The $29 Billion Power Bill: Inside PJM’s Capacity Shortfall

The power grid is hitting its limits. In this episode, we break down the critical results of PJM’s capacity auction for the 2028/2029 delivery year, where the price cap of $325 per megawatt-day was hit for the third consecutive auction.

We examine the widening gap between skyrocketing electricity demand and lagging supply, and what this means for future grid reliability.

Key Topics Covered:

The 6,830 MW Shortfall: Discover why PJM cleared 149,181 MW of capacity, falling significantly short of its desired 20% reserve margin and leaving the grid at a tight 15% margin under worst-case scenarios.

The Data Center Cost Explosion: Actual and forecasted data center loads have added a staggering 6.3billiontothisauctionalone,makingup4629.4 billion) of all capacity-related charges over the last four auctions—costs that ultimately fall on everyday ratepayers.

The $555 Simulated Reality: Learn why PJM's simulated auction shows that without the current price cap, prices would have soared to $555/MW-day across most markets, and reached nearly $777/MW-day in the Chicago area.

Why New Generation is Sitting Out: We look at why only 525 MW of new supply cleared. Skyrocketing costs for gas turbines—which are up 15% in a single year and 50% over the last five years—combined with manufacturers being sold out through 2030, mean developers are struggling to build new generation.

The September Backstop Auction Loophole: We explain how developers might be bypassing the main auction to wait for September's Reliability Backstop Procurement (RBP) auction, which offers longer contracts (2 to 15 years) and a higher average price cap of $555.

The Order 2222 Mystery: We discuss the surprisingly low turnout of demand response and why absolutely zero aggregated distributed energy resources (DERs) showed up to the party, despite being eligible to participate for the first time.

As FERC Chair Laura Swett warns that this consecutive shortfall is an "alarm bell" for the industry, we ask the tough question: Is PJM’s complex system of price caps and regulatory patches sustainable, or are we heading toward a Texas-style market with no capacity guarantees at all?

If you enjoyed the video, please Like, Comment, and Subscribe to stay updated on the future of the energy grid!

Transcript

Why PJM And ERCOT Matter

SPEAKER_00 0:02

Try as one might, it's pretty hard to escape the gravitational field of news from PGM and ERCOT. They seem to be where tomorrow's power grid is being shaped these days. It's not that other markets don't matter, but the fault lines between data center-driven demand and supply are most evident in these two markets. And the resulting regulatory responses have repercussions that will extend well beyond these two markets into the rest of North America. Plus, together, these two grid operators serve close to 30% of US GDP. So that also is important.

The 2028 2029 Auction Results

SPEAKER_00 0:36

Today we'll discuss the capacity auction results for PGM's 2028-29 delivery year. That's June 1st to May 31st. They were, in a word, consistent, which means there were no huge surprises. Once again, the price hit the $325 per megawatt day administratively approved cap that PGM put in front of FERC. And once again, the desired quantities of capacity fell short of the amount deemed necessary to meet the planned reserve margin of 20%. Let's get into the numbers and discuss what they mean and why they matter. First, the amount of capacity that cleared the auction was 149,181 megawatts. Second, that amount was 6,830 megawatts below what PGM calculates as necessary to meet its desired 20% reserve margin. That can ensure grid reliability under worst case conditions. An increased shortfall from last year's 6,516 megawatt shortage. More on that later. Third, the total level of new supply, including uprights of existing generation, where they increase the availability of existing assets to generate more power, well, that was just under 525 megawatts. Not very impressive in terms of the ability to attract newcomers. And fourth, and the number everybody pays attention to, the auction price once again cleared at the price cap, this time at $325 per megawatt day. That's the third auction in the row in which that cap has been hit. One last number to focus on.

What Prices Would Be Without Caps

SPEAKER_00 2:07

PDM ran a simulated auction, absent the price caller, the floor sits at $175, the caps at $325, and it found that in all markets except the Chicago area, the price would have cleared at close to $555 sounds the cap. The Chicago area served by Commonwealth Edison would have been even further pummeled at almost $777 per megawatt day. Putting those numbers in perspective, from 2022 to 2025, before the massive data center forecast of loads entered the picture, prices ranged from $28.92 to $50 per megawatt day. And this new higher price, much higher price, pretty well explains why data loads are such a political hot potato and a lightning rod for opponents who fear price and rate increases resulting from new data load, their right to do so. Per an email to Utility Diet from Joseph Bowering, the head of the Independent Market Monitor, who reviews PGM and reports to the FERG, in this latest auction, data center actual and forecasted load increased costs by $6.3 billion. With a total of $29.4 billion of increased costs over the past four capacity auctions resulting from forecasted and existing data center load. That's 46% of total capacity related charges. But what this means is that all of the ratepayers are paying for this new data load that's coming into the grid without sufficient supply to meet it. So now we have the numbers.

