Is a New Era of Electricity Prices Beginning?

Electricity prices have become a major political issue in the United States, with policymakers increasingly focused on rising utility bills and the costs of meeting growing electricity demand. At the same time, renewable energy has often been blamed for driving prices higher.

But what does the data actually show?

Ryan Hledik of The Brattle Group discusses research conducted with Lawrence Berkeley National Laboratory on U.S. electricity price trends. The research finds that, nationally, electricity prices have largely tracked inflation, though significant regional differences tell a more complicated story. Hledik explains the factors that really drive electricity prices, the role of renewable energy, natural gas, and infrastructure investment, and why electricity costs vary so dramatically across the country.

Hledik also explores whether 2025, when electricity prices rose faster than inflation nationally, marks the beginning of a new era of rising electricity prices, or a temporary departure from a longer-term trend.

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.

The price of electricity has become a major political issue in the United States. Last fall, gubernatorial candidates in Virginia, New Jersey, and other states campaigned on promises to address rising electricity costs, reflecting voter concern over power bills. At the national level, recent announcements such as the Trump Administration’s rate payer protection pledge have sought to insulate consumers from the costs of powering AI data centers. Separately, the administration has argued that renewable energy has driven electricity prices upward.

Yet recent research provides a much more nuanced view of the trajectory of electricity pricing in the US, and the range of factors that influence prices. Notable, the research finds that, on average, electricity prices have grown in line with inflation over the past half decade, though significant regional variation complicates efforts to identify any single price driver. And in 2025, electricity prices outpaced inflation nationally, raising the question of whether the year was an outlier, or an indication of faster price growth to come.

On today’s podcast, we’ll be talking with one of the people involved in that research. Ryan Hledik is an alumni policy advisor with the Kleinman Center, and a principal with the Brattle Group, an electricity market consultancy that, together with the Lawrence Berkeley National Lab, has examined recent US retail electricity price trends. On the podcast, we’ll look at the insights that the research provides, the role renewable energy and demand growth may or may not be playing in electricity prices, and explore whether the forces that have moderated prices in recent years may be losing their grip. Ryan, welcome back to the podcast.

Ryan Hledik: Yeah. Andy, very happy to be here. Thanks for having me on again.

Stone: We were on, what, two or three years ago, talking about virtual power plants, which remain an issue.

Hledik: That’s right. Yeah, no, always excited to have the opportunity to talk with you about these timely issues.

Stone: Okay. So, let’s jump into the conversation today. I want to start with this recent research, which actually extends back to 2025. What was the motivation behind this joint effort between Brattle and Lawrence Berkeley, and what role did you play in the research?

Hledik: Sure. Well, we started this research right at a time when I think affordability and concern about rising electricity prices was starting to become an issue. It hadn’t quite yet become the headline, election-defining issue that we have seen it become more recently. But we were starting to get more questions about electricity rates, and why are they rising, and why is this happening? And so Lawrence Berkeley National Lab took the lead on this study. A researcher there, Ryan Reiser, was the one who led the charge. But they invited us at Brattle to come in and be a partner on it, and help them with some of the research. So, you know, we were very happy to have that opportunity to partner with them on this. And I think it led to some really interesting outcomes, in terms of the complexity of this issue.

Stone: So one of the most surprising findings from the report, again dating back to the original report which was last fall, is that national electricity prices have generally tracked inflation over the past half-decade or so. Actually, the report goes back to 2019, which is when you start that look. And I wonder if you could explain that finding, and how it may or may not align with this popular view that seems to be running right now that electricity prices really have been surging.

Hledik: You know, I think it’s a nuanced question. I think one of the interesting takeaways of the analysis was the conclusion that electricity prices have been rising in nominal terms, but they haven’t been rising faster than other stuff, the cost of other household goods. So that was, I think, important context for the conclusion of our paper.

At the same time, it is important to acknowledge that while electricity prices haven’t risen faster than the cost of other goods, the cost of other goods has been rising at a pretty rapid rate over the five- or six-year period that we analyzed. So while we do think it’s important context to understand that electricity is not an outlier in terms of inflation, it’s also important to understand that it is something that has become more expensive for the average American household. And that is partly why we’re seeing so much attention being paid to this issue right now.

Stone: I wonder if you could take it back a little bit further. So nominally, prices have been in line with inflation. But if we look prior to 2019, has the last six- or seven-year period been different from what we’d seen before in terms of the increase in electricity prices?

