By trying to placate regulated states—letting utilities “opt out” from its capacity market—PJM finds its RPM idea under fire.
Bruce W. Radford is editor-in-chief for Public Utilities Fortnightly.
While the PJM Interconnection has made no major changes to its prototype capacity market since it proposed the idea a year ago in August, and though it has won a tacit OK from federal regulators for many of the plan’s key elements, don’t expect to see a slam dunk when the time comes for a final review of the controversial idea, known as the Reliability Pricing Model, or RPM.
Instead, consider the fate of RPM’s closest cousin, the divisive LICAP plan once proposed for New England that has now crashed and burned. As with PJM’s RPM market, New England’s location-specific ICAP plan had begun with the idea of a sloping demand curve, as pioneered in New York, envisioned as a purely administrative construct designed as a market surrogate to mimic a theoretical market-based demand by consumers for electric generating capacity, assuming such demand exists.
But the New England model proved far too complex and expensive to gain the trust of local regulators or politicians, and the same could prove true for PJM.
Of course, it is true that the Federal Energy Regulatory Commission (FERC) has endorsed PJM’s very controversial plan of accepting and clearing bids by merchant-transmission developers in the same auction with supply bids from power producers, on the theory that grid expansion adds to capacity resources, as does the construction of a new power plant. And yes, FERC has sanctioned PJM’s idea of conducting forward auctions to procure electric capacity resources for future years, with payments withheld unless developers deliver on schedule. FERC also has agreed to set the market-clearing quantity and price at the intersection of a supply curve derived from actual bids, with an artificial, downward-sloping demand curve set by administrative fiat. Moreover, FERC has OK’d RPM’s locational attributes, plus RPM’s design allowing utilities (also known as load-serving entities, or LSEs) to “opt-out” and skip the RPM auctions, choosing instead to satisfy capacity obligations with self-supplied assets or simple bilateral contracts. (See, Initial Order on Reliability Pricing Model, Docket Nos. EL05-148, ER05-1410, Apr. 20, 2006, 115 FERC ¶61,079.)
Nevertheless, many still question whether PJM can adapt RPM to accommodate those states and utilities within its footprint that remain fully regulated, simply by making the market optional, and whether all these separate plan elements can work together in harmony, without producing distortion or discrimination.
In particular, it remains unclear whether an electric capacity market can succeed in its aim—to encourage development of infrastructure and thus assure compliance with reliability standards (which now carry the force of law, by the way)—if industry players can flit in and out of the bidding, as if on a whim, seeking opportunities for arbitrage. Instead, many think that PJM’s opt-out rule only will increase discrimination between pro-choice and non-choice states, when coupled with a market process built around an administrative approximation of true market demand.
Consider this comment from Exelon, ordinarily seen as a pro-market company. Speaking in Washington, D.C., at a technical conference held at FERC June 7-8, Exelon Executive Vice President (and former FERC Chair) Elizabeth Moler went right for the jugular, playing the “Congress card” to make sure that she got the commission’s attention:
“We need to avoid gaming, to avoid the double-counting … to avoid the discriminatory regime issues which I don’t think this commission wants to chance on the Hill, frankly, and to provide a comparable regime between those who are opting out and those who are not.”
Eventually, New England’s LICAP had spurred such contention that FERC had no alternative but to invite the combatants to settle the matter among themselves, and in so doing to consider any reasonable model that might work to attract long-term region-wide investment in generation, including a wide range of models not fully considered in the evidence and testimony collected earlier in the case record. (See, “LICAP: A Mad Dash to the Finish,” Public Utilities Fortnightly, November 2005, p. 22.)
As a result, New England ended up with a FERC-approved Forward Capacity Market. The FCM features “descending clock” auctions to procure supply- or demand-side resources for future delivery three or more years out, when suppliers become eligible to receive payment. The FCM also allows LSEs to self-supply their own resources through owned assets or bilateral contracts.
The auction bidding in New England will open at a high price level chosen as representative of twice the cost of developing new generating resources (the cost of new entry, or CONE), with lower bids accepted until capacity needs are met, setting the clearing price. Initial bidding patterns and clearing prices dictate the CONE value and thus the range of bids for future auctions. (See, Order Accepting Proposed Settlement, Docket No. ER03-563, June 16, 2006, 115 FERC ¶61,340.)
Now, in similar fashion, FERC has declared that it cannot yet determine whether PJM’s RPM model as proposed is just and reasonable, and instead has elected to schedule settlement conferences to see whether the PJM region can agree on a compromise. Moler suggests, however, that settlement talks have hit a serious snag over the opt-out question:
“It’s complicated. It’s something we shouldn’t do on the fly. … It is festering in the ongoing settlement conferences and it’s worthy of an important body of work.”
As of mid-July, following months of proposals, comments, protests, and technical conferences (one in February; two early June), plus an initial FERC ruling and a PJM brief on the issues, the list of contested issues and the scope of the disagreement remains unchanged from late 2005, when this column last covered the topic.
Arguments still rage over: (1) capital cost estimates; (2) energy revenue offsets; and (3) boundaries for LDA market zones (Locational Delivery Areas). In particular, doubt persists on how to integrate merchant transmission RPM bids with PJM’s administrative backstop procedure known as the RTEP (Regional Transmission Expansion Plan). That’s because RPM would reward grid bids that clear the market with a locked-in, competitive, value-of-service revenue stream, while RTEP mandates grid upgrades if merchant projects go wanting, but only pays them cost-plus rates, funded through a region-wide cost allocation, according to a recent FERC order. (See, Docket No. ER06-456, May 26, 2006, 115 FERC ¶61,261, and FERC Docket No. ER06-880, filed June 19, 2006, with PJM proposing to amend the cost-allocation formula.)
More recently, PJM announced $1.3 billion in grid upgrades mandated through its first 15-year RTEP plan, approved June 23, 2006. (For a more complete rundown of issues and problems, please refer to last year’s “Commission Watch” column on the RPM model, “PJM’s New Game,” Public Utilities Fortnightly, December 2005.)
Lately, however, with Exelon’s Moler leading the charge, the focus has switched to the opt-out clause, which establishes a “Fixed Resource Requirement” for utilities choosing not to bid in the RPM market.
Can PJM’s downward sloping RPM demand curve (known as the VRR, or Variable Resource Requirement) co-exist with the FRR, especially if opt-out utilities can jump in or out of the capacity market, electing to buy or sell at opportune times so as to arbitrage the RPM price against the going price for capacity in bilateral contracts? Moler, for one, believes the answer is “no.” (See Figure 2 for the evidence.)
Elsewhere, however, the opinion goes the other way, such as at American Electric Power (AEP), which has indicated it would likely favor the opt-out path.
AEP likens concerns of gaming and arbitrage by opt-out utilities to “childhood fears of monsters under the bed.”
In fact, at the technical conference held at FERC on June 8, AEP’s Craig Baker (senior vice president, regulatory services) told why regulated utilities won’t be able to cherry-pick the RPM market, selling surplus capacity into RPM when prices go high, or choosing to buy from it when prices fall:
“Regulated states,” as Baker explained, “do not allow you to build surplus generation to sell into the market. … If you do that, you are entering into the merchant world.
“I can give you a case in point where we had built generation, expected load to be there, and then didn’t get allowed in rate base.
“We are not going to build a 5,000-MW generator to meet a 2,000-MW load. It doesn’t happen.”