The effects of the January 2014 polar vortex have led to big changes in the PJM capacity market, as the board has approved changes to tighten up operations and improve reliability.
The polar vortex storm of Jan. 6–8, 2014, saw temperatures plunge across the Midwest and Mid-Atlantic, causing a spike in demand for electricity and natural gas. Peak demand for electricity was 25% higher than typical January peaks. This high demand drove up prices for both power and gas. PJM reported that electricity billings for customers in January 2014 were one-third of the entire year’s total billings in 2013.
The worst came on the evening of Jan. 7, when the region set a new wintertime peak demand record of 141,846 MW just as power plants started to fail in the cold weather. During the peak demand hour, 22% of all generation capacity—including coal, gas, and nuclear—was out of service, well above the normal range of 7% to 10%.
Plant outages were due to both mechanical failure—frozen pipes and gearboxes—and contractual failures, especially by natural gas generators. Most merchant gas generators didn’t have firm contracts for fuel delivery, and when demand rose for space heating, there wasn’t enough to go around. Dedicated gas storage options are limited, as delivery companies typically rely on “just in time” delivery.
The crisis also exposed a problem with demand response (DR). Most DR resources are geared toward cutting demand during lucrative summer peaks, not during winter.
A Proposed Solution (That Puts Pressure on Less-Reliable Fossil Plants)
To address these shortcomings, PJM has proposed a new “capacity performance” product that is intended to incentivize power plant owners and DR providers to provide more certainty during peak periods, in winter as well as summer. This could mean further winterizing power plants, adding dual-fuel capability, and signing firm fuel contracts for gas. For demand response, it could mean a greater emphasis on year-round performance rather than summer peaks.
But according to Andy Ott, executive vice president of market operations for PJM, the changes go beyond weatherizing the system for winter cold. “It’s a fundamental change in what capacity is,” he told POWER in an interview. “It does deal with the winter issue, but it is much broader.”
“If we had simply made an edict to winterize power plants, and 30% of the fleet was getting inadequate revenue, their response would be to retire.”
Overall, he thinks the changes will serve to cull the weakest and least reliable plants from the herd.
“We have some units that have been hanging around, high-cost units that run only occasionally, just living on capacity payments, but that don’t perform well,” he said. “They will get priced out of the market and get replaced with more efficient resources. Like units with a 70% forced outage rate, they don’t run when called but still get a capacity payment.”
“Some units won’t be able to make the transition—that’s the whole point.”
The PJM board okayed the new rules and submitted them to the Federal Energy Regulatory Commission (FERC) for approval on Dec. 12. They are largely based on ISO-New England’s “Pay for Performance” market design that was approved by FERC in June.
Following ISO-New England’s Lead
ISO-New England was moved to revamp its Forward Capacity Market system because it was “not achieving its most basic objective: ensuring grid reliability in a cost-effective manner,” according to the ISO-New England website.
When that system operator filed its “Pay for Performance plan,” in January 2014 it noted that “resource performance has been declining and the poorest performing resources continue to be paid.”
For example, in response to extreme hot and cold events in 2013 and 2014, “the ISO alerted all resources well in advance to be available and ready to perform. Nevertheless, several generators were unavailable when they were needed most.”
Pay for Performance will involve a “two-settlement system.” Suppliers will commit three years in advance to supply capacity through the Forward Capacity Auction but will face a second settlement, during the delivery year, based on actual performance during scarcity conditions. Resources that don’t deliver will have to cover their obligations by buying power from other suppliers in the pool that can deliver.
“Put simply, resources that perform well will be paid more, and resources that perform poorly will be paid less,” the system operator declared.
PJM Seeks to Create Appropriate Incentives
The goal is similar with the new PJM filing, to create “stronger incentives within the existing capacity market structure . . . to encourage needed investment by both new and existing resources,” according to PJM President Terry Boston.
Under the proposed rules, PJM will transition over the next three years to a single “Capacity Performance” product in its forward capacity auctions (known as RPM). Capacity Performance products will have to be available year round, and “must be capable of sustained, predictable operation that allows the resource to be available to provide energy and reserves whenever PJM determines an emergency condition exists.”
Meanwhile, current capacity products (Capacity Resource and Base Capacity Resource) will be phased out because they “cannot meet the region’s reliability needs throughout all portions of the year.”
PJM will be able to designate “performance hours” during emergency conditions, when resources will be paid to provide energy. PJM expects to call about 30 performance hours a year, though frequency will vary with the weather.
Every resource will have a “performance obligation,” its share of system requirements during those emergency hours. If the resource fails to deliver on its obligation during compliance hours, it will pay a penalty. There will be “no excuses,” according to PJM.
The Road Not Taken
Staff had originally proposed in their Aug. 20 paper that a Capacity Performance resource must be able to operate 16 hours a day for three consecutive days under extreme weather conditions. This was roundly supported by generators, such as Exelon, Calpine, and PSEG—owners of 38 GW of generation capacity in PJM—in joint testimony.