Hledik: Looking further back than our study period, I believe what we would have seen there is actually that electricity has risen at a lower rate than inflation. So, has become less expensive relative to other goods. More recently, the trend has been electricity prices rising with inflation. And then in just the past year, we’ve seen electricity prices start to rise more quickly than other goods. So it’s something to keep an eye on. Obviously, the conclusion you draw depends on the timeframe that you’re using to analyze the price trend. But it is definitely something that has been changing over time.

Stone: Could that tie into the fact that over the last half decade or so, we have started to see some demand growth that had not been there for the decade prior to that? Is there a tie-in with that, do you think?

Hledik: That will be a very relevant factor as we move forward. One of the really interesting takeaways from our analysis was, when we looked back over this five- or six-year period, starting in 2019, we actually saw that the states that experienced the most load growth over that period were also the states that experienced the most significant declines in the inflation-adjusted electricity prices. So that was a really interesting takeaway, given a lot of the concern right now that has emerged around data centers and the impacts that they could have on retail rates.

Stone: Well, that’s an interesting point because if we look at the basic law of supply and demand, more demand, higher prices. But you’re seeing the opposite, at least to some degree. Can you tell us a little bit more of what’s behind that?

Hledik: Sure. Yeah. When we looked back over this historical period, what we saw was those states that were experiencing that load growth— I think North Dakota was the state that experienced the most load growth over that period— those states had a lot of spare capacity on their system to accommodate that load growth without needing to go out and spend a lot of money on a lot of new infrastructure to serve that load. And when you’re in that situation, what that allows you to do is take the fixed cost of the system that you’ve already invested in and spread that out across more customers or more megawatt hours. And the result of that is, you’re sharing in the cost of the grid with more customers, and that puts downward pressure on rates for everyone.

Stone: So we’re not seeing dramatic build out of the grid to match that new load, but we’re basically absorbing that new load into a system that already exists. And the costs are spread more evenly or more broadly.

Hledik: That was the case historically. Yeah. And now the question as we look ahead, as a lot of that spare capacity has been used, is there a way for us to continue to replicate those conditions going forward?

Stone: Okay, so we’re going to get to that point in just a moment because that’s a really critical potential inflection point here. But I want to jump to something else here for just a moment. A common narrative that we have also seen in recent years is that renewable energy has been a major driver of rising electricity prices. And your research concluded that some of the states with the highest electricity price increases had renewable portfolio standards to support clean energy, and also had policies to support rooftop solar, such as net metering policies. What does the evidence actually show about the impacts of utility scale renewables as well as smaller rooftop solar, and the policies that support them on electricity prices?

Hledik: Yeah. You set that up really well. Because what we found in the analysis is there are three different answers to that question around the impact of renewables on electricity prices. The first takeaway that we had was wind and solar are a pretty inexpensive form of energy. So the states that have access to very rich sources of wind and solar energy, and are going out and are making competitive market-based procurements of that energy, did not see rates going up. And if anything, we identified some modest downward pressure that resulted from the fact that those states have access to a more diverse, more cost-competitive resource mix. So that, broadly, I think in my view, made it clear that it is not just the case that any state that’s dependent on renewables is going to see higher electricity prices.

The second part of the question is what happens in states that have the types of renewable portfolio standards, renewable policies that you are describing, where the utilities are being asked to go out and procure renewables above and beyond what would just happen on a strictly market procured basis? In those states, we did identify on average that the result of those policies could put some upward pressure on electricity prices. It varied across the states that we looked at. On average, the impact was pretty modest. It was about a quarter of a cent per kilowatt hour. And in the states that we identified as having the most extreme impact, it could range upward to about a cent per kilowatt hour. And that’s, I think, a pretty logical conclusion of our finding. If you’re asking companies to go out and buy something that isn’t otherwise supported by the market and pay a premium for that, there will be a price tag associated with that. But I think in those cases, it’s also the fact that policymakers and decisionmakers are going into those policies understanding that that’s a likely outcome, and concluding that the cost of paying for out-of-market renewables is more than offset by the avoided costs associated with climate change and some of the associated social impacts associated with that.

So that was, to me, an interesting conclusion. There has been some other research that’s been conducted on the topic by some academics at MIT that did not identify a relationship between renewable portfolio standards and energy prices. So this is definitely, I think, an area that continues to be fertile ground for further research. But just interesting to note some of the different outcomes that research has found on that topic.

