only retires you from the capacity market. You can still participate in the energy and ancillary market.

Now, dynamic de-list and static de-list bids are slightly different. They basically allow resources to de-list themselves from the auction for one capacity period. Dynamic de-list basically is based on a threshold that is established by New England. For example, in FCA 13, they had established a dynamic de-list bid of $4.30 per kilowatt month. If the auction prices basically fall below $4.30 per kilowatt month, then any resource from the supply can dynamically de-list from the auction without giving any justification.

On the other hand, in the static de-list bid, you basically have to specify what is your static de-list bid before the auction. And, static de-list bids are actually higher than dynamic de-list bids, and so you have to basically file with FERC and IMM, and get an approval from IMM, you know, to get your static de-list bid approved. Any unit that is submitting a static de-list bid is basically expecting a capacity price higher than the dynamic de-list bid, which is $4.30 per kilowatt month, and has to basically get an approval from IMM.

George: Thanks, Himanshu. What I want to clarify, also, here is that all the lists, with the exception of the dynamic de-list, are submitted before the auction. The dynamic de-list bid is proposed during the auction. That's a big difference between those two types...you need to think those two different categories, before the auction and during the auction. Dynamic de-list during the auction.

Chris: Next question is, you say the reserve margin is 19 percent based on the auction results. That understates the margin. According to NERC, ISO New England has a firm capacity reserve margin of about 27 percent. So, why isn't the auction clearing to zero as it should?

Himanshu: The 19 percent is a margin that we mentioned is basically based on the cleared supply, you know, that is receiving capacity revenues and it's clearing in the auction. The 27 percent is a margin that you are quoting, basically includes both cleared and uncleared supply. And, this is a phenomena that we have seen also in PJM where, you know, your total with a margin including cleared and uncleared supply is more than 30 percent, but if you look at just the cleared supply with the margin, then it's only 22 percent.

Your question that why isn't the auction clearing at zero, that would be true if the auction had a vertical demand curve. The way auction works is basically it's an intersection of supply curve and the demand curve. If you had a vertical demand curve and if you had excess supply, then the auction prices will either clear at zero, or will clear at a floor, which ISO might have established. And, that's what used to happen in New England before FDA 7. They actually had a vertical demand curve, and they had auction floor price of around $3 to $3.50 per kilowatt month.

And because there was so much excess supply, you know, the capacity prices were clearing at the floor. But now that is not the case. It is now like a convex-shaped demand curve. In the end, it's the intersection of supply and demand, which basically determines how much supply is going to end up clearing the auction. Anything that basically bids higher than that intersection point ends up not clearing the market.

Chris: Okay, great. Thank you, Himanshu. Next question, please talk more about the trends for both DR and battery storage, and specifically, do you know how far out of the money the other bids for batteries were since only 5 megawatts cleared?

George: Yeah. We don't know what that offer is. So we...

Himanshu: Yeah. The bid information is confidential, so we don't know how much, like, the offer bids for batteries were out-of-the-money, but we do expect, if batteries are not subsidized, as of today, they definitely are very expensive to economically come online or to economically clear the market.

Chris: Great. Next one, if ISO New England thinks that some gas-fired generation is not firm, why are they clearing it through the auction, through the FCA? If the pay-for-performance reforms don't succeed in making gas-fired generation fuels secure as it was designed to, why don't they just fix the provisions, for instance, by increasing the penalty levels? But you'd love to.

George: Yeah. That's a good question. So, I guess this question was asked to New England ISO. And, I was expecting the penalty rates to fix the system. Don't forget that the resources have not get used to this...the PI was implemented last May, and the sorted [SP] condition that we observed, it was just one certain condition during the Labor weekend. And you should leave this market design plenty of time and the resources to understand what they have to deal with to expect a response, and I don't think that that was available.

Himanshu: Yeah. And, I would like to add that I don't think, basically...I think penalty rates in New England are going up. They are going to become, like, $5,500 per megawatt hour in 2024 auction. I think the main problem is the system doesn't have scarcity right now, and that's mostly driven by the fact that there's so much excess supply in the system since FCA 9 that, you know, there was a time when New England actually went into shortage.

And now, since then, New England now has, like, 19 percent to 22 percent as a margin, and that's just reflecting cleared supply. If you actually add the uncleared supply, you probably add 25 percent to 30 percent of the margin level. When you have so much excess supply in the system, it doesn't basically result in any scarcity hour. So nobody takes those scarcity premiums or those risk penalties in a serious way.

If you see like...and this is the same situation that we are seeing in PJM where you have, like, total with a margin of more than 30 percent, if you include the uncleared supply. And, you know, if basically we have a recurrence of polar vortex where the system...like, both PJM and New England saw substantial amount of scarcity hours, I think that would be the year when you will actually see substantial amount of penalties, and people will start taking some of these performance penalties more seriously.