Reserve Margin Risk And Cap Tradeoffs

SPEAKER_00 3:39

What does this portend for electric energy future? Well, let's start with the fact that not enough supply attendees showed up at the party. PGM sets its desired reserve margin, as we mentioned, at 20%. PGM arrives at its reserve margin number by simulating thousands of scenarios across seasons, looking at demand, the likelihood of unplanned generator outages, and also determining how much power it can expect to import from neighboring grids. And then they determine how much they think they're going to need. Well, in this auction, they fell 6,500 megawatts short, leaving them at just under a 15% reserve margin. So it makes it a little bit harder to keep the lights on in a worst-case situation. The caps, even as they offer some relief to energy consumers, well, they also create the problem because they don't provide enough of an incentive for new generation to join the party. Predictably, politicians who argued for the caps celebrate how much they save their constituents, and they generally don't focus on the shortfall. And just as predictably, energy economists and planners and regulators, including PGM, note that this type of market intervention with price callers will increase future supply shortfalls. In its summary document, PGM noted that, quote, a price cap, dot, dot, dot, can reduce the amount of investment and therefore supply in the PGM region.

Why New Supply Stays Away

SPEAKER_00 4:59

Personally, I wasn't surprised that almost no new generation came to the 28-29 based residual auction with liquor and gifts for two reasons. First, there's a potentially more attractive reliability backstop procurement auction plan for September, in which generators, new generators, can establish bilateral contracts with large loads, with delivery starting in the 2028-29 year, and you can get a far longer term from 2 to 15 years and a potentially higher price. The average price of that auction is capped at 555. Then there's the fact that PGM's recent white paper, written in May, which is worth reading, it essentially says the current approach to capacity is no longer working. It's not sufficient to ensure resource adequacy and new approaches should be considered. The white paper also noted that the cost of developing a new gas-fired combined cycle reference plan is much higher than the estimated cost of new entry, which is close to that $555 per megawatt day price. And that is because of recent supply chain issues and a significant demand supply imbalance with new turbines. GE, Mitsubishi, Siemens, they're essentially sold out through the end of the decade with orders well above existing factory production limits. Lazard, the financial analyst, they just released levelized cost of energy numbers in their 19th annual report, where they look at the all-in cost to generate a megawatt hour of electricity for different types of resources. And they indicated that a megawatt hour from a gas-fired combined cycle plant has increased in cost every year since 2019, with the 2025 to 2026 price increase soaring by 15% over the single year and up 50% over the past five years. So with the price, again, capped at $325 and these rising costs of generation, the shortfall should come as no surprise. We'll likely see another shortfall in the 2029-2030 auction that takes place in December.

Demand Response And DERs Underperform

SPEAKER_00 6:56

What did surprise me was the anemic amount of demand response and distributed energy resources that cleared the auction. PGM noted this was the first year that aggregated distributed energy resources, so-called DERs, under FERC's order 2222 were eligible to participate. And you might remember 2222 basically directs the grid operators to treat aggregations of DERs over 100 kilowatts like any other wholesale asset. But none of them showed up to the auction. And at the same time, the total demand response number declined from 7,299 megawatts last year to 7,017 megawatts this year. I'm personally at a loss to explain this one. So perhaps somebody watching this vid can clue me in. I would have thought that these higher prices in recent years would have elicited a bumper crop of demand response in DERs rather than the meager harvest we've been

FERC Sounds The Alarm

SPEAKER_00 7:47

seeing. The supply shortfalls weren't lost on FERC chair Laura Sweat, who commented this was the second year with a shortfall compounding the quote, alarm bell, unquote. FERCA scheduled a technical conference this week on July 23rd to address these issues, but don't expect any magic bullets. These issues are challenging, they involve billions of dollars of winners and losers, and there's no easy fix here. Well, thanks for watching, and we'll see you again soon.