|
|
|
|
|
After Interviewing Generators, Brattle Claims Currently Proposed Texas Changes Would Not Assure Resource Adequacy
June 1, 2012
Email This Story
Copyright 2010-
The Brattle Group's report on ERCOT Investment Incentives and Resource Adequacy, prepared at the request of ERCOT, claims that none of the PUCT's existing proposals addressing the energy-only market, "would likely support a target reserve margin consistent with the 1-in-10 criterion, unless much more price-setting DR [demand response] were to participate in the market."
"Scarcity prices would be too infrequent to support the target because if reserve margins are high enough to make load shedding very rare, scarcity pricing events would also be quite rare," Brattle claimed
Brattle's claims are immediately suspect because much of its report was based on interviews with a, "broad spectrum of generation developers and lenders," who have an incentive to understate their willingness to invest in the current market design in order to "reform" the market into one much more tilted towards generation rather than load, and which would assure them greater returns on any investment they would have proceeded with under the current design.
Indeed, Brattle even encountered this in its report. Brattle noted that, "Many generators in ERCOT stated in our interviews that the energy-only market is excessively volatile and uncertain, and that they would prefer that ERCOT adopt a structure with more forward price certainty, such as a forward capacity market."
However, Brattle noted that, "[i]nterestingly, in our recent review of PJM’s forward capacity market, we heard many similar concerns about capacity price volatility and uncertainty."
Brattle further noted that, "[a] number of PJM suppliers proposed to extend the forward period of the capacity market, or extend the capacity market into long-term products reminiscent of long-term PPAs or a regulated planning construct."
In other words, capacity owners are never satisfied, and will never be satisfied until they can have both the "certainty" of a regulated market where they are guaranteed a set rate of return, and the infra-marginal revenue received from uniform clearing price markets (whereas normally, in return for guaranteeing cost recovery, ratepayers receive cost of service power).
Accordingly, investors' words must be discounted, and greater weight should be placed on their actions. Over 10 years, the ERCOT energy-only market has continued to attract investment, and despite cyclical predictions of reliability violations, has never recorded a resource adequacy shortage as new investment has always occurred.
While Matters is not saying that no changes need to be made in the market, there should be caution against claims that the current PUCT proposals would fail to assure resource adequacy, given the experience under the current market design, which is admittedly less efficient at providing investment signals.
Brattle said that if the offer cap is increased to $9,000, a reserve margin of approximately 10% could be achieved without reducing the frequency of scarcity prices below the level needed to support investment.
On average, the 10% reserve margin achieved with a $9,000 offer cap would result in approximately one load-shed event per year with an expected duration of two-and-a-half hours, and thirteen such events in a year with a heat wave as severe as the one in 2011. In years with less extreme weather than 2011, however, load shedding would be expected to occur less than once in ten years.
However, Brattle concedes that if several thousand megawatts (MW) of price-responsive demand were added, those resources could prevent involuntary load shedding and set prices at customers’ willingness to pay, thereby increasing reliability and softening (but not eliminating) price spikes.
With this much demand response, ERCOT’s energy-only market design could support the current bulk power reliability target under a $9,000 price cap, Brattle said, though it immediately dismissed this notion by stating such a high demand response penetration would take years, not months.
Brattle offers no definitive recommendation as to an end-state resource adequacy approach, but says its analysis suggests that either the market design needs to be adjusted or the reliability objectives need to be revised.
Brattle outlined potential end-state market designs as including:
1. Energy-Only with Market-Based Reserve Margin
2. Energy-Only with Adders to Support a Target Reserve Margin
3. Energy-Only with Backstop Procurement at Minimum Acceptable Reliability
4. Mandatory Resource Adequacy Requirement for LSEs
5. Resource Adequacy Requirement with Centralized Forward Capacity Market Brattle did not recommend any specific model, but listed the policy tradeoffs of each.
As described by Brattle, here is a summary of each design
1. Energy-Only with Market-Based Reserve Margin: 2. Energy-Only with Adders to Support a Target Reserve Margin: 3. Energy-Only with Backstop Procurement at Minimum Acceptable Reliability: 4. Mandatory Resource Adequacy Requirement for LSEs: 5. Resource Adequacy Requirement with Centralized Forward Capacity Market: Brattle conceded that the "Energy-only with market-based reserve margins” design is theoretically the most efficient option because it allows customers to choose the level of supply based on prices and their value of avoiding curtailment, without having to pay for costly reserves they may not want. "It also provides strong incentives for resources to be available when they are needed most," Brattle said.
"We believe that energy-only, perhaps with rare backstop procurement of short-term resources as needed to support a very minimal reserve margin, might be the most aligned with the Commission’s demonstrated philosophy to let the market work. However, this would require managing public expectations about reliability implications and the potential for periodic high spot prices," Brattle said.
Of course, Brattle saw the need to defend centralized capacity "markets" devoting a section of its report to discredit what Brattle termed "myths" of the centralized capacity "markets."
Specifically, Brattle said that it was a myth that capacity markets cost more than energy-only markets, stating that, "It is not correct that capacity payments increase all-in customer costs."
However, one need only compare energy commodity rates (e.g. non-distribution) in ERCOT with those in retail choice states with capacity markets. When an all-in (generation and distribution) retail rate in ERCOT can be procured for less than the generation and capacity rate (excluding distribution) in the Northeast retail choice states, the savings from the energy-only design are clear.
