Coal's Raw Deal

After years of voicing emotional complaints of little help in forging concrete improvements, ("Electricity is different." "You can't change physics!"), opponents of electric utility restructuring at last have begun to build the kind of logical, coherent and step-by-step arguments that sway regulators and resonate with laymen and politicians.

Recall that the Federal Energy Regulatory Commission (FERC) intends to revisit its decade-old Order 888, and the pro forma Open Access Transmission Tariff (OATT)—and you begin to see a real counter-revolution.

For the missing link—the driving force, if you will—look no further than right under your nose. Look at coal, the fuel that still forms the hidden backbone of the electric utility industry, but which has remained sadly out of fashion for most of the past decade. Experts now say we need more coal-fired generation, despite the environmental concerns. However, they also believe that the basic regime now in place at the regional transmission organizations (RTOs), with its day-ahead energy markets, financial transmission rights (FTRs), locational marginal pricing (LMP), and a bid-based and security constrained dispatch, exerts a marked bias against coal-fired plants.

The full story can get quite complicated, but it all starts with the planning horizon—the forward time period on which investors are willing to bet their dollars.

Since the industry’s very beginnings, utilities have tended to plan for the “long run”—the time frame in which “we’re all dead,” as the British economist John Maynard Keynes once said, famously. Utilities have done quite well with that. By contrast, the RTOs focus on the short term.

It’s not that FERC’s RTO markets don’t encourage investment in infrastructure. To the contrary: they do. But it’s the wrong kind—the simple gas turbine, the quick fix, the easy buck.

Under RTO practice, system operators use software to calculate grid conditions in real time, yielding instantaneous marginal prices (the notorious LMPs), plus a set of linked financial derivatives (their partners in crime, the congestion-hedging FTRs). These market signals impose their will on every project, favoring the short-term fix. Yet the base-load coal plant tends to do its best work when bundled up in a 30-year contract.

Simply put, RTO practice creates less risk and uncertainty over the nominal short-term wholesale price of power, but more risk and uncertainty over the long-term cost of transmission. That spells trouble for the coal-fired plant, sited far off at the mine mouth, needing long-haul transmission over a long-enough term to pay back the capital costs.

The Problem Illustrated

Consider this example, outlined in a letter sent to Midwest ISO (MISO) President and CEO Jim Torgerson by a coalition of municipal utilities serving retail load in Illinois, Indiana, Wisconsin, Michigan, Missouri, and Minnesota, dated May 31, 2005:

A 600-MW coal unit costs roughly $1 billion to construct. If we instead constructed 600 MW of gas-fired peaking capacity, it would cost us only about $300 million.

At today’s coal prices, the variable production cost from a base-load coal plant would be less than $20 per MWh. In comparison, variable production cost

from a gas-fired plant would be about $80 per MWh.

Now pause for a moment. If the investor lived in the pre-RTO world, and paid the coal plant’s high capital cost, he’d be assured delivery of the coal plant’s $20 power, and would risk only the chance of a transmission outage, or perhaps a curtailment under TLR rules (transmission loading relief). Under this older regime, governed by Order 888 and FERC’s pro forma OATT, the coal plant, after qualifying as a network resource, could sign a contract guaranteeing physical transmission rights at a price certain over a long term, say 30 years. The transmission provider would undertake the obligation to maintain the grid network to afford source-to-sink delivery capacity to support the plant and its contract. And if redispatch or grid upgrades might ever be required to maintain delivery capacity, the plant owner would pay no more than its load-ratio share of such costs. This delivery risk was modest—quite a bit less than the energy price risk of committing capital to coal as a boiler fuel.

But today, in RTO markets, the coal plant developer deals with minute-by-minute congestion markups and LMPs that often reflect natural gas on margin. Now return to the letter:

We today can spend $1 billion on coal-fired capacity and still face the risk that our delivered energy price will reflect the $80 production cost of gas-fired capacity. We do not need to spend an extra $700 million to obtain $80 energy, but this is the risk we are required to take today under MISO rules if we invest to serve our customers in a cost-effective manner over the long term.

(For the full text of the letter, see Comments of Midwest TDUs, FERC Docket No. AD05-7, filed June 27, 2005.)

This problem occurs if a load-serving retail utility invests in a base-load coal unit but lacks FTR hedging coverage, or must pay full value to buy it. Voicing near-identical concerns are groups such as ABATE (Association of Businesses Advocating Tariff Equity), the Coalition of Midwest Transmission Customers (CMTC), and the Transmission Access Policy Group (TAPS). To quote attorney Robert Weishaar, Jr., representing ABATE and the Coalition, the investment or contracting decisions "are skewed against the coal unit."

