What’s the best way to handle rising US electricity demand? Contrary to what some large utilities and regulators think, it’s not building new fossil gas plants. In this episode, Eric Gimon and Michelle Solomon, coauthors of a new report from policy shop Energy Innovation, make the case that utilities have more effective options to address both short- and long-term demand.
Text transcript:
David Roberts
For the last 20 years or so, electricity demand in the US has been relatively flat — grid planners have gotten used to load growth on the order of 0.5 percent a year — and that fact has deeply shaped utility incentives, policy approaches, and the electricity sector generally. But as the research firm Grid Strategies declared in a December 2023 report, “the era of flat power demand is over.”
Between 2022 and 2023, grid planners’ forecasts for electricity demand growth doubled — and they are still rising. The main culprits in the short term are new manufacturing and industrial facilities and, especially, data centers. The general wave of AI hype has inflated data center growth numbers flying everywhere. And soon enough that short-term growth will be swept up in longer-term growth from the electrification of the automobile fleet, home heating, and industry. There’s a whole lot of load growth coming down the pike.
This sudden paradigm shift has many utilities (particularly in the southeast and midwest, like Georgia Power and Duke Energy) and their state regulators spooking and reaching for what's familiar, i.e. new fossil gas power plants and/or delaying the retirement of coal plants.
But are gas plants the best way to deal with this new demand? A recent report from the policy shop Energy Innovation argues that they are in fact the worst — that there are other short- and long-term solutions that should come first. This is rapidly becoming the central question of the energy transition, so I'm eager to talk to two of the authors of that report, Eric Gimon and Michelle Solomon, about the other tools utilities can bring to bear on short-term demand growth and how the system needs to evolve to cope with long-term demand growth.
Without further ado, Eric Gimon and Michelle Solomon, welcome to Volts. Thanks so much for coming.
Eric Gimon
Thanks, Dave.
Michelle Solomon
Yeah, thank you so much for having us.
David Roberts
This is, to me, the kind of white-hot center of the energy transition right now, which is how do we deal with load growth that's back both short-term and long-term? So, Michelle, I want to start with you. Maybe you can just talk a little bit about these forecasts. I will say, just as an outsider, that when utilities say, "Oh my God, all of a sudden we've discovered this massive new demand, it's upon us. Before we have a chance to do anything, we got to have some gas plants." I get suspicious. So, can you explain sort of what's happening with these forecasts?
And why does it seem like this has so completely taken everyone by surprise?
Michelle Solomon
I think one thing that's really important to kind of get straight is that when we have these reports coming out saying that, you know, the load forecasts are doubling, as you mentioned, for a while, the growth in electricity demand was flat. So a doubling over something flat may not be as big as everybody is thinking. So, I think that's one important thing to start with. And then, yeah, just again, we've been out of practice with growing. We haven't been prepared to build new infrastructure. And so, we're kind of shifting into this new phase. And honestly, we're going to need to have a lot more demand growth than what we're seeing with these forecasts in order to decarbonize our economy.
So, I think these are some first initial growing pains, and it's a little bit of a litmus test for who is really going to be serious about meeting climate goals and reducing carbon emissions as we go forward.
David Roberts
Yeah, I mean, what I want to say is, if you're panicking about this, just wait five or ten years, you're going to see some serious load growth. And I'll say the growth of data centers is relatively fast and relatively sudden, but people have been forecasting this coming load growth. I mean, we've been talking about electrifying everything for years now. It's not a great mystery that this was going to happen. I just feel like utilities should have been preparing already and it's a little rich now that they're acting like they're being caught flat-footed.
Michelle Solomon
Yeah, I think it's also important to think a little bit about the specific timing right now. We do have new EPA rules coming out in the coming weeks that will regulate new gas plants. And so, many utilities have expressed that they are concerned about these rules and kind of spinning up hype right now, right before the rules come out. You know, it's likely not a coincidence.
David Roberts
Interesting. And one other thing about the growth, which I don't feel like I have a good sense of, is how localized is it? I mean, we're sort of like, I feel like we default to talking about this as kind of a vague, abstract national problem, but these data centers are not being evenly distributed over the country. So, how localized is this growth?
Michelle Solomon
Yeah, I think that's right. And I think Eric can speak a bit more to kind of where specifically the data centers are showing up. But one thing that we looked at in our report was just what does the NERC, which is the reliability kind of entity for North America, you know, where does NERC see this new load show up? And actually, the regions where it sees the most new load by percentage are not lining up with the regions that are kind of sounding the alarm about load growth and gas plants. So it's mostly the biggest regions by percentage are actually the Northwest and the Southwest.
And, kind of, some of these Southeast regions are a little bit lower by percentage. Eric, I don't know if there's anything more you want to add about the kind of the data center specific hotspots.