Brattle says that capacity "markets" do not raise retail rates because, "Capacity payments only replace the 'missing money' that results from high mandated reserve margins depressing energy market prices (by lowering market heat rates and avoiding scarcity prices)."
However, Brattle ignores that by subjecting energy and capacity to separate uniform clearing prices, the total cost to retail customers is increased, because the mix of plants with the lowest fixed costs (as determined by the capacity market) may not be the mix of plants with the lowest marginal costs (as reflected in the energy market), as newer, more efficient plants that produce energy more economically will have higher upfront going forward costs than depreciated but inefficient plants. Essentially, new plants which may lower the overall costs to customers (despite their high fixed costs) are "crowded out" because they cannot clear the regulated capacity market, regardless of their beneficial impact on energy prices.
Brattle said that another myth was that "capacity markets overpay existing generation." However, Brattle never even attempts to refute this myth, other than weakly offering that the same thing occurs in the energy market, so it's OK ("These concerns overlook the fact that energy-only markets similarly pay old and new resources the same price to reward their equal contribution to providing power when resources become scarce").
However, this ignores the fact that while energy is truly a fungible product, capacity, at least in terms of solving the missing money problem which capacity markets are intended to address, is not, regardless of what FERC has found. In other words, while all capacity may meet the reliability target in the same manner, if load were not compelled to pay all capacity the same price, load would not. Instead, load would only pay those at-risk units a supplemental capacity payment, while any units that are in-the-money in the energy market, and therefore would be available without any additional capacity payment, would not receive a capacity payment from load.
Turning back to Brattle's report, in addition to, and regardless of the overarching policy path selected by the PUCT, "we recommend enhancing several design elements to make the ERCOT market more reliable and efficient ... [including]: (1) increase the offer cap from the current $3,000 to $9,000, or a similarly high level consistent with the average value of lost load (VOLL) in ERCOT, but impose this price cap only in extreme scarcity events when load must be shed; (2) for pricing during shortage conditions when load shedding is not yet necessary, institute an administrative scarcity pricing function that starts at a much lower level, such as $500/MWh when first deploying responsive reserves, and then increase gradually, reaching $9,000 or VOLL only when actually shedding load; (3) increase the Peaker Net Margin threshold to approximately $300/kW-year or a similar multiple of the cost of new entry (CONE), and increase the low system offer cap to a lever greater than the strike price of most price-responsive demand in Texas; (4) enable demand response to play a larger role in efficient price formation during shortage conditions by introducing a more gradually-increasing scarcity pricing function (as stated above) so loads can respond to a more stable continuum of high prices, by enabling load reductions to participate directly in the real-time market, and by preventing price reversal caused by reliability deployments; (5) adjust scarcity pricing mechanisms to ensure they provide locational scarcity pricing signals when appropriate; (6) avoid mechanisms that trigger scarcity prices during non-scarcity conditions; (7) address pricing inefficiencies related to unit commitment but without over-correcting; (8) clarify offer mitigation rules; (9) revisit provisions to ensure that retail electric providers (REPs) can cover their positions as reserve margins tighten and price caps increase; and (10) continue to demonstrate regulatory commitment and stability."
Email This Story Copyright 2010-
• There is no regulatory imposition of a planning reserve margin requirement, nor are there out-of-market interventions to support target reserves or adjust energy prices. Energy prices are usually set by marginal generation offers. When all generation resources are fully utilized, the price rises until price-responsive demand curtails itself voluntarily and the market clears at load’s marginal willingness to pay for power. The price can rise to very high levels or reach an administratively-determined cap at VOLL if involuntary curtailments are required.
• An energy-only market in which reserve margin outcomes are expected to be lower than acceptable, and market rule changes are used to increase prices to support additional investments. Such actions include: 1) further increasing the high system offer cap, the low system offer cap, or the PNM threshold; (2) expanding the responsive reserve requirement, which would in effect structurally withhold more generation capacity and increase prices; (3) relaxing market power mitigation rules; (4) considering an LMP adder, as some stakeholders have suggested; or (5) introducing various types of capacity or availability payments as a separate, explicit revenue stream as has been done in Spain and a number of Latin American countries
• Energy-only market featuring backstop procurement provision that is triggered when anticipated reserve margins fall below a minimum threshold. Capacity levels would be allowed to vary from year-to-year above and below the target reserve margin, but would not be allowed to drop below the minimum acceptable reserve margin. Such a “minimum acceptable” reserve margin would have to be far enough below the target to allow for market-based outcomes to prevail in most years
• Imposition of resource adequacy requirements on LSEs, including locational minimums for LSEs in load pockets. The resource adequacy requirement itself would be determined administratively based on reliability studies. LSEs would be required to buy or self-supply enough capacity to meet their peak load plus the mandated reserve margin or else face a penalty.
• ERCOT would hold an auction in which it procures forward capacity obligations on behalf of all load 3 to 4 years prior to delivery. During the delivery year, the cost of that procurement would be allocated to LSEs. LSEs would be able to hedge against capacity auction costs through self-supply or bilateral forward contracting. Incremental auctions would also be needed to facilitate economic adjustment to new information and manage supply- and demand-side risks between the time of the initial auction and delivery.
|
|
|
|
|