America’s Biggest Grid Is Losing Its Balance

Here’s a dilemma: Large chunks of fossil fuel-powered energy generation are scheduled to fall off the U.S. electric grid in the next decade thanks to economic and regulatory pressures. Even larger chunks of renewable energy generation have not yet been approved to connect to the grid and may not be for years, if ever . Meanwhile, data centers and electrification have kicked off the f irst notable demand growth for electricity markets in over 20 years. On top of all that, the grid has become increasingly vulnerable to climate change-fueled disruptions, whether from solar power being knocked out by hail or natural gas lines freezing in an ice storm.

In some parts of the country, the solution to this dilemma is relatively simple. In much of the Southeast and -west, large utilities that own power plants are simply building more natural gas power plants . In California, regulators are mandating that utilities procure enormous amounts of energy storage, and have rejiggered residential solar rules to encourage more combinations of solar panels and batteries. And Texas is planning to lend billions of dollars at low interest rates to help finance natural gas plant construction.

Then there’s the PJM Interconnection, the 13-state electricity market serving much of the East Coast and Midwest, run by the country’s largest regional transmission organization. Despite PJM’s constant warnings about natural gas and coal generation retiring, it has not been able to bring new generating resources online in a reasonable timeframe. The grid operator — technically a non-profit — has neither the regulatory muscle nor the financial firepower to shape new energy generation to its preferences; its interconnection queue got so long, it instituted a two-year pause on new applications .

While many of PJM’s problems are unique to its particular circumstances, they’ve gotten so severe in recent months, it calls into question whether the decades-long project of structuring electricity generation, transmission, and distribution into something like a market is even working anymore.

“The whole premise is that a capacity market is about efficient entry and efficient exit,” Abe Silverman, an assistant research scholar at Johns Hopkins and former New Jersey utility regulatory official, told me. “We’re squeezing the tube on the entry side and letting very few new entrants in.”

According to PJM’s independent market monitor, at the end of last year, there were just over 7 gigawatts of natural gas projects in the queue, about half of which it expected to go into service eventually, while some 24 gigawatts to 58 gigawatts of coal and natural gas is expected to retire by 2030. There were over 200 gigawatts of renewables projects in the queue, the market monitor said, but only around 30 gigawatts that’s expected to go into service, and for the purpose of a capacity auction, only about 11 would count.

But for power market observers, the sirens really started going off at the end of July, when PJM held what’s called a capacity auction, which determines the price companies get paid to supply energy-generating capacity over and above forecasted peak demand in order to avoid blackouts . By the end of the five-day process, the cost of that capacity came out almost 10 times higher for than the previous PJM capacity auction — $14.7 billion, compared to just over $2 billion in 2022 — a signal that supply, demand, and reliability dynamics within PJM are seriously imbalanced.

That almost certainly means rate increases for consumers. In Maryland specifically, some residential electricity bills could rise anywhere from 2% to 24%, a monthly change of $4 to $18, according to the state’s Office of People’s Counsel .

What that almost certainly does not mean is a huge amount of new generation coming online. “In an efficient capacity market structure, the market starts sending higher price signals and generators start coming on-line,” Silverman told me. “Usually when you see high prices, you would expect more of a response from the supply side.”

In PJM, however, “new generation cannot come online quickly,” according to a letter from a group of consumer advocates in PJM states , therefore “the high capacity market prices are not an effective signal for new entry but instead a windfall for the owners of existing generation.”

Ironically, the high prices were due, in part, to PJM applying a formula it typically reserves for renewables to coal and gas plants, which “derates” the capacity they’re able to offer in times of stress, e.g. during a winter storm. Historically , coal and gas got high ratings because high winds and cold temperatures was considered unlikely to disrupt their production, while solar and wind scored much lower. But after 2022's Winter Storm Elliott, during which natural gas lines froze and caused a mass blackout, PJM knocked down the rating for combined cycle gas plants — the most efficient kind of gas plant, which recaptures heat exhaust to produce more power — from 96% to 79%, and for combustion turbine natural gas plants from 90% to 62%. Wind got a bump, while solar was rated down.

In other words, “PJM doesn’t view all these megawatts as reliably as they did before Elliott,” Nicolas Freschi, a senior associate at Gabel Associates, which does energy and environmental consulting for federal agencies, told me. That meant some 26 gigawatts of projected coal and gas capacity disappeared from the auction, according to S&P Global Commodity Insights .

The environmental activist community has long argued that gas is less reliable than utilities and the public seem to think it is, and that this should be taken into account with grid planning. The gas derating was “a good thing,” Claire Lang-Ree of the Natural Resources Defense Council told me, “because that means what we're paying for in this auction is actually reliable. It's a truing-up of the system.”