Eric Gimon
Yeah, well, one thing to keep in mind about data centers, even though it's a lot of growth and AI is on people's minds, we've built more distributed PV generation, something like two or three times the current data center use. So just to keep that in perspective, it's something that's growing fast, but it's growing fast from a small single-digit number percentage of the total load. There is a significant impact on a kind of case-by-case basis, especially when data centers are kind of venue shopping for smaller co-ops and utilities that can give them a good deal, and that can be very difficult for those entities. I don't have a lot of sympathy for Duke Energy or Georgia Power that have been dragging their feet on a lot of the energy transition stuff and have frankly ridiculous things in their IRP.
We were just talking to one of our colleagues today about what's in the Duke IRP —
David Roberts
Just quickly, for listeners, tell us what an IRP is.
Eric Gimon
An IRP is an Integrated Resource Plan that was a wave of reform maybe 20 years ago to get utilities to motivate what they were asking commissions for. And so, they started doing these plans that are supposed to explain why they want the keys to the kingdom, as it were. And they've become better and better at kind of gaming the IRPs to get what they want.
David Roberts
I'm not an expert in this by any means, but my impression is that, especially from some utilities like Duke, the IRP feels a little bit sort of like reverse-engineered from what they want, rather than the other way around.
Eric Gimon
Yeah, I mean, for example, in the Duke IRP, there is a limited amount of solar allowed to come on because their transmission team has told them that "this many projects we can deal with in terms of interconnection." And then they're assuming that the interconnected projects are only of the 50 megawatts to 75 megawatts size, whereas there are much larger solar projects out there. That's just an example of the kind of obstacles. Now, it's not always just like a pure greedy capitalist pig kind of mindset that's driving this. You know, it could be that there's some planning engineers or a certain culture that really wants gas plants and is pulling the levers that way just because that's what they're comfortable with, as you said in the intro.
David Roberts
Right. Yeah. And it's a combination of, you know, comfort and also this speed required now. So it's one thing to tell them to do things they're not comfortable and familiar with. It's another thing to tell them to do those things really quickly or else face consequences.
Eric Gimon
Speed is a bit of a red herring, we feel.
David Roberts
Oh, yeah.
Eric Gimon
Gas plants don't get built that fast.
David Roberts
Right. This is another question I had about this. Like, if you got data centers coming on in a year or two, what's a gas plant going to do for that? It takes at least like two or three, five years. What's the —
Eric Gimon
Three to five years to permit and build?
David Roberts
Yeah.
Eric Gimon
Also, they're proposing combined-cycle gas plants.
David Roberts
Right. Not even peakers.
Eric Gimon
Not even peakers. Right. And the real needs are kind of peak needs from these new resources.
David Roberts
Yeah, it seems a little shady. So, I wanted to ask you, Eric, about — you in this report cite the EIA, the Energy Information Administration, the sort of national energy research body — its 2007 electricity demand forecast. You know, because everybody's talking about this 20-year plateau. We've been on this plateau in electricity demand, but that's not what EIA forecast 20 years ago. 20 years ago, it was forecasting electricity demand to rise, rise, rise. Like the line went up and right. And in reality, the line did not go up, it just went right. So, tell us a little bit about why those forecasts turned out to be wrong.
And it's funny, it's amusing, ironic, I don't even know what the word is, that those forecasts were also based on panic about data centers. So, tell us that story.
Eric Gimon
Well, it's an interesting story. It actually goes back to sometime around 2015, 2016. I was trying to understand claims that natural gas was the savior that had been reducing our greenhouse gas emissions. And so I looked at the inventory of how gas was replacing coal versus renewables replacing coal, you know, with a kind of better return on greenhouse gas emissions reductions. And I kind of had a paper going, and Amory Lovins came by the office and I told him about it. And he was like, "Well, what about energy efficiency?" I was like, "Oh, Amory, always on about energy efficiency."
David Roberts
Wait a minute. You're saying that Amory Lovins brought up energy efficiency to you? What?!
Eric Gimon
So, I went and looked, and it was clear back then that we were already seven, eight years into a flat forecast. It was a little hard to see exactly because we had gone through the great financial meltdown. So, maybe that had an effect. But that's when I started tracking down some of these sources about what codes and standards were doing and also the collected data from AEEE on energy efficiency in every state. And once you start kind of collating all these different efforts, you saw that that explained a pretty big chunk of the "missing demand." And in fact, energy efficiency had done a lot of the lion's share of reducing greenhouse gas emissions, if you think about the counterfactual being, whatever the EIA thought at the time.
So, when this came up and we were trying to push back on this narrative, it was fun to just go back to those same sources and update. And I don't think anybody was expecting a 20-year pause. And so, what's important to take from this isn't necessarily that, "Oh, this is all smoke and mirrors, you know, this load growth is never going to happen" because we don't really know. And as you said, if we want to electrify everything, we've got a lot of load growth ahead of us, right? I mean, we think maybe 3% a year would cover some of the more extreme cases for just electrifying industry.
So, we have to be ready for it. So the way we thought about this is a near-term set of recommendations, things that are faster than gas plants, and a long-term set of recommendations about getting ready for growth.
