North America’s Looming Electricity Supply Shortages

In December, the North American Electric Reliability Corporation, or NERC, released its annual assessment of grid reliability across North America. The results were concerning.

NERC, which is the organization responsible for setting grid reliability standards, reported that electricity supply is struggling to keep up with rapidly growing demand across much of the U.S. and Canada. In several major grid regions, electricity shortfalls could occur under challenging conditions within the next one to three years.

On the podcast, Abe Silverman, assistant research scholar at the Ralph O’Connor Sustainable Energy Institute at Johns Hopkins University, discusses the threat of electricity supply shortages with a focus on one area of the grid, in particular the PJM Interconnection. PJM is the largest regional grid operator in the U.S., serving 65 million people in the eastern part of the country. PJM recently announced that it, too, could face a capacity shortage as early as 2026.

To date, the grid operator has undertaken a complex set of actions to address its challenges, with more efforts on the way. Silverman explores PJM’s looming supply shortfall and examines the steps it’s taking to shore up supply. He also explains the controversies that some of these actions have raised.

Full Transcript

Andy Stone: Welcome to the Energy Policy Now podcast, from the Kleinman Center for Energy Policy at the University of Pennsylvania. I’m Andy Stone. In December, the North American Electric Reliability Corporation, or NERC, released its annual assessment of grid reliability across North America. The results were concerning. NERC, which is the organization that’s responsible for setting grid reliability standards, reported that electricity supply is struggling to keep up with rapidly growing demand across much of the U.S. and Canada. In several major grid regions, electricity shortfalls could occur under challenging conditions within the next one to three years. This challenge is being felt across the continent, but the responses to it vary by region.

On today’s podcast, we’ll focus on one specific area, the PJM interconnection. PJM is the largest regional grid operator in the US, serving 65 million people in the eastern part of the country. PJM recently announced that it could face a capacity shortage as early as 2026. To date, the grid operator has undertaken a complex set of actions to address its challenges, with more efforts on the way. On today’s podcast, we’re going to explore PJM’s looming supply shortfall and examine the steps it’s taking to shore up supply, as well as the controversies some of these actions have raised. Here to talk about all this is my guest, Abe Silverman. Abe is an Assistant Research Scholar at the Ralph O’Connor Sustainable Energy Institute at Johns Hopkins University, and he is also former general counsel for the New Jersey Board of Public Utilities. Abe, welcome to the podcast.

Abe Silverman: Hey. Thanks, Andy. It’s great to be here.

Stone: It’s great to have you back. I think the last time you were on the podcast was a couple of years ago, when we also discussed reliability concerns in the PJM market. Here we are again, although this time, I think it feels a little bit more urgent than we last spoke.

Silverman: It’s certainly getting a lot of folks interested, including a lot of politicians that, you know, have traditionally paid a little bit less attention to these markets. And all of a sudden, everybody is interested.

Stone: So I want to start us out by talking about the challenge that the electricity system faces today, kind of big picture. It’s really unprecedented in the modern era. We’re experiencing rapid change in the system, and this has left us in a position today where there’s growing concern about the potential for power shortfalls. Can you tell us what is changing and leading to this decline in electricity supply, relative to demand?

Silverman: Yeah, that’s a great question. And we really do have to talk about this both as a supply story and a demand story. And really, what’s going on is the supply is aging, particularly in the PJM fleet. And there’s a lot of retirements of existing fossil fuel generators that are leaving the system as they age out. And you know, at the same time, we’re having a lot of trouble getting in new supply onto the grid because of siting and permitting and the interconnection queue — the dreaded interconnection queue, which I’m sure we’ll talk about. And then at the same time on the demand side, really, for the first time in a generation, we’re seeing an incredible amount of new demand for electricity. Particularly in the PJM region, largely driven by data centers, but also by electrification of transportation and buildings, and really the return of manufacturing to the region. So it’s kind of this crazy story where you have supply really challenged, and you have demand growing extremely fast, and the resulting collision between those two is raising a lot of eyebrows.

Stone: As I mentioned in the intro, this is a national problem, but we’re going to focus on one place, one market in particular, and that’s PJM. I think PJM is interesting because it’s the largest electricity market. It’s also where a — you could say, a perfect storm of developments and a few about faces have complicated efforts to address supply concerns. And that that’s, I guess, culminated in PJM’s recent warning that electricity supply may come up short within the next couple of years. So it may appear obvious what I’m going to ask. But what, specifically, does PJM mean when it says it could face a quote, “capacity shortage”? End quote.

Silverman: PJM has really been talking about this issue for quite a while. It really started with the release of their 4Rs report in 2023, where they highlighted that the amount of new generation supply coming into the market was less than the amount of retirements and less than the amount of load growth. And since then, I think the situation has only gotten more concerning, from PJM’s point of view. And it’s fair to say that there’s actually a fair amount of debate between various people as to, you know, is PJM actually in a supply crunch, or is this something that can get managed through?