Under typical RTO practice, neither the power producer nor the load-serving utility reserves or acquires physical transmission rights for a transaction, either short- or long-term. Instead, the payment of a regional grid access charge (a zonal "license-plate rate in the typical case) will suffice to guarantee delivery. If congestion arises, the parties simply "buy through" it. They pay the locational marginal price, and can hedge that risk by acquiring FTRs, which pay back the cost to the holder. But the cost of the hedge can eat up coal's advantage. And the FTRs will cover only short terms of one month, six months, or 12 months, at the longest. If you want a longer FTR, to hedge the 30-year useful life of your coal plant, you are out of luck.

This bias against coal has not gone unnoticed on Wall Street. In a report prepared in September of last year, and widely quoted, Moody's described what it called "a potential risk" in the short-term marginal pricing model commonly used by RTOs:

"Without long-term contracts for transmission rights and price certainty for the transmission of energy from new generation facilities, cost recovery in the long term may not be assured.

“The certainty of cost recovery,” added Moody’s, “represents a major factor in the credit assessment of financing for new generation projects.” (“Credit Issues Resurface as New Electric Generation Projects by Public Power Utilities Take Center Stage,” Moody’s Investors Service, Sept. 2004.)

Standard and Poor's had voiced the same concern a few months earlier:

"Pricing data associated with hourly nodal prices should provide market signals for use in planning for investment in transmission and new generation. Yet generators may realize that the benefits will be ephemeral." (Makeover for California’s Power Markets, S&P, July 2004.)

FERC's Answer

No surprise, then, that FERC says it has received many inquiries from electric market participants in regions governed by RTOs, asking for longer-term transmission rights, to approximate the delivery price certainty that plant developers used to get with a multi-year physical transmission contract. So this past spring, FERC invited the utility industry to offer ideas on how to create long-term transmission rights (LTTRs) in markets with locational pricing. And FERC raised the stakes by releasing a staff paper at the same time suggesting that RTO practice, with its LMPs and short-term FTRs, might represent a step backward from Order 888 as originally conceived:

"OATT transmission service, once obtained, appears to provide better long-term price certainty than the current RTO transmission service." (FERC Docket No. AD05-7, staff paper at p. 7, filed May 11, 2005.)

The staff’s conclusion raises a potential conflict with FERC policy, which states that RTO tariffs must offer services and rights “equal or superior” to the pro forma OATT.

Other factors are at play, however. The evidence appears to show that locational pricing improves the short-term efficiency of generation dispatch and grid performance, and thus lowers overall energy prices, according to the latest "state of the market" reports issued by the grid operators in the Northeast. Also, LTTRs can invite "phantom congestion," which comes when players reserve physical grid capacity but then don't use it.

As a power project developer, Tenaska suggests that FERC should "let the market decide." However, Tenaska also observes that market-friendly schemes for participant funding (awarding service preferences to developers who expand the grid) represent simply another system of long-term delivery rights. Thus, according to Tenaska, the key issue at FERC is not whether LTTRs can function in LMP markets. Rather, says Tenaska, FERC must decide whether to restrict the offering of LTTRs to market participants that fund grid upgrades, or to extend LTTRs to include players willing to buy them at prices in excess of the marginal cost incurred by the RTO to provide such rights.

The California Department of Water Resources asks why FERC has waited so long to act. The DWR charges that FERC had stressed the need for long-term grid rights way back in 1997, in one of its first decisions governing the California Independent System Operator (Cal-ISO). (See 80 FERC ¶61,128.) The DWR adds that the problem really is worse than it appears, since RMR generation (regulatory must-run) represents still another form of unhedgeable transmission risk. The DWR claims that Cal-ISO “now frequently supplants” its market-based congestion management scheme in favor of reliance on must-offer generators that the ISO selects unilaterally.

Cal-ISO commented on June 27 that in its work to develop a new market design—the MRTU project—it is currently "modeling" CRRs (congestion revenue rights, an FTR-equivalent) of only two lengths: annual and monthly. It adds that it is "unlikely" that longer-term rights will be available in February 2007, when it hopes to have its new market up and running.

Here's an idea: Keep a certain percentage of transmission capacity separate from the RTO market, and make it available for long-term physical grid rights.

Yet experts say that such "carve-outs" erode the efficiency gains from LMP and a bid-based, security-constrained dispatch. FERC apparently agrees, as it has spent much time and effort attempting to limit the extent to which utilities can retain their grandfathered transmission contracts once the RTO moves to a so-called "Day 2" market. (See, e.g., 108 FERC ¶61,236, Sept. 16, 2004.)