Stone: Yeah. I’m glad you brought up that MIT research. That was from MIT Center for Energy and Environmental Policy Research. And they did point out a couple of things. The title of that research is basically saying, “Don’t confuse correlation with causation,” in those states. And they also say, quote — and this is at the top of their paper, they say, “We find that RPSs and utility scale renewables are robustly correlated with lower residential electricity prices.” So again, a different perspective here, right?

Hledik: Half of a different perspective. I think both our studies and theirs agreed that again, the market-based renewables procurements can put downward pressure on rates. Where our conclusions differed was around the renewable portfolio standards.

And I think that’s largely attributable to different methodologies that were used. They used econometric analysis, statistically analyzed the relationship between renewable portfolio standards and prices. Our analysis was largely based on studies that had been conducted in the states with those policies, and their estimates of the incremental cost associated with those policies. Some of that can be approximated through the price of renewable energy credits, right? That tends to reflect the premium associated with out-of-market renewables purchases. And so some of our analysis tied back to that fundamental research.

Stone: So on the residential solar component, the solar metering, you also saw some rate increases. But that was a rate design issue more than the technology issue. I wonder if you could explain what’s going on there.

Hledik: Sure. What’s happening there is essentially the opposite of the historical data center scenario that we were discussing a minute ago. You know, when you have net energy metering policies in place, and residential customers are paying a largely flat volumetric cents-per-kilowatt-hour rate for their electricity, when they install rooftop solar, they’re able to reduce their bill nearly to zero, essentially, over the course of a month, when you pay a fixed charge. And as a result of that, you’re spreading the fixed cost of the grid across fewer kilowatt hours. And so customers who don’t have rooftop solar can end up seeing their rates increase to offset the loss of revenue from those other customers. Who are still, by the way, using the grid, even though they’re not paying for it.

So we have seen that having a positive or upward impact on rates. It tends to not be a very significant impact in most states, just based on the level of rooftop solar penetration that they’ve seen. Where the impact was material was limited to a couple of states where that rooftop solar penetration has been higher. And those states have started to reform their net energy metering policies to try and address this issue.

Stone: So if renewables are not a primary explanation for rising electricity rates, what are the major factors that have historically driven electricity prices in the United States?

Hledik: Yeah. That’s the big question. And to try and answer concisely, I think I would say that the category of drivers that have had the biggest impact on rates are, over the historical period that we analyzed, investments that we have needed to make in the grid that weren’t necessarily related to load growth. So, you know, there are portions of the distribution system— kind of the wires that that take that take electricity from those big substations that you see when you’re driving down the interstate down to your home— there are portions of that grid that are over 80 years old at this point. So utilities are just needing to go out and make investments just to replace aging infrastructure and maintain the same level of reliability that we’ve had for decades.

Unfortunately, that need to make those investments has happened right at a time when the cost of that power systems equipment has gotten a lot more expensive. There are still some supply chain constraints coming out of the pandemic that haven’t been totally resolved. And so we have seen the cost of some of that equipment rising at a much faster rate than inflation over the last four or five years.

So it’s unfortunate. It’s kind of a perfect storm situation. We’re needing to make those investments right at a time when stuff’s gotten more expensive. And that is kind of the, I think, underappreciated driver of a lot of the price increases that we’ve seen over the historical period that we analyzed.

Stone: And as you mentioned in the report, there is a lot of regional variability, which we’re going to get to in just a moment. But before we do that, I want to ask you another related question. The reports briefly make note of price stabilizers that have allowed pricing to stay relatively stable over the last half decade plus. And question is, are those stabilizers now weakening? But I want to ask you, what are those stabilizers that have generally kept pricing at least in line with inflation?

Hledik: You know, I think there are a couple answers to that. One is, even in the states— going back to the North Dakota example— in those states that have experienced load growth over a historical period, that load growth has been fairly predictable and at an anticipated rate. So the big difference I think that we’re seeing now, looking forward, compared to the last five or six years, is load growth is coming online at a rate that we could plan for and at a rate that wouldn’t drive costs. And so you’re adding kilowatt hours to the system. You’re also, depending on where you are, needing to make some investments to accommodate that, but it’s happening at a rate that you can plan for. And that alone has allowed electricity prices to be fairly stable over this period.

