Commission Watch

What everybody missed in setting up the regional grids.
Fortnightly Magazine - March 2005

Commission Watch

What everybody missed in setting up the regional grids.

While the electric utility industry has largely agreed on what elements to include in a standard market design (SMD) to govern wholesale power trading in a given region, recent experience shows that the regulators from time to time have overlooked a number of things.

These omissions keep cropping up even in the most successful of the regional markets (New York, New England, and PJM in the mid-Atlantic). They are proving difficult to correct or reconcile for the various independent system operators (ISOs) and regional transmission organizations (RTOs) spread out across the country. And the Federal Energy Regulatory Commission (FERC) has not issued much in the way of guidance.

Consider a brief sample of three nagging questions-"holes" in the market, if you will, that few people fully anticipated. These problems undermine the cost efficiencies that we looked for under a market regime, but by no means do they take us to the end of the list:

  • The Gen Plant Rate Case.

    Despite having "deregulated" the generation sector, RTOs now find themselves conducting virtual rate cases for power plants in an effort to set a value for capacity. RTOs can be seen gathering evidence and reviewing all manner of actual and hypothetical costs and inputs for the gen sector, such as labor, taxes, fuel, insurance, maintenance, heat rates, and efficiency profiles. What can be done?

  • Must-Run Gen Costs.
  • Who should pay the steep and unpredictable reliability must-run (RMR) costs that RTOs incur in real-time to dispatch "regulatory must-run" power plants to erase deviations from day-ahead schedules and stabilize the grid?

  • Gen Plant Retirements.
  • With Order No. 2003, FERC has mandated procedures to get new plants interconnected with the grid, but how do we decide when to retire a plant? Who signs off on that? It may seem to be a simple matter, but RTOs can wreck incentives for new investments if they don't take care in managing these retirements.

    These questions remained largely unanswered when ISOs and RTOs began installing their market regimes. In fact, they may not have fully realized that the problems existed in the first place, or their full implications. Yet, these questions now take center stage in a handful of disputes pending at FERC.

    In one such case, the New York ISO (NYISO) has proposed to reconfigure the sloping demand curve for its ICAP market (installed capacity), which helps power-plant owners recover fixed costs by rewarding them for the long-term capacity value they contribute to bolster the adequacy of regional power supplies. But power producers and utilities question the ISO's assumptions regarding benchmark levels of cost and profitability for new gen units ().

    Why should the regional grid operator be the one to decide the ideal hypothetical prototype for a new gas turbine, and then spend more time tallying up what costs would be for that make-believe unit?

    In a second case, ISO New England believes that high and unpredictable RMR costs have discouraged participation in the day-ahead spot market by so-called "virtual" traders-buyers and sellers who do not serve load or own supply resources. And if virtual players flee the spot market, prices could become more volatile.

    Recently the ISO has raised ire by proposing to fix the problem by re-allocating the cost across the entire spectrum of real-time load, rather than to the smaller universe of those who cause the deviations. The case poses a fundamental question: Should regulators treat RMR as an energy service, cleared through the auction market, or as a reliability service, billed to those taking transmission service? ()

    In the third case, investors complain that PJM has reneged on promises to certify the proposed Neptune undersea merchant transmission line. They claim a priority position in the new project queue, and say PJM is dragging its feet unfairly on a system impact study, while it tries to figure out what to do about a sudden rash of later-dated announcements of power plant retirements ().

    The Neptune case poses an intriguing question: If a power-plant owner wants to retire a unit that qualifies for regional planning purposes as a network resource, should the RTO force that owner to jump through the same hoops as required for new projects seeking to obtain the same status? Why didn't FERC anticipate this problem in its landmark Order No. 2003?