So what does it mean? You know, really what it means is, we have this very elaborate system to ensure the reliability of the bulk power system. And the details get very complicated. But effectively, we have these markets, and we tune them so that we don’t have more than one loss of load event — so one blackout as a result of not having enough generation on the system — once every 10 years. And, you know, that’s probabilistic, and PJM is always talking about ways to improve that metric. But really, what we’re talking about is making sure that there is enough generation to meet load needs under a variety of weather conditions, and a variety of financial conditions, such that at the end of the day, we really have a reliable system the vast majority of the time.

And that’s — you know, that’s pretty important to our way of life, right? Both economics and and, you know, the the livelihoods of people, and health and safety for the population. And you only have to look at what happened with winter storm Uri a couple years ago in the Texas area, where they had a really catastrophic grid failure that resulted in in several hundred deaths. So when we talk about these things, it’s very serious.

So now, what is PJM actually saying? PJM runs a forward market to make sure that we have the right amount of generation in the right places on the grid, ideally on a three-year-forward basis. Nhat is designed to make sure that power plants are built when there is a need for new supply. And I think part of the real concern from PJM is, we don’t have the same forward visibility that we would like, the way these markets were designed to have. Instead, through a variety of pretty complicated reasons, we’re actually now looking 12 months into the future, instead of having a crystal ball that looks 36 months into the future, the way the market was designed. What PJM is saying is that they are concerned that the amount of generation supply we’re going to have starting on June 1, 2026, is not going to be sufficient to meet the full reserve margin that PJM says is safe. And so it’s not like having a slight shortfall in the amount of generation means there’s an instant catastrophe. No. It means that, on a probabilistic basis, blackouts are slightly more likely than we would want to see them. Or really, load shedding, if we got to a point where the supply was not sufficient to meet the demand on the grid.

Stone: And I understand that that means, when we’re looking at periods where there is particularly high demand or other stresses on the grid, that’s when that real shortage in the risk of an actual blackout is going to be higher. Not so much under normal operations. Is that correct?

Silverman: Yeah. Generally, when we think about, when is the danger of grid reliability? You know, over the last 30 years, people have generally considered the middle of the summer as the time when, if the grid was going to run short of supply, it would happen on that really hot day in August. But over the last decade, particularly starting with the polar vortex in 2014 and winter storm Elliot in Christmas Eve, 2023, as well as what happened in Texas with winter storm Uri, there’s been this real realization that actually the biggest threats to system reliability often occur on those incredibly cold days in the winter. And so we have really struggled on those cold days to run the grid successfully.

And there’s a lot of reasons for that. Partially just, you know — I mean, electric generators are machines. And we all know machines don’t necessarily like cold weather. So when the weather gets cold, things tend to break. Things tend to freeze. Along with that, we’ve also had some real challenges getting natural gas supplies. Winter Storm Elliott was a big story about lack of natural gas supplies that were leading power plants to come offline. So you have this recognition by grid planners and reliability analysts that the resource adequacy risk, which is how we talk about these things, is really shifting from the summer into the winter. And certainly, you know, climate change and the rise of extreme weather is helping to contribute to that.

Stone: You know, I want to take a step back for a moment to something you said several minutes ago. And you talked about these capacity markets and the capacity auction, and the change in the timing of the capacity auction, which really kind of fits into this whole crisis that we’re potentially facing in the PJM market. But before we go into that timing issue, I want to make sure that people understand how PJM — which is a competitive wholesale electricity market — ensures, or is designed to ensure capacity through competitive forces. Could you talk us through that mechanism, the capacity market that PGM uses, again, to ensure that it has a enough generation in the future to meet its demand needs?

Silverman: Let’s go back, really, prior to the modern formation of PJM. Back then, individual utilities were responsible for going out, building, owning, contracting with sufficient supply. As we went through industry restructuring in the late ‘90s and early 2000s we really took that resource adequacy obligation away from the individual utilities and said, “No, we’re going to have a centrally procured pool of power plants that we’re then going to allocate across the entire region.”

Stone: And these are all competitive, right?

Silverman: It’s a competitive market, that’s right, where the supply and the demand are going to cross at a certain point on the demand curve, and that’s what’s going to set the price. And the key element here is that PJM says we need X amount of capacity. We need it in specific locations. And then we need a reserve margin, so that if we’re wrong about the guess on the weather, or we’re wrong about — you know, a couple power plants break on a critical day — that we’re not stranded. That we have a generation in reserve. So you basically, at the simplest, the very simple level, you take the total amount of load that you expect to serve. You say, “We need that much generation.” And then, on top of that amount, we also say, “We need somewhere between 15 to 20% spare capacity.” What we call the reserve margin. And that generally, if you have somewhere in that 15 to 20% excess, you will have a fairly reliable grid, because it allows for things to go wrong. It allows for load forecasts to be incorrect. And yet it still gives enough slop in the system to recover from from events.