Another point, again, going back to that North Dakota example, is, we haven’t been in a situation where our grid has been very capacity constrained. And so when you aren’t facing those types of constraints, prices aren’t spiking to some of the extremes that we’ve started to see more recently in certain markets.

Stone: Right. It’s interesting here in the PGM region where I live. For a long time, PGM was criticized for having too much capacity, right? So it could absorb all that new load. And that has come to an end. So maybe that’s part of that price stabilization period that’s ended.

Hledik: Absolutely.

Stone: Okay. So we did mention earlier that the price of electricity has generally, across the country, tracked inflation, but there are large regional variations. As you point out, California’s electricity prices have increased at twice the national average in recent years. That’s a prime example of, again, this regional variation. What are the drivers in California that are primarily behind this more dramatic rise in electricity prices over the last few years?

Hledik: The short answer is wildfire risk mitigation costs. You know, I think California and a few other regions both have needed to spend money preemptively to address the risk of wildfires or extreme weather or natural disasters, and also have had to spend money to repair their power grids after there has been a wildfire, after a hurricane has blown through, or a winter storm. And so that that combination of proactive and reactive investment in grid hardening has been a big driver in California, definitely the number one explanatory driver of the price increases that we’ve seen in that state, and has also contributed to some of the price impacts that we’ve seen in some Gulf states and East Coast states as well.

Stone: I want to point out here that this is what makes this conversation so difficult to boil down. What are the drivers of electricity price increases? Because they are so regional and regionally different. So California prices have been up. We’ve also seen, in the mid-Atlantic and up into New England, higher than average price increases as well. What’s going on there?

Hledik: There are a couple factors. One is, those regions, like much of the U.S., still tend to be fairly dependent on natural gas as a source of fuel for their electricity generation. And natural gas prices are variable. And when natural gas prices go up, those are going to flow through to your electricity rate because a lot of the generators in your region on the East Coast are paying for natural gas in order to produce electrons.

The natural gas story has been more one of volatility than it has necessarily been one of a consistent upward trend in prices. So that depends partly on what historical period you’re analyzing. But in general, especially looking back just over the last year for which there was available data, from 2024 to 2025, what we did see there was an increase in natural gas prices and that being probably the biggest driver of increases in electricity prices over that last 12-month period.

So that’s one. And then I think another factor, like you said in PJM specifically, is the sudden increase in capacity prices associated with unexpected load growth.

Stone: And let’s go ahead and jump to that capacity price issue for just a moment. Capacity prices in PJM have gone up dramatically over the last couple of years, on the order of 12 to 14 billion dollars per year. And they’ve only been capped or limited because the governor of Pennsylvania, Josh Shapiro and some other governors got together and basically negotiated agreement with PJM to cap those capacity prices. So capacity prices are a really big issue. And they are definitely painful to places in PJM such as BGE, Baltimore Gas and Electric Zone, which has had some constraints.

But it’s very interesting. The market monitor for PJM recently pointed out that despite the dramatic increase in capacity prices costs in PJM, capacity is still a relatively small portion of the overall bill. So I wonder if you could put that a little bit more into context for us.

Hledik: Sure. Yeah. I mean, that’s another good example of how nuanced and complex this is. It’s so hard to distill any of this down to just like an eight-word headline.

Stone: And capacity prices also can be very political because they are a response to grid operator policies and things like that.

Hledik: Exactly. Yeah. But you are right. I think when I was looking back— and again, it depends on where you are in PJM— but capacity prices could be five times or six times higher than what they were a couple of years ago. So capacity prices have shot up to that cap.

But the point about customer bills is right. Capacity prices probably only account for 10 to 20 percent of a household electricity bill in PJM. So it could be the case that the capacity price is increased by a factor of five, and your electricity bill has maybe only increased by 10 percent. And it’s unfair to say “only”, because a 10 percent increase in an electricity bill is noticeable. The point is it’s not a 5x increase in your electricity bill, right? The majority of your bill is to pay for the energy that’s being generated that you’re consuming, and it’s to pay for the poles and wires that are delivering that electricity. So again, it’s just important context to understand like when you hear about a 5x increase in a capacity price, to really understand what that translates into in terms of an actual bill impact for an average household.

Stone: It does seem that that focus on capacity pricing— again, not to underplay the importance in the role that it plays. But it can make it easy to overlook some of the bigger energy-related cost impacts that are there, right?