At the same time, she acknowledged, the auction result was “a bad thing insofar as it was the driving cause of the price spike,” which also means huge payouts for power companies.

“Despite the decrease in capacity credit, the higher capacity prices will impact the capacity revenue received for projects in PJM, generally increasing it,” S&P analysts wrote in August . By way of example, S&P looked at one natural gas plant in Ohio and found that its project per-megawatt-hour net revenue in 2026 would increase by 40%.

Morgan Stanley estimated that major power producers such as Texas-based Vistra and Maryland’s Constellation Energy would see a boost to their earnings before interest, taxes, and amortization of $700 million to $800 million each.

And yet in both Texas and PJM, many analysts (not to mention the gas industry) still see gas as the solution to a shortfall exacerbated by gas’s documented vulnerability. That’s due to its ability — at least on paper — to generate large amounts of power at any time of day.

So far, however, only one power producer with a large natural gas fleet, Calpine, has publicly indicated that it will aggressively pursue development in PJM. Calpine operates a 76-facility fleet that includes 66 fossil fuel-fired plants from California to Massachusetts. “The PJM market needs and values reliable, dispatchable, non-duration-limited power” the company said in a press release . (These are all industry code words for natural gas.) Calpine said it was “accelerating its PJM electricity generation development program following market signals indicating higher demand for reliable power,” and that it was looking at “multiple new locations in the PJM region, particularly in Ohio and Pennsylvania.”

Other companies have been more cautious. “It is only one auction, of course, and not long enough out in the future to be starting a new project,” Vistra chief executive Jim Burke said in an August earnings call. Morgan Stanley analysts noted that because the next auction is in December, “we don't foresee enough time to build significant new generation capacity. There are only 18 months between the auction and the start of the delivery year, which doesn’t leave time for permitting, interconnection queue timing, and construction because they are behind.”

S&P forecast that only one natural gas project under construction in Ohio could possible bid into the next auction. And while stock and bond analysts are more focused on the prospects for new natural gas plants, they are not particularly optimistic they’ll come online any time soon. “Merchant newbuilds remain marginal under our assumptions, indicating price signals may need to improve further to incent merchant new entry,” Guggenheim analyst Shahriar Pourreza wrote in a note.

Todd Snitchler, the head of the independent power generator trade group Electric Power Supply Association, noted to me that the July auction price was “coming off a record low,” and that the “abnormally” low prices in the previous two auctions — which were then followed by a lengthy delay — “suggested that assets should be leaving, and not coming on” — a trend PJM and other electricity market overseers have been warning about for years.

“One auction does not make a trend make,” Snitchler said.

If prices stay high, however, some analysts think power producers will eventually start trying to build new natural gas plants in PJM. “Investors don’t want to start building extremely expensive projects until they’re sure this price environment is sustainable,” Freschi told me.

Instead of beckoning new gas construction, clean energy and ratepayer advocates want PJM to focus on interconnection reform so that its existing queue — which is overwhelming renewables — can finally make its way onto the grid.

In a statement to Heatmap, PJM said its new system of evaluating projects in groups instead of on a first-come, first-served basis will lead to 230,000 megawatts being processed over the next three years. The PJM spokesperson also pointed to Calpine's announcement as a sign that the capacity auction was bringing new investment.

“We need investment in real projects that can get connected to the grid quickly, as opposed to the speculative projects that have clogged the queue in the past,” the spokesperson said. “Our reformed interconnection process encourages projects with the best chance of being built, and we are weeding out some of those that have been hanging on for years past receiving an interconnection agreement from PJM and who have not moved to construction.”

“Generators should submit their new project queue positions today,” the spokesperson added.

But like so many projects clogging the queue, these reforms are speculative, and in the end the restructured market, where new supply supposedly responds to high prices, simply may not work on its own terms. Some of this is due to policy in PJM states — you’re unlikely to be able to build a new natural gas plant in Democratic-controlled states like Maryland, New Jersey, or Illinois, and Guggenheim’s Pourreza wrote that “any new gas generation will be clustered in [Pennsylvania, Ohio, and West Virginia],” which could both lead to lower capacity prices in some areas and a more unbalanced market as new gas capacity becomes concentrated geographically.

But even in areas that are famously friendly to fossil fuels and have less complicated market and interconnection processes, demand for new gas has not smoothly resulted in gas plant construction. In Texas, which has closest thing to a free electricity market that exists in the United States, the state has had to turn to a multibillion low interest rate financing program to entice developers to build new natural gas plants.

May that be a warning to regional transmission planners everywhere. As S&P analysts wrote, “High prices signal the need for new generation, but do not guarantee it.”

Read more: This Is How You Die of Extreme Heat