And so what PJM does is, they run this somewhat complicated market — but at its heart, it’s really simple. It’s just a stock exchange for capacity. And we should talk about what capacity is. It is the ability of a power plant to generate power when needed. So almost think about it like a car lease. I lease a car. I could keep that car in my driveway on some days, but I’m still paying the lease payment. But then I have the flexibility to get in that car and drive around whenever I need it. That’s effectively what the what the capacity market is doing. So PJM is going out and entering into contracts with generators so that the generator will be there, ready to respond when necessary. And in exchange for taking that money, every generator sits there and puts a bid into the market for every five-minute increment of the 8760 hours of the year. And they have to either put in a bid or explain why they can’t. And then that’s what PJM uses to actually run the grid on a day to day basis.

Stone: Basically what we have here is we have this market, and it is the price signals that will again ensure that we have enough capacity when we need it. Explain that price signal.

Silverman: So PJM capacity markets are set in this rather abstract concept called dollars per megawatt day. So for every megawatt, for every day, you get a certain dollar amount. And that price is set very intentionally. It’s to encourage competition between generators so that a more expensive generator that says “I need more in capacity payments” is going to have to compete against someone who’s maybe willing to come in and do that job more cheaply. And the market will then procure the least cost suite of resources.

But now there’s a really important concept embedded in all this, which is, PJM recognizes that at some points we are going to need a private investor to step up and say, “I’m willing to commit a billion dollars for a brand new power plant.” And so what happens under the PJM market design is that when there’s excess generation supply on the grid, the prices are relatively low. And then as the supply and demand come closer into balance, the price starts to rise. And we have this whole concept called new entry pricing, or net cost of new entry, where eventually the prices in the capacity market will start rising and generators will either make investments to keep their existing generators online, or they’ll defer retirements, or if the prices get to a high enough level, they will actually commit to building new power plants and bringing them into the market.

And so the concept is, you know, that as the supply demand conditions tighten, the prices start to increase, and then can get very high if the market is really signaling the need for new entry. And the other element of this, which is really important, is PJM doesn’t just set one price for its entire footprint. You know, PJM is 13 states plus the District of Columbia. So it’s a huge geographic footprint. And within that region, within the PJM region, there are places where there is simply not enough generation, or where load is growing, or where there’s transmission constraints that prevent power from flowing across the whole PJM system. In those cases, PJM comes in and says, “Okay, under our market design, the price in that locational deliverability area, or LDA, starts to increase and can get higher.” So for example, in this past market, you had relatively high prices in Virginia and Maryland zones. And so that was really sending a price signal to generators that additional supply is located in those places, and hopefully going to attract new investment into the market.

Stone: So that is the invisible hand of the market guiding investment in resources where they are needed, to make sure there are enough resources in all areas and that pricing is reasonable for consumers. However, we’re seeing a situation now where we may not have enough capacity despite these market forces and these market signals. What is happening right now to put this supply-demand balance in jeopardy?

Silverman: So, we have this platonic ideal of how the market should work, and then we have reality. And the reality is extremely messy. And I would say there’s three ways that reality has intruded into the PJM market design over the last couple of years. One is that instead of having the market on a regular schedule with stable rules, a variety of factors have conspired to cause some really significant delays in how PJM is conducting its markets. So ideally, the way the market was initially envisioned, PJM would hold its auction three years in advance. So if there was a reliability problem, or if we were seeing we didn’t have enough new generation coming into the market, it would send a price signal three years into the future that would tell investors, “Hey, we need new generation here.” And for a variety of reasons, really leading back to some regulatory dysfunction that happened at the main regulator for PJM, the Federal Energy Regulatory Commission, back in the 2019 time frame, PJM got off track on having its three-year- forward market. And now, instead, is having an auction one year in advance.

So, you know, I like to think of the capacity market — the way it’s designed is, it’s supposed to act as a crystal ball so that we can see three years into the future and invest capital where needed. Instead, our crystal ball is on the fritz. And instead, we’re looking one year forward. Which, you know, as you can imagine, is not really enough time to build a major new power plant. And so part of what’s happened in the last couple of years, we don’t have that forward price signal the way you would want.

Stone: So you’ve got a market, you’ve got this capacity market, you have an auction. Supply tightening in a certain part of the PJM market is reflected in a high price in that auction. And that auction signals the need for new development, and that development of the new resources will be ready three years in the future. So we see the tightening. We have the market auction. And now we’ve got three years, as you mentioned, to build this new infrastructure, which takes time. So it will be ready three years in the future. Now, though, that’s been compressed, say, to one year, and you’ve got this capacity auction. High prices signal the need for new resources. But you can’t build this stuff in one year, right?