    The Gen Plant Rate Case

    Perhaps the greatest irony of electric market design lies in the fact that, despite having "deregulated" the generation sector, the most successful RTOs and ISOs have found themselves mired deep in what can only be described as unsanctioned, virtual rate cases for power plants. Certainly, this outcome was hardly anticipated by the thinkers who dreamed up the SMD. A fine example can be seen in NYISO's now pending attempt to restructure the sloping demand curve for its capacity market, known as ICAP, where both power producers and load-serving utilities were up in arms over the ISO's proposal.

    The case illustrates how the ISO, to determine the graphic parameters for its demand-cost curve, must examine a host of cost factors associated with new gas turbines. Such factors include: (1) plant-heat rate; (2) capacity factor (including such variables as seasonal, temperature-driven variations in plant performance capabilities); (3) offsets for revenues earned; (4) fuel cost volatility; and (5) the shape and profile of customer load, as affected by weather and other variables.

    This task looks a lot like what the state PUCs traditionally do in a full-blown rate case (for a utility that still owns generation). However it differs in two key respects.

    First, the ISO does not begin with historic data already compiled and submitted by the utility to document a given test year, based on a concrete set of assets actually owned by a utility. Rather, the ISO puts on its own case, backed by data collected most likely by a consulting firm of its choice. It then attempts to estimate a hypothetical best case for an ideal, speculative and mythical power plant, to set a benchmark compensation rate for capacity value (assuming there is such a thing), to offer a carrot for merchant plant investors.

    Second, the ISO is new at the game, and does not build on established case law, with generally accepted rules for cost normalization. Consider the New York case, for instance.

    Various plant owners argue that the NYISO has skewed the estimated customer load profile and load shape by failing to normalize degree-day measurements across a long enough historic period. They fault the ISO for relying on nameplate capacity values identified for gas turbines in the manufacturer's marketing brochures, without collecting data on historic plant performance. They say the ISO has boosted estimates of turbine energy revenues beyond historic experience (the higher the revenues, the lower the capacity price) by tacking on a speculative "scarcity" adder. The ISO assumes that at some point during the year (say about 20 hours), energy prices will spike out of control (higher than levels warranted by market fundamentals), and plant owners will earn extra-marginal revenues.

    With wide-scale divestiture of generating plants across the Northeast, we have shifted a huge slice of the traditional utility rate case over to the ISO/RTO. There it rests, alive and vigorous, but without the legacy of rate-making principles and due-process guarantees developed through long-standing practice at the state public utility commissions.

    Must-Run Gen Costs

    According to the New England ISO, charges assigned to market players to cover the cost of dispatching power plants in real-time to respond to deviations from day-ahead schedules averaged only $8.50/MWh in Connecticut and $5.50/MWh in Northeast Massachusetts and Boston (NEMA/Boston) in the summer of 2003. But last fall, in October 2004, real-time RMR charges in NEMA/Boston reached as high as $74/MWh, and exceeded $60 on three days in the same month.

    That means that charges to adjust plant operation to cure second-level contingencies in the power delivery system sometimes ran as high as the cost of the underlying energy.

    Noting these high costs, the ISO proposed in January to redesign its model to allocate real-time RMR costs to the entire universe of real-time load. That would alter the current practice of assigning real-time RMR costs only to those market bidders responsible for the deviations.

    The ISO chose to revamp its allocation in light of anecdotal evidence that at least one virtual bidder had fled the day-ahead energy market to avoid getting hit with RMR costs. Such a hit would prove difficult to avoid, because, as one virtual bidder, DC Energy LLC, has pointed out, "nearly all cleared virtual trades result in real-time deviations." Moreover, the ISO did not want to discourage virtual bidding, as most experts consider that virtual bids are desirable, as they tend to increase market volume and liquidity. That encourages real-time price convergence with day-ahead results, while depressing price volatility and risk. Yet, to many others, the ISO's proposal would socialize RMR costs, and abandon the classic principle that cost-causers should pay.

    But are the bidders really causing the costs? Is there another way out? Who has the real incentive and wherewithal to mitigate RMR costs?

