The ISO graples with the politics of scarity.
Fortnightly Magazine - July 15 2003

The ISO graples with the politics of scarity.

In regions that have embraced electric industry restructuring, such as New York, New England, and the mid-Atlantic states, where independent system operators (ISOs) have taken over and the standard market design (SMD) has grabbed a foothold over bulk power transactions, one fascinating question still dogs theorists and policymakers alike:

Is a power supply shortage really all that bad?

For more than a century now, the typical power industry engineer has dedicated his life's work to guaranteeing freedom from blackouts for every man, woman, and child, to a certainty that approaches five "nines" to the east of the decimal point. And the engineers have largely succeeded.

But the SMD puts little faith in such efforts. It relishes abundance and scarcity in equal measure, as it weighs the contrasting forces of supply and demand to reveal the energy price and clear markets. During those brief intervals of greatest demand, it relies on gas-fired turbines to supply power to serve the peak. But those turbines seldom run except during periods of extremely high prices. In fact, they cannot recover their fixed costs and high operating costs without the sort of price spikes that scarcity can produce.

The SMD won't work without scarcity, but regulators and politicians cannot survive in a world where they cannot protect ratepayers and voters from high power prices.

Nevertheless, there's no verdict yet on whether regulators will tolerate scarcity or seek to banish it entirely. In fact, it's a good bet that they'll try to have it both ways. If you want evidence, just look to New England, where the ISO has recently proposed three new radical pricing schemes that on first glance might seem to run at cross-purposes. These proposals would:

(1) reduce high prices when a supplier with a dominant market share (a "pivotal" supplier) starts submitting unusual or suspicious bids;

(2) guarantee that high-cost, gas-fired peaking turbines that aren't very profitable most of the time can raise bids and prices as high as necessary to recover all their costs, including fixed costs and capital investment, by setting a "peaking unit safe harbor" (PUSH) threshold; and

(3) ensure that prices will climb to the moon during a period of extreme scarcity by permitting the real-time energy price to rise to the absolute highest allowed level (the region's $1,000 bid cap) and perhaps a bit more, depending on line losses and congestion uplift.

If this policy package seems a little hard to believe, turn to our "Commission Watch" article ("When a Cap Becomes a Floor," p. 14), where I lay out a few more details.

Figuring out how this came about requires a look at rules governing reserve margins, plus an understanding of how certain shortcomings in computer software have literally backed the ISO into a corner.

NEW ENGLAND'S MOVE TO BOOST PRICES AS HIGH AS POSSIBLE DURING A SHORTAGE OWES A DEBT TO ECONOMIST DAVID PATTON, the independent market advisor for ISO New England and the New York ISO. Patton's theories turn the familiar "duty to serve" upside-down. If you read him closely, he seems to be saying that it's more important for grid operators to maintain operating reserves than to deliver electricity to the customers themselves.

As Patton explains, reserve requirements are just that; they are requirements. As Patton sees it, "The SMD model and ISO operators must dispatch all available energy resources in order to maintain the required reserves." Thus, during shortage conditions, when the ISO can meet energy demand only by compromising the reserve requirement, the market is not really clearing. "Although energy demand is met," says Patton, "the operating reserve requirements are not satisfied."

Patton's answer during such periods is to boost the price to the top-all the way up to the safety net price cap of $1,000 per megawatt-hour-since the market needs more supply to clear. And the first and most important place to inject that supply is to bring the backup reserve up to speed.

"The relevant demand in this case," says Patton, "is the demand for operating reserves … the operating reserves have become the marginal source of supply."

You'd think that the utility's duty to serve has now become the ISO's duty to maintain reserves. As Patton puts it, "One must figure out what the market operator would have paid an incremental energy supplier to provide one megawatt of energy-allowing the operator to restore 1 MW of its operating reserves."

Of course, you don't have to be an expert to guess that this idea has won brickbats from the utilities that will be left paying for this peak-priced power.

In a protest filed by attorneys John Coyle of Duncan & Allen and Robert McDiarmid of Spiegel & McDiarmid, the coalition argues that scarcity pricing will not encourage gen plant construction.

Logic dictates, they say, that the opposite will occur: "Rewarding existing generators with higher prices when supplies are scarce will 'incent' generators to withhold supplies" and seek out expanded opportunities to reap monopoly rents.

MEANWHILE, SHORTCOMINGS WITH COMPUTER SOFTWARE STILL MAKE IT DIFFICULT FOR GRID OPERATORS TO PROTECT ALL THE HIGH-COST GAS TURBINES. This problem arises because certain operational limitations will sometimes force system operators to dispatch plants out of merit, when the bid would not have qualified on the margin. Thus, even when the system operators dispatch a high-cost gas turbine, they don't always add all the turbine bids into the pot to determine the locational marginal price (LMP). Listen to the New England ISO.

"In LMP markets," says the ISO, "the price is not set by simple stacking algorithm at each location.

"Some of the DCA peaking units [peakers that operate in designated congestion areas] are inflexible, with lengthy start-up, minimum-run, or minimum down-time characteristics."

As the ISO explains further, the inflexibility of some units is such that during extreme peak periods the units must run continuously for several days at a time so that they will be available during a few hours each day.

"Assume an inflexible unit with an offer price of $500/MWh operating at its low limit, but the nodal LMP is $100. … If the market rules were changed to permit the inflexible unit to set nodal LMPs at $500 ... a market participant could easily find itself forced to buy incremental energy to serve load at $500/MWh at the same time that its own generating resource priced at $100 is available but not dispatched.

"In a worst case scenario," adds the ISO, " these perverse incentives could diminish [our] ability to operate the system."

In its landmark order, issued in April , the Federal Energy Regulatory Commission told the New England ISO to allow all those high-cost peaking turbines to set the LMP price in the day-ahead energy market, even when they receive a subsidy under the new PUSH bidding rule.

So far, the New England ISO has said it cannot meet FERC's request because of the software problems noted above. But it says it will allow three-quarters of those subsidized turbines to set the LMP.

As the ISO notes, that should serve well enough to provide "rough justice" to FERC's SMD vision, but not a full guarantee of cost recovery.

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