Silverman: Exactly right, exactly right. And, you know, the nice thing about a three-year-forward auction is, I’m a power plant developer, I could put a bid into that auction and see if it clears. You know. And even if— I may actually bid into a number of markets before I finally get the go signal. And that’s good. That’s the way it’s designed. And so at least I have that three years to bring the project into market, whereas right now I just don’t.

So that is one element of what’s going on in PJM, but the other elements are also super interesting. So the second major thing that’s going on in the PJM system is that we have load growth. Really, for the first time in a generation, we are seeing really significant new demand coming in. You know, it’s very interesting. When you go back and look, you can see the last time we had really significant load growth was the 1990s. You know. And really — you know, you can even go back further in time. Nineteen-fifties, we had 9% annual load growth, largely driven by new appliances. And then in the 1960s air conditioning came in, and you had 7% load growth. A little bit in the ‘70s, ‘80s and ‘90s.

And then in the 2000s, we really stopped seeing load grow. And you know, part of that was because we had more efficient light bulbs. You know, LEDs. A lot of other reasons. We didn’t have a huge amount of economic growth, generally. Industry was using less electricity overall. But then, really, in the last couple of years, load growth came back with a vengeance, largely —  though not entirely — driven by data centers. And in particular, artificial intelligence data centers, who are starting to come into the market in a really big way, such that we’re now seeing 3, 4, 5% load growth. And if you even look forward a little bit further than that, PJM is projecting 10% load growth by 2030, largely on the backs of data centers.

So at the very time we need to be looking into the future to accommodate this new load growth, the PJM market is kind of on the fritz, and simply isn’t sending that forward price signal. So you have less visibility into the future. And at same time, we have explosive and frankly, unanticipated load growth. You know, I’m very envious of anybody who saw this coming and bought Nvidia three years ago. They’re probably not listening to this podcast right now because they’re on some Caribbean island.

But, you know, we have seen this trend come in, where the load growth is changing even over the course of 12 months. Like PJM — you know, they do an annual load forecast. In 2024, they had one number, and then in 2025, they had to up that number again by really significant amounts. Multiple power plants of new load growth coming in every year.

Stone: Our perception of the future is changing so dramatically quickly, it sounds like.

Silverman: Yeah, absolutely. And so, you know, that would be challenging even if the market was working great. That would be challenging, to meet all that new load growth. And because the market isn’t working so well right now, we have all these other issues, it’s been very difficult to respond to that load growth appropriately.

And then the third chink in the armor, as it were, is the fact that the PJM system is having a really hard time bringing new supply to market. You know, this is the dreaded interconnection queue. The interconnection queue is this list of new generators that want to come into the PGM market and sell energy and capacity to PJM consumers. Years ago, the interconnection process took about two years. You sort of knew what you were going to expect, and it was within that three-year time window that you’d need to bring new resources into market.

Well, starting in the early 2000s, we started seeing a ton of new renewable generation coming into the market, and the PJM queue process kind of buckled under the weight of all those new projects wanting to come in. So this is a good news, bad news story. The good news is, the clean energy revolution is working. Developers want to develop clean. Buyers want to buy clean. But we can’t get it through the regulatory process to actually bring it to the grid in a timely manner. So at the same time that we need this new supply to meet the data center demand, we are seeing a real bottleneck in bringing new resources to market

Stone: Now, PJM did move to address that bottleneck. It reformed its interconnection process a few years ago. Has that solved the problem?

Silverman: It hasn’t. You know, it’s a step in the right direction. But the solution is a multi- year solution. And in the meantime, you’ve had several years of very little throughput through the interconnection queue. Now PJM is speeding up, and they will be the first to tell you that, yes, they had a couple of years where projects were not moving through the queue. But now, they’re trying to make up for lost time and push it through as quickly as they can. I think at the end of the day, you are still not seeing enough generation coming out of the queue to meet the anticipated demand.

And the other piece of it that’s a little bit complicated, and I think gets very, very political, very fast, is that the vast majority of generation that’s currently in the queue is intermittent generation. When we say intermittent, what does that mean, right? It’s wind and solar, for the most part. And so a huge part of the PGM queue is made up of these intermittent resources, which do have a lower capacity value than, say, a more traditional gas or coal or nuclear resource. You know, which, of course, operate independently of weather.

And so we have this kind of really unfortunate confluence, where load is growing extremely fast. The amount of new generation coming into the market is very low because of these constraints in the interconnection queue. And the PJM market is only operating in this sort of emergency, short term basis. So that is where — you know, that sort of sets the stage for what’s been going on in the last few months, where, you know, PGM has just been filing change after change after change to their market. And we’re seeing complaints come in talking about the PJM market, asking FERC to make all sorts of changes on an emergency basis.