    During the initial debate in New England, some stakeholders had suggested treating RMR operations as a transmission service and assigning the costs to transmission owners, to create an incentive to upgrade the grid. In fact, according to the ISO, the majority of real-time RMR charges result ultimately from a violation of reliability contingencies for transmission-and not for generation.

    Some municipal utilities have opposed the new ISO method, saying it hurts utilities that self-schedule their own resources, because self-scheduling tends to depress day-ahead locational marginal prices (LMPs), creating a higher price differential against real time. Calpine echoes that point, insisting that the new ISO method will encouraging under-scheduling:

    "Yes," says Calpine, "you will get price convergence in day-ahead and real-time markets, but each will be understated."

    Other utilities complain that the proposed allocation method will make it impossible for them to hedge the risk of real-time RMR charges, since, if they serve load, they will take a hit no matter how good of a job they do in covering their day-ahead positions.

    The best solution may well lie in the counterintuitive idea of treating RMR power-plant operation as a transmission service, rather than energy or generation, according to PSEG Energy Resources and Trade LLC, which filed comments in February to protest the ISO proposal. To understand why, consider the subtle difference between the twin concepts of a load-serving entity (LSE), which enters the market to procure supply to serve demand, and "network load," describing the physical phenomenon of a local distribution utility company (LDC) that actually takes energy off the transmission network.

    As PSEG explains, under New England Power Pool (NEPOOL) protocols, suppliers who serve load are not necessarily the same entities as those who take energy off the system. The one acts as a market player. The other operates a physical system. PSEG explains the difference, and how it relates to RMR cost policy:

    "The need to commit RMR units," says PSEG, is not a market product, rather it is a reliability requirement that is incurred based on the physical needs of the system.

    "As such," adds PSEG, "it should be recognized [as] a transmission service [with] costs allocated to network load, like other transmission reliability charges, such as VAR support and blackstart."

    Here's the key concept: While the ISO must dispatch RMR resources to support the physical needs of the system, an allocation to real-time market load throws the cost off on those transacting bids on a financial basis in the wholesale markets. As PSEG notes, it sends a price signal regarding transmission reliability to a class of financial market participants that cannot predict, manage or hedge those costs in any way.

    Plant Retirements

    Next, consider the case of the Neptune merchant transmission line, proposed to run from northern New Jersey (FirstEnergy's Sayreville substation), in the PJM grid, to North Hempstead, N.Y., and designed to ship 660 MW through a direct-current undersea cable to power-starved Long Island. At the downstream end, Long Island Lighting Co., the operating electric utility subsidiary of Long Island Power Authority (LIPA), says it is counting on the Neptune merchant line to provided essential energy supplies to serve its retail electric customers for the summer 2007. Neptune says it can fill the bill, but only if PJM signs off on all required paperwork and issues a final interconnection agreement by March of this year.

    Proposed in December 2000, the line won a completed feasibility study from PJM in 2001, plus a system impact study (SIS) agreement in July 2002, with updates issued several times to reflect project amendments and the withdrawal of a higher-queued project. (That higher-queued project was the 1,100-MW Liberty gen plant, proposed by PG&E National Energy Group, now National Entergy & Gas Transmission, Inc.) On Jan. 21, 2004, PJM sent a revised SIS, billing Neptune for $4.4 million in grid system upgrades required to accommodate the project (which Neptune accepted as a valid deal).

    But since then, utilities and power producers in the region have notified PJM of their intention to retire an inordinate number of power plants. That, in turn, has caused PJM to rethink system impacts.

    But since then, utilities and power producers in the region have notified PJM of their intention to retire an inordinate number of power plants. That, in turn, has caused PJM to rethink system impacts.

    In July 2004, PJM told Neptune that the line could not be built without more grid upgrades, costing another $25 million or so. But then, in the fall, PSE&G announced plans to retire another seven power plants in New Jersey, totaling more than 1,100 MW ().