Hledik: It can. And I think the other point here is this is kind of the market doing what it’s supposed to do. Which is, when you enter into a situation where demand is outpacing supply, prices go up to attract more supply to the market. That’s just kind of the fundamental principle of markets. I think there are some challenges in PJM with this level of load growth being so sudden and unexpected. I don’t know that the market was just designed to accommodate that.

Stone: And to make clear, PJM is ground zero for the AI data center build out in Virginia and other areas, right?

Hledik: It is. Yeah. One of them.

Stone: Yeah, one of them.

Hledik: So I think there are still some market design issues that need to be worked through, given how severe the growth and load has been and how unprepared we were for it. But it is also just sort of important to appreciate that basic point that as supply and demand conditions tighten in the market, generally prices go up. That is what happens.

Stone: I want to point out one other issue here is that the electricity prices have tended to be highest for residential customers, higher than commercial and industrial customers. I just want to make sure we understand why that is.

Hledik: Part of the reason for that is, as residential customers, we’re connected all the way down at the very edge of the grid, right? So we need to pay for all the generation that’s being used to generate the electrons that we’re consuming, all of the transmission lines that you see as you’re driving down the interstate that deliver that move those electrons over long distances. And then we also need to pay for the distribution system, which is what takes those electrons the last mile all the way down to our homes.

A lot of larger customers don’t actually need to connect to the distribution system. They can plug directly into the transmission system. That’s what the data centers are doing. And they can connect at higher voltages. So as residential customers on a per kilowatt hour basis, we are using more of the grid than some of the other larger customers. And so we are needing to pay for that full use of the grid. That is partly what is driving rates higher for residential consumers.

From state to state, you know, politics also play into electricity rates and how price levels are set. And so another question is, how well are the residential consumers groups advocating for them in those rate cases versus larger customers? There’s a political element to that as well. But fundamentally, on a per kilowatt hour basis, the residential sector tends to be more expensive to serve than a larger customer, which is more efficient to serve.

Stone: Okay, so let’s go ahead and jump to 2025, right? So the initial research that you all did that came out last year looked at 2019 through 2024. Again, saw rates rising in line with inflation. You came out with additional research early this spring that looked at 2025 and found that now we see electricity prices actually outstripping the inflation rate. So the question here becomes, is this a one-off or is this a sign of things to come? But tell us what happened in 2025. How dramatically did the rates go up? And what do you think is behind that?

Hledik: Yeah, so we did observe a bump up in rates, like you said, in both nominal and inflation- adjusted terms. The most widely cited cause of that rate increase that we saw was an increase in natural gas prices. Natural gas prices were very low in 2024 compared to historical period. They were higher in 2025. And so again, because close to half of the US generation fleet runs on natural gas, that makes its way into electricity rates. That was a number one driver. And then another part of that is the PJM story that we were just discussing, and the fact that we are starting to see some wholesale markets like PJM reacting to the sudden emergence of load growth with increases in capacity prices in particular.

We also saw that the observation about distribution system investment rising over that earlier historical period, that is persisting as well. And there are a lot of utilities that are coming in, asking for approval to continue to make significant investments in their grid partly for that reason. Not just load growth, but because they need to maintain the system in order to keep it reliable.

Stone: That’s an interesting point. Because last year was a record year for utility proposals to increase electricity rates. I think there were $18 billion in rate increases that were proposed. I’m not sure how much of those actually got approved by state PUCs. But it sounds like the distribution build out is, again, a major issue there.

Hledik: It is. And I think unfortunately for us as consumers, that’s not one that I see necessarily fading away anytime soon. I think there will continue to need to be an investment in the distribution system to maintain the level of service that we’re used to. And that is, to an extent, kind of the new normal that we’re facing in terms of electricity prices.

Stone: So let’s jump to one of the gorillas in the room here, or maybe the biggest gorilla in the room, which is the AI data center build out, right? So I want to ask you, we talked about PJM and the capacity price impact in PJM. But generally to this point, how much of the electricity price increases that we’ve seen — I guess, particularly in 2025 — might also be attributed to AI data centers? Or is that still around the corner in terms of actual bill impacts?

Hledik: Very little. If we’re going to see material bill impacts from data centers, that’s something that we’ll see in the future, not something that we would have seen over the historical period that we analyzed.