Stone: It’s interesting. We’ve been talking for almost a half an hour now, and really all we’ve done is kind of set the foundation for the crisis that we’re in right now. That just kind of gets to the point of how complicated all this is. But you mentioned — several minutes ago, you mentioned winter storm Elliott. And then a couple months after that, in February of 2023, a couple of very interesting things happen simultaneously on the same day, on February 24th of that year. PJM comes out with the 4R Report, which you mentioned, which talks about for the first time, really, “Hey, we’re running into some capacity issues now, and looking out a few years, we may not have the capacity we need to ensure reliability of the system.” Number one. Also on that same day, the PJM board, understanding this crisis and really seeing what had happened with a near disaster avoided during winter storm Elliot, instructs the PJM market and all the participants in that market — the generators, the transmission owners — everybody to get together and find a way to ensure, to move forward, to make some changes to the market to ensure that capacity and reliability would be there in the future. Tell us what happened at that point, and what the first steps were for PGM to really aggressively go at this capacity crisis that’s looming.

Silverman: Yes, exactly. And it’s no accident that the board action took place at the same time that PGM management came out with this report. So let’s back up a second and go to that Christmas Eve 2022. Winter storm Elliott. The grid came within a hair of crashing. And there’s a lot of reasons for that. But at the high point, PJM had something like 50 gigawatts of natural gas that went offline during winter storm Elliott.

And there’s a lot of reasons for that. Some of it was problems on the gas system. Some of it was cold weather events. But we really had very close — and frankly, the closest the PJM grid has come — to a real reliability catastrophe. So PJM, I think, and all of us, spent the next couple of months really thinking about what happened and how do we avoid it from happening again? And that’s when, like you say, the PJM and the PJM board issued the 4Rs report. And the 4Rs report is really interesting because it goes through and really for the first time did a sort of holistic analysis, what we often call a balance sheet approach, to the amount of capacity in the PJM market.

The four Rs were resource, retirements, replacements and risks. So that’s what PJM was analyzing. And they were really doing it on the back of this winter storm Elliott problem that they had. You know, what do you do to avoid having the kind of catastrophic reliability event that we only narrowly avoided in winter storm Elliott?

So PJM came in and said, “There’s a few things that are happening. One is, we have a lot of retirements that are coming onto the PGM grid,” including retirements driven by state policies like the Clean Equity JOBS Act in Illinois, the EPA Good Neighbor rules, some of the coal combustion residual rules, some rules in New Jersey. So there was a lot of clean energy transition work at the state level that was leading to retirements of about 40 gigawatts, 40,000 megawatts of generation on the PJM grid. At the same time, they also said the amount of new generation coming onto the grid through the interconnection queue was extremely low. And part of that was because of the dysfunction in the PGM interconnection queue. But they were also pointing to the fact that even for the handful of projects that made it through the interconnection queue, they were really struggling to get built. And they pointed to some concerns about siting and permitting and financing and supply chain, particularly coming out of the pandemic.

So you have those real constraints on the amount of new generation coming into the grid. We’re losing some of the generation that we do have. And then they’re looking at load growth. And really enormous forecasted load growth. Back then it was electrification. Now it’s sort of switched over to the data centers. And so, you know, those various pieces come together to suggest that we are going to be very short in the amount of capacity we have.

Now, one of the things that makes the sort of head-exploding level of complexity associated with all this is not only are there all these issues, sort of on the ground reality intruding into the sort of standard market design — or at least the philosophical market design we’re trying —but then PJM comes in and makes a series of very important rule changes to the way that they count the amount of capacity participating in the market. You know, you could have, Andy, a whole session on effective load carrying capability, what they call ELCC. And, you know, there’s a lot of math behind it. It gets very, very complicated. But it’s effectively saying, you can’t have a system that is completely dependent on one type of technology. Like, for example, if you had a system that was100% natural gas, and there was a disruption to the natural gas grid, you would obviously — all the power plants would fail. And so there are some linkages between specific fuel types. And, of course, on the clean energy side, if the sun isn’t shining at a particular time, then you would expect to see low capacity produced, or low amounts of energy produced from all the solar farms in the PJM region.

So they came in and recalculated all the various ways of what they call the capacity accreditation for all the different types of facilities in PJM. They made a bunch of other changes to the market too. And so what happened coming out of the 4Rs report is, we had a number of new rules come in, and they all took effect in one auction. So we get to this past summer, and the auction cleared — you know, the PJM capacity market reliability pricing model — they had the auction, and prices came in way higher than most of us had anticipated.

Stone: After these market changes had been made.

Silverman: After these market changes had come in. In some ways, it’s like PJM turned all the dials to 11. And as a result, you know, they put in a much more conservative approach to counting capacity. They made all these changes at the same time. And that, in addition to the retirements, in addition to the load growth, and in addition to the constrained interconnection queue, conspired to significantly increase the consumer costs associated with the capacity market.