    This latest bombshell sent PJM back to the drawing board. At a meeting of its Transmission Expansion Advisory committee, held Nov. 19, 2004, PJM stated that in the past six months it had received requests to retire as much as 5,000 MW of generating capacity. That, said the committee, would require more than $130 million in grid upgrades across the region (but not all from Neptune) to deal with the chaos. And so, as of Dec. 21, when Neptune filed its complaint seeking a go-ahead for construction, PJM had yet to execute an interconnection agreement to authorize the project.

    A Game of Chicken?

    According to ex-FERC commissioner and lawyer Michael Naeve, representing Neptune and its investors, the Atlantic Energy Partners, LLC, the incumbent power producers who already enjoy projected status as network capacity are holding the transmission line hostage. Naeve concedes that PJM rules envision a possible second look (a restudy) after PJM has executed a system impact study and calculated any required upgrade costs. But he argues that FERC Order No. 2003 allows such a restudy only if the earth has moved because of one of three events: (1) a modification or (2) a dropping out of a higher-queued (proposed) project; or (3) a redesignation of the point of grid interconnection for the principal project.

    In this case, however, Naeve implies that PSE&G is playing a game of chicken, choosing perhaps to mothball certain underperforming plants while it waits to see if PJM adopts rule changes that would make them more profitable. But that could leave Neptune up in the air for years, waiting for the smoke to clear.

    As it happens, PSE&G had conceded in a report filed at FERC in November, that it had announced plant retirements because of the "inability of the affected units to recover their costs of operation under the current PJM market design." According to the company, the PJM ICAP market failed to award a proper value for power plants located in constrained areas.

    The company continued: "PSEG Power believes," it said, "that had a proper locational capacity mechanism been in place, the need to retire some or all of these units may never have arisen" ().

    In fact, PJM may terminate its current ICAP market for compensating power producers for plant capacity value, and replace it with a new protocol, known as the "reliability pricing model" (RPM).

    (According to a memorandum issued Jan. 20, PJM was to have held a two-day stakeholder conference in mid-February to vet its RPM proposal and take comments, and then post and review any revisions through mid-March. A final RPM rule would be filed at FERC by March 31, 2005 ()

    Consider the full implications of the Neptune dispute.

    While FERC has designed an orderly procedure for new power projects (gen plants and transmission lines) to win the right to interconnect with the existing grid, it has left incumbents free to play both ends against the middle, distorting the base line and creating a moving target for investors. In fact, PJM rules allow incumbent power plants to go into mothballs for as long as three years, and then resume operations on their old terms, continually certified as a regional network resource, as if nothing had happened.

    Meanwhile the Neptune project is racing the clock. To construct the undersea line, Neptune reportedly has reserved the services of the good ship , one of the only vessels in the world presumed to be able to lay high-voltage DC cable across open ocean. Neptune dare not lose its window, says Naeve, or it will not be able to complete laying the cable in time to meet LIPA's power needs.

    At last report, the was seen in the southwest Pacific, laying cable to connect the island of Tasmania with mainland Australia.

    Entergy's Answer

    One vision of where all this will lead can be found in a fourth case now pending at FERC. Entergy has asked federal regulators to confer a dual blessing on its remarkable plan to create a sort-of do-it-yourself, single-company RTO-lite that will make FERC and the state PUCs happy at the same time!

    Entergy thus asks for two up-front declaratory rulings. First, the company asks FERC to promise that its novel construct for an "Independent Coordinator of Transmission" (ICT) will escape public utility status and thus remain outside of FERC review. Second, it asks FERC to confirm that Entergy will satisfy federal guidelines (the so-called "and/or" rule) on transmission pricing with its controversial plan for participant funding for new "economic" transmission. Under this plan, Entergy would force new project owners to assume financial risk and pay all the incremental transmission costs required to accommodate the construction, including needed upgrades to the existing grid. It would do so, that is, if Entergy's ICT first determines that the project is not part of its regional "base plan" to maintain reliability, but designed to make somebody else's rate cheaper. Yet, after the investor completes the project and pays the incremental rate, he attains privileged base-plan status ().