Stone: And again, I guess the question here comes to, where are those costs apparent? So we have the rate payer protection pledge from the White House earlier this year. Other policies around that, such as large load tariffs, which are seeking to isolate consumers from the price impacts. Tell us a little bit about those efforts.

Hledik: Definitely. You know, I think another important thing to understand here is your electricity rate, you can think of it as a really simple math problem. It’s total costs that are being spent to provide electricity and deliver it, divided by the kilowatt hours of electricity that are being sold, right? The numerator, the top part of that fraction, is growing. We are going to need to make significant investments in the power system in order to accommodate a bunch of new load growth. But the bottom part of that fraction, the denominator, is also growing. Data centers use a lot of energy. They use it essentially around the clock. And so they are growing the sales base while they are increasing the need for investment.

And so whether our rates go up will depend on whether the revenue that those new data centers are paying into the system exceeds or doesn’t exceed the new costs that they’re imposing on the system. So a lot of the work that’s being done around large load tariffs right now is to make sure that these new large customers are paying for the incremental costs that they’re introducing to a power system. And as long as that’s happening, customers should be insulated from the impact, and potentially even benefit if those revenues exceed the costs.

It’s a very simple way of putting it. It is not easy to precisely identify all of the incremental costs that are being incurred to serve a new gigawatt data center, especially when some of the investments that are being made to serve that data center are upstream investments in the transmission system that will benefit that data center, but also will benefit other customers well in terms of reduced congestion and the ability to better integrate renewables into the system and that sort of thing. So it is a complex issue to solve. But it is, I think, addressable problem for consumers.

Stone: That’s an interesting point. Because as we look to upgrade the grid to accommodate data centers as well as just general load growth, and to make the grid more reliable, the question— as you just basically said— becomes how do you allocate those costs amongst the biggest drivers of needs for upgrade to the grid, which are the data centers and everybody else also who’s using the grid, right?

Hledik: Yeah. And I think, you know, a potential opportunity here is, we’ve spent a lot of time talking about how the grid is getting old. It needs to be upgraded. We have hyper scalers that are coming in looking to spend billions of dollars to upgrade the grid so that they can connect their data centers to the system. And so if some of those upgrades and that grid modernization can benefit consumers as well, there is an opportunity for this to be a win for everyone. We just need to make sure that we’re getting the prices right and setting up those large load tariffs and contracts in a way that ensures that outcome.

Stone: I would imagine one place where there could be upward pressure, though, is as you have more entities looking for equipment due to demand for that equipment, that equipment price is going to go up for everyone. Regardless of what are those new, say, generators or transmission upgrades, are really to serve AI data centers or the grid generally.

Hledik: Definitely, yeah. And we’ve been colloquially referring to that as “an indirect price effect.” Again, one of those effects that’s harder to quantify and put into an electricity rate. But there, we have seen some interesting policies. I think we’ve seen this in Minnesota and then also in Pennsylvania, with PPL, where as part of the agreement for a new data center to connect to the grid, they also agree to pay into a fund that will benefit low-income customers. Either through energy efficiency measures, or just a fund that will improve affordability for lower-income customers.

That is a pretty explicit acknowledgement that they are paying above and beyond even the full incremental costs that they’re imposing on the power system. And that is a way that I see them being able to offset some of those indirect effects that are harder for us to measure and allocate to them.

Stone: So we see the need for new infrastructure to meet new load. We do see, as you mentioned,  that these pressures aren’t going to go away any time soon. We’re not going to suddenly see the price of new equipment fall by 50 percent. No fire sales. It’s just not going to happen.

Now it’s interesting, because in the research you point out a few strategies. I think there are three generally that, if implemented, could significantly help to moderate electricity prices going forward. Could you talk us through those? One of them was, basically, put this new load on spots on the grid where there’s plenty of capacity to handle it.

Hledik: Yep. I mean, that’s the easy one. But unfortunately, we’re running out of those locations that have spare capacity. So, yeah, I think to the extent that we can site data centers in locations that have spare capacity, that is definitely the low hanging fruit. And I think that is happening organically as the hyper scalers shop around and try and identify those opportunities.

The second opportunity is to ask the data centers to be flexible. That could mean flexing their own load or that could mean using their own on-site generation to reduce their impact on the grid during hours when it’s constrained. There’s a lot of excitement about that opportunity. Google has demonstrated through a number of arrangements with their utility partners that they can do that. There are still some question marks around the extent to which data centers are able to do that. We’re still answering that question.

