Home

Exelon Proposal Threatens Customer Choice in Texas

February 10, 2012

Email This Story
Copyright 2010-12 Energy Choice Matters

It is becoming increasingly clear that the biggest threat to retail customer choice is not regulators, legislators, incumbent utilities, or consumer advocates, but rather the owners of formerly incumbent generating assets, who now seek to foreclose the retail market after successfully stopping the threat of new entry in most of the RTOs in the U.S.

This objective is achieved, of course, through the compulsory transfer of wealth from competitive retail suppliers to the owners of generating capacity as selected through an auction (charitably called a competitive market) -- an auction whose design inherently favors the owners of largely or fully depreciated assets originally owned by incumbent utilities, and in most cases remain owned by utility affiliates. The process is otherwise known as the capacity market, which forces competitive retail suppliers to subsidize the generation of their competitors -- generation which the competitor can then use to squeeze out competing retail suppliers.

And Exelon now wants to bring this market design to ERCOT, where, if its merger with Constellation is approved and closes, Exelon would own over 4,000 MW of capacity.

"We believe it would be useful for the Commission to explore regulatory measures used in other markets that have successfully maintained required reserve margins," Exelon said in comments to the PUCT in Project 37897, in specifically endorsing the PJM capacity market design.

"Given the success of PJM's resource adequacy construct, Exelon supports the mechanism of a forward market for capacity to provide future price signals," Exelon said.

Aside from future price signals, a capacity market in ERCOT would allow asset owners to force competitive retail electric providers to subsidize the asset owners' generation, with the potential for the elimination of customer "choice" in retail electric providers, as asset owners would used subsidized capacity payments to create an oligopoly.

Consider how an asset owner who also owns a retail book could operate its retail supplier under an ERCOT capacity market.

Currently, if an asset-owning retail supplier wishes to use its generation to serve its customer book -- particularly if the assets are otherwise uneconomic except during scarcity conditions (e.g. peaking units the supplier wishes to use for load following during super peaks rather than supplemental purchases at the system-wide offer cap), the supplier must recover the fixed costs of maintaining such generation from a source other than the energy market due to the plants' infrequent dispatch, which means the supplier must either raise its retail pricing accordingly, or accept lower margins from these customers.

Indeed, Exelon notes the challenges faced by marginal generation in its comments. "It costs millions of dollars a year just to keep a plant running. An energy-only market with average prices centering on variable costs results in shortfalls in cost recovery," Exelon states. In other words, the types of peaking assets which are most useful for retail suppliers in serving retail load at the super peaks are often uneconomic, and cannot be justified solely from energy market revenues and their load following capability.

Such valuations are borne out by market participant behavior. Despite their benefits in serving unhedged load at super-peaks, one market participant with significant retail load recently sought to mothball generation including units at S.R. Bertron and Greens Bayou, because the costs of keeping such units ready to serve load outweighed any benefits. Consider what would happen, however, if the asset owner did not have to recover fixed costs of these uneconomic units from its own retail customers, or the energy market.

Under a capacity market, if these assets cleared the capacity auction (as all existing assets likely would do in a capacity-short region), competing retail electric providers would now be paying for the fixed costs of these uneconomic power plants. Now, the asset-owning retail supplier would no longer have to recover the costs of maintaining the availability of such generation from its own retail customers, or in the form of lower margins. These plants can now be used by the asset-owning retail supplier to serve its super peak retail load at a lower price than the supplier would incur through the balancing market (since the variable costs of all but the marginal unit will be lower than the balancing market clearing price, and since the supplier will also avoid collateral costs).

With its fixed costs of maintaining the generation subsidized by all retail electric providers in the market through a capacity charge, the asset-owning retail supplier could lower its retail pricing, giving it an unfair competitive advantage in customer acquisition. Or, rather than lowering retail pricing, the asset-owning retail supplier could use the higher retail margins resulting from its subsidized capacity revenues to purchase competing retail suppliers.

In either case, it is not hard to see how a capacity market necessarily distorts the retail market, and will inevitably lead to an oligopoly. The only retail suppliers which will survive would be those owning generation and receiving capacity payments from their competitors. While the load may still be competitively served, it hardly represents the choice currently afforded in the ERCOT retail market, or any truly functioning market.

Exelon attempts to assuage opposition to a capacity market by claiming that such a market will not increase costs, but its arguments lack merit.

First, Exelon offers that, "under the RPM construct, energy and ancillary services revenues are deducted from capacity prices to ensure no over-recovery of generator costs, thus enhancing the efficiency of the market."

Even under the original design of the Energy & Ancillary Service Offset, load argued that the mechanism insufficiently deducted energy and ancillary services revenues from capacity prices. However, while the mechanism's prior efficacy may be debatable, there is no question that the mechanism, going forward, will not fully address the "over-recovery of generator costs."

That is because, as previously reported, FERC, at the behest of generators, has adopted a change that will maintain higher capacity prices even in years following high energy prices, by setting the highest point on the Variable Resource Requirement Curve equal to the greater of gross Cost of New Entry (CONE) or 1.5 times Net CONE (see 1/31)

"[U]se of an energy-only market to signal supply shortages does not somehow reduce the cost of generation," Exelon claims.

Matters does not dispute that market design does not change the actual cost of generation itself, but market design certainly impacts the cost of such generation borne by load, or rather, the revenues received by generators.

"Regardless of the resource adequacy policy employed, the cost to build a new generation plant is the same and thus the cost to consumers would be the same under any regulatory regime," Exelon said.

Again, on the first part, Exelon is correct, the cost to build a new generation plant is the same regardless of market structure.

However, just because a power plant costs the same under an energy-only market and capacity market, doesn't mean that load will pay the same amount for such generation under both scenarios. Indeed, the use of a capacity market essentially guarantees that the total cost of generation (capacity plus energy and ancillaries) will be higher than would be the case in an energy-only market.

It's simple to understand why. It's important to remember that the capacity market and energy market are separate markets, with each clearing different products. However, unlike with energy and ancillaries, there is no "co-optimization" of the energy and capacity markets -- that is, there is no mechanism to ensure that the markets produce a combination of resources to meet both objectives at the lowest price.

Consider that the capacity market, which clears three years ahead of the energy market, clears the units with the lowest going-forward costs, which generally favors existing assets, without any regard for their costs to produce actual energy. In other words, because they have been largely or fully depreciated, 50-year old steam units, while marginal in the energy market because of their inefficiency, are essentially guaranteed to clear in a capacity market.

However, by guaranteeing these units' fixed cost recovery, it ensures that three years after the capacity market clears, they will be setting energy market prices, and load will be suffering high clearing prices.

The capacity market essentially creates barriers to entry and energy market price relief, by subsidizing resources which clear the capacity market (which favors depreciated incumbents) with capacity payments, while denying such payments to resources with higher upfront fixed costs (such as new, more efficient units) -- even though these resources will provide relief to load in the form of lower energy market clearing prices.

An energy-only market removes this inefficiency and aligns revenues with dispatch order, incenting the generation with the lowest combination of fixed and variable costs, because all such costs must be recovered in the same market.

Email This Story

HOME