Responding to FERC's request, the Sacramento Municipal Utility District (SMUD) claims that the problem of carve-outs eroding efficiency "is easily avoidable" if physical rights holders are given tradable rights. That, says SMUD, would encourage resale of unused contract rights in a secondary market, avoiding phantom congestion. SMUD cites the example of the WesTTrans (www.westtrans.net), a voluntary group of electric transmission operators in the Western Interconnection. SMUD claims that WesTTrans operates an OASIS site that, contrary to the norm, allows potential transmission service customers to query information from each provider at one time, rather than individually, which facilitates resales of physical grid rights.

SMUD's proposal typifies the feeling out West, where grid operators rely largely on a physical-rights model. SMUD adds, for instance, that the fledgling GridWest group plans on using physical transmission rights.

In any case, even MISO, borrowing heavily from PJM's design to build its own market, has conceded that absent any long-term rights to transmission, it seems likely that new or incremental investments in capacity "are likely to be suboptimal."

Working Within the System

Can the RTOs retain their purely financial grid rights, but simply offer FTRs with longer terms, to help out the coal plant developers?

Five years ago, the New York ISO had offered two- and five-year TCCs allocating as much as 15 percent of system-transfer capability to the five-year rights. Demand was lackluster, however. According to DC Energy LLC, only seven participants bought 2-year TCCs at the Spring 2000 auction. That number rose to 20 in the Fall 2000 auction, with 5-year TCCs awarded to 9 participants. Prices lagged accordingly. Two-year TCCs for transactions from Zone G (NYC) to Zone J (Hudson Valley) cleared in the Fall Auction at an average of $1.87/MWh. The equivalent TCCs for five-year terms cleared at only $0.66/MWh. Some say the ISO reserved too much capacity for the longer-term rights. Others say the market in New York was simply too new to inspire confidence.

In New England the RTO says only one participant has inquired explicitly about FTRs longer than one year. That participant serves load in Vermont, the only state in New England with vertically integrated utilities that retain a franchise obligation to serve load. (Five of the six New England states, says the RTO, feature retail access with service obligations of 12 months or less.)

The prime objection to longer-term FTRs is the added risk of default.

To issue a set of FTRs for auction, system operators must estimate grid network topology and predict conditions expected throughout the term of the right. They must conform the FTR issue to grid capacity (the "simultaneous feasibility" test).

According to the PJM RTO, events such as transformer outages or large-scale grid outages due to new line construction can affect long-run grid topology. Other relevant events might include power-plant additions or retirements at key locations, unexpected load growth, or changes in enrollments among retail-choice customers. In particular, the New England RTO bemoans the lack of forward schedules for planned transmission outages.

A retreat from license-plate rates also might affect FTRs. That raises concern for PJM, as FERC recently asked for comment on whether PJM's license-plate rate design might discriminate against new RTO members, such as American Electric Power, which has recently joined PJM. That's because developers who build new "reliability" transmission can socialize the costs across the entire region under PJM's RTEP system (Regional Transmission Expansion Plan). But AEP, which brought new high-voltage lines to PJM when it joined, must recover the line costs only from ratepayers within its license-plate load zone. (See 111 FERC ¶61,308, May 31, 2005.)

Another item is congestion revenue, which funds payouts to FTR holders.

In New England, the RTO notes a negative trend since October 2003, with congestion revenues consistently falling short of FTR payoffs for the most recent FTR auctions. PJM, meanwhile, notes no serious problems in shortfalls of congestion revenues, but offers data showing that FTR payouts historically have failed to match the target market value of FTRs (see Figure 2, “FTR Payouts in PJM.”)

These reports raise the issue of who bears the risk when revenues fall short.

Who Covers the Risk?

On page 15 of its white paper, the FERC staff asks whether LTTRs should be "fully funded," whereby any shortfall in congestion revenue would be recovered through uplift payments collected, perhaps from transmission-owing utilities (TOs). Ideas on this issue run all over the map.

Morgan Stanley wants fully funded FTRs, with the TOs passing costs through to ratepayers. This model describes the practice in New York, where the ISO issues fully funded rights. Full funding does not apply in PJM, however, where the FTR holder shoulders the risk. PJM thus argues that a buyer or holder of FTRs ought to be prepared to pay for the cost of a revenue shortfall as the price of gaining a hedge.

Perhaps the last word on this problem should belong to one Neal A. Fitch, a regulatory specialist at Reliant Energy Inc.

Mr. Fitch suggests that RTO skill in managing real-time grid operations may not translate to the speculative world of financial derivatives:

"While it has historically been reasonable," he writes, "for the RTO to [act] as the market operator and/or clearinghouse in short-term markets … it is less clear whether or not the RTO should pay a significant role in long-term financial markets that ... deal exclusively with risk management.

"It may be the case that [others] can provide ... long-term FTRs without the added complexity."