And there’s a couple different ways to think about that. The capacity market, on average, has cost consumers PJM-wide about $7 billion a year. For the last couple of years, it’s actually been much lower than that, somewhere in the $2 billion range. And so after the auction for the ’25-‘26 delivery year — so that’s the June 1 of 2025. The total cost to consumers of this capacity market went from $2.2 to $14.7 billion. Which is just an enormous year-over-year change.

Stone: That’s $11 billion more from one year to the next.

Silverman: Exactly. And I do think it’s important to talk about the average as well. So it’s about a doubling of the 10-year average cost for capacity.

Stone: Still a good $7 billion.

Silverman: Still, $7 billion. You know, that’s real money among friends. And I think it’s also important to talk about what this means for real on the ground consumers. So you know, it actually is surprisingly difficult to track through exactly what an increase in capacity costs does to your bill as a consumer in PJM. But generally, transmission and distribution is about half your bill. About half of it is energy. And about 20% of that bill is attributable to the capacity market. So you know, it’s a multiple times increase to that portion of the bill.

And so for Maryland, where Office of People’s Council in Maryland did a really nice analysis, they talked, about a customer in Baltimore would see about $16 a month in additional costs as a result of this one capacity auction. Which, when you’re thinking about energy poverty in America — which is a very real issue — and general affordability, that is an enormous increase. You know, on order of $200 a year for your average customer. And that’s a lot, attributable to this one change in the market. And of course, it’s not the only thing happening. We’re just seeing cost increases coming all over the place, including from increased transmission investment in other places, resulting in some really significant bill impacts for consumers.

Stone: I want to jump in on that point for just a moment. The governor of Pennsylvania, Josh Shapiro, became very upset about the results of this capacity auction from July of 2024 that we’ve been talking about. And he said, quote, “Pennsylvania rate payers face potentially the largest unjust wealth transfer in the history of U.S. energy markets.” He was upset.

Silverman: Yeah. And, you know, one of the great energy economists, Peter Crampton, talks about how when the lights are on, politicians view electricity as dull and technical. So long as the cost is reasonable, the lights are on, politicians are unlikely to participate in market design discussions. Well, you know, what’s happened is the politicians are involved. Because while the lights are still on, the reliability risk is increasing, and certainly the prices are increasing quite a bit.

So you do, you have Governor Shapiro, along with a number of other political entities in the PJM region, all of a sudden — you know, it used to be that the energy portfolio was kind of a sleepy — not really. Hasn’t been that way for years. But, you know, you certainly have real increased focus on the affordability issue. And so you have a number of entities, both consumer advocates in PJM as well as PJM itself and Governor Shapiro, all coming in and calling on FERC to make, effectively, emergency changes to the PJM market. And there’s a lot of moving pieces there. But one of the things that’s a little bit scary is — you know, without changes, without the filings that are currently pending on FERC —and, you know, spoiler alert, we don’t know what what FERC is going to do with these filings yet.

Stone: And this is in addition to the changes that you were just talking about that led up to the high prices in the 2024 auction last July.

Silverman: Correct. This is a second year, a second round of really significant changes. But if FERC does not approve these changes, then we are potentially looking at another doubling of the already extremely high prices for 2025, that would occur in 2026. And so as a result, you have PJM itself coming in and making a number of proposed changes that FERC has not yet ruled on. You have Governor Shapiro filing a complaint, asking FERC to impose a price cap on the market. And a lot of it is premised on this really simple idea that the market cannot meet the goal which was established, which is to incent private investors to invest capital to bring new generation supplies into the market, chasing those higher prices.

Stone: Couldn’t you see this —hey, the markets are doing exactly what they’re supposed to do. We saw high prices, right? So that means at some point in the future, we’re going to get an influx of new generation to make sure we’ve got enough capacity, and then the prices will go down. Isn’t this the way markets are supposed to work?

Silverman: Again, the pesky reality is coming in. Because we have this constraint in the interconnection queue. So investors, capital, cannot deploy, for love or money, new generation fast enough. Because they simply can’t get it through the study process, through the interconnection queue. And so if you think about, you know — all these markets are premised on the idea that there is a ready supply of new generation that is willing to come in in response to high prices. But that new supply has effectively been cut off by the interconnection queue debacle. And so there’s kind of this really interesting question. If prices rise, and firms can’t respond to those price signals because of a regulatory constraint, what’s the point in letting the prices get that high?

Stone: So what PJM is looking for, I believe, in these latest filings — and I don’t want to oversimplify here — but they seem like a lot of quick fixes. And this could get technical pretty quickly. But I think it might be — if we can just glance off the surface of this, just to give listeners just some some sense of what’s happening. There’s a bunch of acronyms here, right?

Silverman: Yep. Yep.

Stone: There’s SIS, and there’s RRI, and there’s RMR. And again, these are attempts to bring capacity on really quickly in the short term, so that we don’t see this doubling of prices that you warn could be happening in these capacity markets. Would you mind diving into any one of those acronyms that kind of catches your fancy?