George: Don't forget that the system is not constrained in terms of megawatt liability. The system is constrained in terms of energy. That's a big difference, that the PI pay-for-performance, it's a capacity measure. So, fixing the limitation on the energy needs to come in the different parts of the market design on the energy side than the auxiliary reserve status as New England ISO plans to do when the implement permanent measures.

Chris: The next question is, is the Interim Winter Energy Security Proposal available to dual-fuel gas/oil units?

George: Yes. Actually, dual-fuel units, depending on their oil storage capabilities are allowed to participate.

Chris: Okay.

Himanshu: And I think in one of the previous questions we didn't answer, like, the trends on the DR, because we answered the battery storage specific question but we didn't talk about the trends in DR. Overall, I think, in terms of trends in DR, in New England, basically, DR comprises of either passive demand response and active demand response...and passive demand response historically has been mostly energy efficiency, which is driven by, you know, state mandates and state budgets that New England has.

We do expect there will be continued growth in passive demand response going forward, but we also believe that New England has very aggressive targets for energy efficiency, and that I think mostly are driven by the linear cost curve that they assume. And we think that once you utilize the low-hanging fruit, it is going to become incrementally more expensive to implement new energy efficiency measure. We believe that although there will be increased in passive demand response going forward...or energy efficiency going forward, but it will be less than what ISO-NE is forecasting in their self-report.

Then, we do expect that there will be some incremental participation on passive demand response from other types of resources, like what we have started observing, like, you know, solar participating as a...distributed solar participating as passive demand response. There could be some incremental increase in passive demand response from, you know, distributed solar participation. And then, on the active DR side, you know, in the past, we have been observing active DR to decline over time, but we did observe in this auction that active DR basically marginally increased. And that was, again, mostly because New England has recently changed the definition of "active DR."

And then, there are other type of resources, you know, that basically participated as active DR in this auction. You know, there was little bit amount of battery storage that cleared as active DR, and there have been other type of resources as well. So, we do expect that active DR might, like, you know, marginally increase in the long run because of incremented participation from resources such as batteries, solar, which sometimes instead of participating as generator, participate as DR because of their small size. But we don't expect this to be like a material driver forward.

Chris: All right. Thanks, Himanshu. I think this is a follow-up to one of the questions you were answering earlier about the pay-for-performance provisions and their impact, another question. If there's no scarcity of firm capacity, are the concerns about fuel secure supply baseless?

George: No. You should consider some...you have only gas-fired generators in New England, and you face severe...and you have plenty of them, 200 percent reserve margin, and so you face severe constraints on the gas supply. So, having megawatts without fuel, that will not do you any good in terms of liability.

That's exactly what I mentioned before about the fuel security and the capacity market, two different...they're related, of course, of concepts [SP] but I do agree with the treatment of the ISO-NE of separating them and putting the fuel security under an auxiliary service on that energy side instead of the capacity market. PJM is moving a similar direction.

Himanshu: It's basically, you know, the step taken to mitigate something that could happen. I give you an example, like in 2018, when the cold snap happened, we didn't observe, like, significant scarcity in New England, but that doesn't mean the system was completely reliable. If you actually look at 2018 generation stack in New England, 50 percent of the dispatch or generation was coming from oil resources.

Now, imagine a market where all these oil resources retire just because, you know, the capacity prices are not high enough and these oil resources are not economic. And we have seen in FCA 13, there was significant de-listing of oil resources. So, if those oil resources were not there on the hindsight in 2018, then definitely system would have been on scarcity. In order to avoid to reach to that situation, New England basically wants to add some form of fuel security.

Chris: Next question is, why was the demand curve vertical previously, in the earlier auctions?

George: That was the standard design across the United States. So, New York City initially implemented this concept of demand curve to recognize that things are not binary, either you have the...but there's some incremental value. After you met the target reserve margins which differ by region, or for a specific level, let's say 16 percent New England, at 16.1 percent the liability contribution of the resources is not zero. It declines, but it's not zero. This marginal liability, impact MRI, which is actually concept for the convex scale New England implements is not utilized in all capacity markets with exception of MISO, but is thinking of using the same thing. PJM has it, New York has it, so we believe it's a superior design.

Chris: Great. I think we'll close on this one. Does ICF provide auction clearing price estimates for future bids? And I think the answer is yes, and the answer would be, "Please reach out to George, Himanshu, or Rohit, the contact information shown on the screen, and they can talk to you about ICF services in regards to that." And, obviously watch for white papers and things in advance of next year's auction as well.

With that, I'd like to close, and remind you again before we go to please fill out the survey. We really do appreciate the responses, and use what you say to help guide us in the development of future webinars and the implementation of future webinars. And, with that, thank you very much for joining us, and you will be receiving an email shortly with the recording. Thank you.