    According to Entergy, all of this would occur within a protected, free-market, FERC-free zone. In the eyes of the law, transmission service would remain bundled with energy, so that the state PUCs get the last laugh and don't lose any jurisdiction.

    To FERC's eyes, that would represent the ultimate in unintended consequences.

    Regulatory Roundup

    Resource Planning & Procurement California OK's long-term resource plans for the state's major investor-owned electric utilities, in the process adopting two controversial rules. First, in evaluating the cost of purchased power contracts extending 3 years or more, utilities should add in a 20% cost factor to reflect the damage done to balance sheets because debt rating analysts view such contracts as the equivalent of debt obligations. Second, in weighing the cost of fossil-fired generating resources, utilities must account for the damage done by global warming by tacking on a cost ranging from $8 to $25 per ton of expected CO2 emissions. Cal. PUC Decision 04- 12-048, Dec. 16, 2004.

    Generation Behind the Meter PJM RTO issues status report that stakeholders are working on plan to permit municipal utilities and cooperatives-not just wholesale customers-to net any behind-the-meter generation against associated load to determine liability for charges for network transmission and ancillary service. Also, munis and co-ops could choose whether to count BTM capacity in full (w/o netting) in satisfaction of their installed capacity (ICAP) obligation. FERC Dkt. ER04-608-2, filed Jan. 3, 2005.

    Wind Power Interconnections FERC proposes rules for wind power facilities to qualify for interconnection to the wholesale power grid, requiring wind facilities to maintain real-time SCADA communications capability with transmission providers, and to demonstrate the ability to continue operations to add stability to the grid even during a low-voltage emergency. FERC Dkt. PL04-15, Jan. 19, 2005.

    Construction Contract Bidding California OK's rules barring utilities from using reverse auctions in soliciting bids for energy construction contracts - a procedure by which bidders compete for contracts by trimming their bids until the lowest bid is accepted. Cal. PUC Decision 04-12-056, Dec. 16, 2004.

    Transmission Outages Citing unconstitutional interference with interstate commerce, a federal district court overturns Kentucky law that had required state's electric utilities to give priority to in-state native-load retail electric customers in deciding which service to curtail first to deal with an interruption or outage affecting intrastate wholesale power transmissions. Ky. Pwr. v. Huelsmann, Civ.A.3:03-47-JMH (E.Dist.Ky.), Jan. 18, 2005.

    Fuel Cost Recovery Connecticut hikes rates for Connecticut Light & Power by 10.3 percent above level of January 2004, saying that need to recover rising costs of fuel can circumvent ongoing retail rate freeze. Conn. D.P.U.C., Dkt. Nos. 03-07-01RE04, 03-07-02RE05, Dec. 22, 2004.

    Mitigation of Power Prices Federal appeals court overturns FERC order that had OK'd plan by NY ISO to mitigate wholesale power prices set in regional day-ahead auction through an automated computer protocol that examined anomalies in bidding behavior and potential price-setting impact of outlier bids. Court says FERC did not justify approval after it had found that NY plan in some cases would mitigate price spikes caused not by market power or collusion, but by legitimate market fundamentals, such as scarcity or weather-driven load. Edison Mission Energy v. FERC, No. 03-1229 (D.C.Cir.), Jan. 14, 2005.

    Return on Equity California OK's new rates of return on common equity for test year 2005 for SoCal Edison (11.4%), and PG&E (11.22%). It asked utilities in their ROE evidence to account for the potential impact on capital structure of long-term purchased power contracts (3 years or more) viewed by credit analysts as the equivalent of debt. Cal. PUC, Decision 04-12-047, Dec. 16, 2004.-B.W.R.


    Articles found on this page are available to subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.