And then there also is that the system operators are kind of catching up in terms of their willingness and ability to rely on that flexibility and say, “Okay, even though you’re a gigawatt data center, we’re going to connect you with only 800 megawatts of capacity because we believe that you’ll flex the other 200 megawatts when you need it.”

Stone: That’s a totally new paradigm for the utilities, by the way. They’re not used to operating like that.

Hledik: Exactly. So yeah, we’re still trying to figure out what the data centers can do, and we’re also still trying to bring the utilities and the system operators along at the same time. But huge opportunity there.

And then the third piece, which is what I get most excited about, is not even necessarily asking the data centers to be more flexible but asking other customers in that region or that state or that utility service territory to be more flexible. And not asking them to do it, but paying them to do it. And the most efficient outcome could be that Google comes into a service territory and says, “We are going to offer a heavily subsidized or even free battery to any customer in the service territory that wants one.” In return for that, the utility needs to be able to dispatch those batteries for grid services from time to time in order to provide the capacity that’s needed to accommodate this new data center.

Stone: Can I get that for my house? That sounds great.

Hledik: Yeah. My hand is raised. I will be first in line if that comes to Portland, where I live. But we are actually working with Google on that topic right now, exploring partnerships with a few different utilities that see this as an opportunity to build out their own demand side management portfolios, pay customers in the process, and then also get the data center load connected more quickly than it could be otherwise.

Stone: Are there specific areas of the country where that might be happening?

Hledik: The most recent example, which was just a few days ago, was, again Google and Voltus, a demand response aggregator, signing an agreement to bring 100 megawatts of new demand response capacity to the PJM market. So that’s one example. And we’ll be seeing that happen elsewhere as well.

Stone: And that’s interesting, too, because when you think about demand response you often think about commercial businesses, industrial businesses, that really can move the needle on this. But going back to our VPP discussion from a couple of years ago, if you can aggregate my house and your house and everybody else’s house is into one whole, then that becomes significant as well. And I guess that’s where the Google battery plan comes into play.

Hledik: Exactly. Yeah, no. And when we’ve looked— studies we’ve done in the past have identified about 200 gigawatts of untapped demand flexibility potential nationally. And that was even before we were talking about some of these new emerging models around distributed batteries. So there really is a lot of potential there.

Stone: And again the question is who’s going to pay for that? Maybe a Google. Something like that.

Hledik: Yeah, exactly.

Stone: So Ryan, I want to ask you a final question here. This whole question of electricity pricing could go a lot of different ways. And there are a lot of unknowns as we look into the future. What are the forces, either positively or negatively, that you see that could keep electricity prices in check or let them explode exponentially, I guess, in a worst-case scenario? What’s your thinking about that?

Hledik: Sure. Well, like you said, data centers and other large customers are the massive gorilla in the room right now. I think we have, from an electricity price impacts perspective, a huge opportunity there. If we can continue going forward to create those types of conditions that we observed in North Dakota historically, this is an opportunity to continue to put down more pressure on rates by asking those new large loads not only for the cost of the new infrastructure that’s being developed, but also to pay for the existing grid that they’re also sharing, and the use of it with other customers.

So I think the big opportunity we have in particular is oriented around that model that we were just describing, where a hyper scaler could come in and pay for flexibility. And by virtue of creating that flexibility on the system, be accommodated at a lower cost than we’ve seen even historically in terms of investment that was needed for the grid. If we can create those conditions, we are going to create a scenario where customers benefit.

My concern, where I see there being a risk is that’s new. Like we were just discussing, this is not — we’ve been doing demand response for decades. That model is new. And I am worried that it’s going to take years for us to work through some of the new regulatory questions that emerge around that. How much of that new flexibility is really incremental? How much of that can we attribute to the data center? What should those incentives be?

Those are all important questions. My hope is that those are questions that we can answer quickly, and not let the perfect become the enemy of the good, or else I think it’ll take us five years to answer those questions and by the time we’ve done that the ship will have sailed and we will have just spent all this money on traditional infrastructure and not taken advantage of this opportunity that we have right now to transform the demand side of the industry and actually make it as flexible as it could be.

Stone: Ryan, thanks for talking.

Hledik: Thank you very much.