“The purpose of the eligibility requirement is to reduce the likelihood of non-performance by allowing only reliable generators to offer a CP product in the first place,” the generation companies wrote in a filing.
In the end, PJM management did not follow this path. “PJM is not dictating how a unit is capable or becomes capable of meeting the performance obligations; only that the unit does meet such obligations,” it wrote in its FERC filing.
Wind and Solar Could Benefit
In an interesting Robin Hood twist, PJM will take the penalty payments from under-performing resources and pay them to those that deliver more than they have committed to during those hours. Penalty payments will be derived from the Net Cost of New Entry (CONE), the cost of building a new combustion turbine or combined cycle gas plant, divided down to the hourly level. That currently ranges from $120,000 to $200,000/MW-year, or upwards from $325/MW-day.
Ott estimates these payments could be worth as much as $4,000/MWh.
This will be especially important to variable generation like wind and solar. Wind is given a capacity value of 10% to 13% of nameplate output, while solar gets 30%, based on typical output during periods of peak demand. If these resources produce more during the proposed “performance hours” they stand to reap huge benefits.
“In the winter of 2014 we had 2,500 MW of wind at peak, with only 500 MW of capacity obligation,” explained Ott. “Under the capacity performance system they would have been paid $4,000 per MWh for the 2,000 excess MW. That’s pretty good.”
This would also apply to generators that can exceed their standard output for short periods and to imports from outside PJM.
Moreover, variable and seasonal generators can buddy up with other resources to make “coupled offers.” Solar, peaking in summer, could join up with winter-peaking wind to increase the odds of performing during peak hours year-round.
The Challenge of Year-Round Requirements
Nonetheless, the new rules have raised concerns from energy efficiency and renewable energy stakeholders, based on the year-round requirements.
Jennifer Chen of the Natural Resources Defense Council compared the fleet of different power plants and DR resources in PJM to the Fellowship of the Ring, the band of hobbits, wizards, humans, and elves in the Lord of the Rings series.
“Each has different strengths and weaknesses, but as a diverse team, they overcome all sorts of obstacles,” she wrote in a blog entry. “PJM’s proposal seeks to ensure that every team member can more or less individually triumph over another Polar Vortex by requiring nearly all resources to be available every day of the year even though these types of events happen only once every 10 to 24 years for a few hours.”
“This proposed all-year availability requirement would largely eliminate from the team (drive out of the capacity market) renewable and demand-side resources with availability dependent on weather and consumer behavior,” she adds. “But these were the very resources that helped preserve reliability last winter, and their elimination is a casualty of requiring year-round availability for seasonal needs.”
But PJM is concerned that the DR resources that played a critical role in dealing with the polar vortex did so only on a voluntary basis. “The significant value provided by Demand Resources during these winter events and the lack of performance obligation demonstrates a need” to make them mandatory, the system operator wrote in its filing.
To make demand response more firm, PJM proposes to eliminate the seasonal DR products and replace them with a single year-round product, with much more stringent requirements. DR resources would now “be required to reduce load on any day of the year, for an unlimited number of interruptions,” and curtailments “will no longer be limited to a maximum of 10 hours in duration.”
As noted above, like other resources, demand response would be able to make coupled offers, marrying winter and summer resources. (For more on DR issues, see “FERC Order 745 and the Epic Battle Between Electricity Supply and Demand.”]
Expected Effect on Prices
Ott expects the changes to raise capacity costs by $5 billion a year in 2018/19 but to reduce energy costs $2.4 billion a year by increasing supply during peak periods. This works out to about 0.2¢/kWh or $2 to $3 per month per household. Because energy costs are 80% of the charges in PJM, “even a modest change in the energy price will offset the capacity price,” according to Ott. In more extreme weather periods the savings will be larger. If the system had been in place in 2014, during the polar vortex, he said, it would have saved $7 billion.
“It’s like insurance,” he said. “You’re paying up front for capacity, and you may or may not need it. But in an extreme year, it more than pays for itself.”
Others think the effect could be much greater. Former Illinois Power Agency Director Mark Pruitt thinks the capacity performance reforms could raise capacity prices to $272/MW-day, 118% higher than today’s prices. In an analysis for Crain’s Chicago Business, Pruitt estimated the changes “will funnel more than $560 million in additional revenue [in 2020] to five of Exelon’s six Illinois nuclear stations.” (Note that this sort of capacity market change is what Exelon has been saying it needs to keep its nuclear units economic.) Energy prices paid by ComEd customers would climb 1.4¢/kWh, or 19%, from today’s 7.4¢, he thinks.
UBS market analyst Julien Dumoulin-Smith sees a similar impact, with capacity prices rising from $120/MW-day in the most recent auction (for delivery in 2017/18) to $204/MW-day the following year.
If approved, the rules will phase in starting with the 2016/17 year and be fully implemented by 2020/21.
To improve winter capacity during the transition, PJM will procure up to 2,500 MW of additional capacity resources for the period December 2015 to March 2016 using a reliability-must-run mechanism.
—Bentham Paulos is a freelance writer and consultant specializing in energy issues.