Silverman: Yeah, yeah. And I think we can generally bucket them into one of three buckets. It’s either rule changes for how PJM is running its market — things like price caps, what we sometimes refer to with the reference unit, the shape of the demand curve. All these sort of, you know, really weedy changes to the way the auction structure is set up. So that’s one bucket of changes that they’re making.

The second is on identifying new sources of capacity that can come into the market. So you mentioned RMR, reliability must run units. These are generation units that are on their last legs. They’re getting ready to retire. And PJM says, “No, you can’t retire yet, because we need you for system reliability,” usually on the transmission grid.

Stone: If you go away, we don’t have anything to replace you right now.

Silverman: Exactly. And usually the solution is to build a new transmission line that will alleviate the constraint that that retiring power plant was addressing. Now, traditionally, those reliability must run units were excluded from the capacity market. One of the changes that PJM has done is to come in and say, “No, we’re going to count those going forward.” Another one — again, trying to bring in that extra amount of supply — they are going to make all generating units bid into the capacity market. Whereas today, a lot of particularly renewable generators, are not required to bid into the market. PJM has required them to bid into the market, which would add a couple 1000 megawatts of additional supply.

You also mentioned the surplus interconnection. Which is — you know, there’s not very much free lunch in the energy industry. Surplus interconnection is one of those free lunch kind of proposals where you basically take an existing power plant and you say, “Okay. Put a battery behind the meter of your existing power plant.” Or add solar to a gas-fired power plant. Basically using the existing interconnection and just putting more capacity behind that meter. So that is a real common sense proposal that just lets stuff get deployed very quickly because it doesn’t have to go through the interconnection queue.

So those are sort of two of the three changes. And then the last one is really interesting, where PJM has kind of, you know, “Break glass in case of emergency.” PJM has said, “We think we are in such a dire supply situation,” they’re going to invite a certain number of generators to come in and bypass the constraints in the interconnection queue and move directly into construction. So the idea is that these are shovel-ready projects that PJM will give a one-time pass to let them come in a lot more quickly than they otherwise would.

Stone: There’s a lot of resources waiting in line, and you’re saying to these resources, “Hey, you guys can essentially jump into the queue, jump ahead of guys, people, resources that have been waiting for several years. You’re going to jump ahead and get ahead of them.” Right?

Silverman: That’s right, exactly. And not only that, but PJM is going to select the resources that can jump ahead based on their size and their accreditation value. Their ELCC-weighted accreditation value. So in practice, what type of generator is both big and has a relatively high capacity accreditation? It’s largely natural gas. And this has, you know, sent some real shock waves through the sector. Because all of a sudden you have PJM basically saying, “Well, we are going to take special emergency action to bring to market new natural gas fire generators.”

And, you know, that’s a very controversial thing to do, and it leaves a lot of regulators and sort of grid watchers very conflicted. I’m conflicted on this one, because at one level, you really want PJM doing whatever it has to do to break the log jam in the interconnection queue. And yeah, it does seem like we’re headed towards a supply crunch. But at the same time, there’s something not quite right about cheating and cutting in line. You know, my 7-year-old daughter would be very quick to call someone out who cut in the lunch line at school. And, you know, you can’t help but think that PJM is sort of allowing people to cut in line.

But of course, if it’s the choice between that and an unacceptable reliability risk, you almost have to grin and bear it. And we will see very shortly whether the five FERC commissioners are willing to grin and bear it, or whether they’re going to say, “Nope, the rules are the rules.” And we’ll have to see — you know, we’ll know in the next month or two the fate of all of these various proposals.

So in addition to all the things that PJM has proposed, you also now have Governor Shapiro’s complaint that was just filed right before the new year asking FERC to take emergency action to put a price cap on the upcoming auction. And you know, a lot of Governor Shapiro’s concerns really revolved around, why are we going to send these incredibly high prices, if there’s not enough supply in the market to respond?

Stone: You know, Abe, what you’ve been talking about here in this interconnection queue and the jumping of resources over resources that are already waiting for several years to get in — and these resources being fossil fuel, primarily, it looks like a lot of natural gas — really points to another big problem that we haven’t talked so much about. And that is the lack of available transmission. And correct me if I’m oversimplifying here. But if transmission had been proactively built over the years so that it would be possible to quickly bring these new wind and solar resources that are backlogged in the interconnection queue onto the system, we might be able to bring in enough of those resources to balance themselves, to bring on battery storage, where the need to potentially go back to fossil resources wouldn’t be so great. I mean, there’s a planning fault here that somebody is responsible for. Comment on that, and am I on the right track?

Silverman: Oh, yes, you are speaking my language. A lot of my work is actually in trying to get states and utilities and regional transmission organizations like PJM to proactively plan out the transmission grid. The challenge in processing interconnection requests is really just a lack of sufficient transmission headroom. You know, the challenges associated with connecting data centers and other new loads to the grid is also really indicative of a lack of transmission. And so, you know, we have had 20 years where we really — particularly in the PJM region — have not identified proactive transmission upgrades that would just ease a lot of these problems and make the whole grid work better.

But this is hard, right? Because we are where we are today. And it’s very difficult when you think about a transmission reform put into place today. Last year, FERC issued its landmark order number 1920, which was all about long term transmission planning. You know, it’s a great order. And it really sets us on a stage to much better long term transmission planning. But we’re not going to see the benefits of that until 2030 at the earliest. Because it just takes that long to get a transmission line built.

But yes, I totally agree with you that had we been doing long term transmission planning all along, we would be in a much better situation today. But it’s also important to recognize, you know, the things like load growth are just adding fuel to this particular fire. And you know, where you have more new data center load coming online in the next couple of years than there is excess generation in the market to supply it, you very quickly run into this sort of question of, what’s it going to do to customer affordability? And how do we think about reliability in this sort of interim period between now and the halcyon future where transmission planning is working, where we have a forward capacity market again, and where the generation queue is unstuck. And I think the real crunch time is between now and 2028, 2029, where you have all these factors — all these confounding factors — coming in and really stressing the grid.

Stone: I want to bring up another point, just to build on what you’ve said. And I think we’re getting close to the end of the conversation here. But I want to bring this up. We obviously have this perfect storm of load growth and policies driving changes in the generation mix, and some misfortune and missteps at the PJM level as well as at the regulatory level — at the FERC level — that has gotten us into this tight situation. But it’s interesting. The CEO of First Energy, which is a utility that operates within the PJM market, has stated, and he stated recently, that he doesn’t think PJM’s capacity market will entice enough new investment into power plants in the market. That’s a quote or a paraphrase. And he suggests that a return to the old regulated rate of return model, the pre-restructuring model for this industry, is needed. My question to you — and this is a big and loaded question — are competitive markets up to ensuring reliability? Or do we need something like the centralized planning that existed when you had one utility that controlled its whole footprint and decided what was built and what wasn’t? Are markets up to this task?

Silverman: Yes. I think they are. But we have to take these kinds of critiques seriously, right? It’s important, you know, when you when you think about questions about restructuring and undoing regulation and putting monopolies back in charge of the generation supply, we need to remember, why did we enact competition to begin with? It’s because that system really did not work very well. You had very high costs, you had delays. You did not have the kind of innovation that we have seen come into the PJM market, and really the markets nationally, over the last few years. So, you know, I say, okay. If we had the utilities in charge, would we have two terawatts of new clean generation currently pending in the interconnection queues? I don’t think you would.

Because the utilities, you know, simply just don’t innovate in that kind of way. And I think it’s a little bit self serving for a utility to say, “Well, we really think we need to do away with competitive markets, because we would like to get back into the business of owning generation, and increase our rate base and increase our earnings.” Now that said, I mean, you do have to talk very seriously about, what do we do about these markets? You know. And if they’re not working, you know, I don’t think the answer is to stop innovating in our market design.

So I personally have spent a lot of time in my career thinking about, how do we make these markets better? And I think it’s a two-phase challenge. One is, you just need to execute on things like the interconnection queue and running your markets on time. I mean, frankly, you know, it’s — I’m very sympathetic to the PJM and the PJM staff who work incredibly hard to get this stuff done. But the last couple of years, they simply haven’t executed on some really core functions. And so, you know, part of it is, it’s not clear to me that a monopoly utility would have done any better job working within the constraints that we have in terms of supply chain. And, you know, what happened after the pandemic, and Ukraine, and everything else. So I think we need to talk about that. But then we also need to think about, all right. How do we make this better? And that’s where, you know, you need to look at alternative market structures and think about, how do we get the capacity markets and the energy markets to be buying the types of resources and the quantity that we want necessary to meet state’s policies, public policy, to meet the Clean Energy Challenge?

And that probably requires a couple of things. It requires executing on the things we’ve already talked about, like the queue. But it also means that we have to sit down and have a very serious conversation with states and consumers and large buyers of electricity and say, “All right. What do we want this long term supply mix to look like? And then how do we arrange our markets so that we are buying more of the types of resources we want and less of the types of resources we don’t want, without endangering reliability?”

Stone: Abe, a final quick question for you here. So, should we be really worried, those of us that live within the PJM footprint, that we’re going to have some outages in the coming years?

Silverman: I can’t say no, categorically. But I will say I think it’s far more likely that a tree is going to fall on the local distribution line leading to your house than that we’re going to have a catastrophic failure of the PJM grid. That said, certainly we need to take these kinds of threats to bulk system reliability seriously. And we really need to just execute on the simple stuff. And that is the best defense we have against the kind of reliability issues you’re talking about.

Stone: Abe, thanks very much for talking.

Silverman: My pleasure. Thanks for having me. It was a lot of fun.