David Roberts
Here's a question. So, when you're talking about why that 2007 forecast proved incorrect, why demand ended up flat, you cite energy efficiency and you cite, you know, voluntary corporate contracts and other things, other factors that are still ongoing. So, one conclusion you might draw from that is just that these current forecasts will also prove incorrect. Energy efficiency will once again flatten demand and we just don't need to worry that much. Do you know what I mean? Like, because my sense, and just tell me if you think this is right. My sense on a kind of global level is just that demand forecasts in general tend to be overblown.
I feel like this has happened before in the US and will happen again. Like, we always are more efficient than we think we're going to be. So, why isn't the lesson here just like, "Let's all calm down?" Yeah, right.
Eric Gimon
Well, I think there's an element of "let's all calm down", like this isn't going to happen overnight. We don't need to build 10,000 gigawatts, you know, some crazy amount of stuff right away. There's two things that are different. One you alluded to, right: We want to electrify everything. So we've scheduled a lot of load growth, a lot of not business as usual. That's going to change the picture. We need to. There's a lot of energy that goes into vehicles, into heating. I have an electric car and I have a heat pump now at my house, and then my load is five times what it used to be.
We had to be ready for that. It's not all going to happen all at once. The other thing is what was driving down that load growth. One thing was energy efficiency measures. Well, we're starting to step off the gas on that. If you look at AEEE, their measure of spending is dropping off. So we need to get back on to energy efficiency. It's not something we should be forgetting about. It's something — it's a force multiplier, right, it makes everything else easier. And then the codes and standards that really drove things starting in 2007 were things that came out in EPAC, the Energy Policy Act of 2005, and there's been some updating of codes and standards and so on.
But, I think it's slowed down a bit. It's not my area of expertise, but I definitely think we could be doing a lot more there, especially on heat and cooling. Right. Every air conditioner could be a reverse heat pump and be much more efficient. So, we've stepped off a little bit on some of the things that have helped us in the past, and then we've got these new loads. At the same time, like you, when I talk to any veteran of the industry, they kind of roll their eyes at the forecast, so you never quite know where we're at, but it's good to be prepared.
David Roberts
Yeah, it's why I find this moment we're in so fascinating. Like, you could easily tell the story, like, "Here we come again with another round of inflated demand projections, and everyone's panicking," and, you know, like, I remember back in the seventies, they were going to build whatever, like two gazillion gigawatts of nuclear power, and that was all because of demand projections. And of course, like, it turned out to just be a tiny fraction of those. So you could just tell the story like, that's happening again, and then you could tell the opposite story, which is like, "No, no, it's totally different, there's a tidal wave of demand coming. You're crazy. Let's get ready. We need to panic and get started." And it's really, really hard to know where to come down in between those two positions.
Eric Gimon
It's a good way to frame it. I think maybe one way to move out of that frame is to think about what are the least-regrets things we can be doing.
David Roberts
Right, yes, exactly.
Eric Gimon
I mean, one thing we can't afford is for electricity prices to climb too fast so that electrification isn't viable for most people. And behind that, we need to be very careful about where we spend money. A lot of the assets that carry electricity around and transmission distribution are being used a lot less than they were 20-30 years ago and not as efficiently. And so one of the sets of recommendations, and Michelle can talk a lot more about that, is just getting more out of our existing grid.
David Roberts
Yeah, yeah, we're going to get to that in a minute. But, Michelle, I wanted to talk to you a little bit about the voluntary corporate piece of this. You know, I think people have been surprised, or at least I have been somewhat surprised over the past several decades at just what a powerful force for good corporates turned out to be. You know, as a raving socialist, I'm temperamentally inclined to not give corporates much credit for things, but, like, they've been really a powerful force for good. And so one of the things that your report raises is if you're asking these companies which have these net zero commitments to come to your state and then you're going to build gas plants to provide them, they're not going to come.
And I guess I'm just, I'd just like to hear more about that. Like, is that, is the voluntary corporate thing really going to serve as a check on this, or ultimately, are they just going to go where things are cheapest?
Michelle Solomon
That's a great question. And, you know, I won't pretend to have a perfect eye into exactly what all of these companies are going to do, of course, but there has been a lot of evidence shown that a lot of these companies, particularly the tech companies, they have clean energy goals and they do want to meet them. And they are potentially willing to pay a premium for clean resources. So, I think they have a lot of power in this situation to go where the utility is going to build the resources that they want. Of course, they want to also get the power.
They want to come online quickly and they want to go where they can get power quickly. But I think it's really important that if they are committed to these climate goals that they do, you know, stick to them and hold the utilities to them as well.
David Roberts
I mean, you kind of have to wonder, if they care so much about renewable energy, why they want to herd into the Southeast in the first place. Like, what is attractive about the Southeast for these facilities?
Eric Gimon
I mean, in the Southeast they have low rates, so low, you know, unit cost, not necessarily low bills, and there's been pretty easy access to transmission and things like that. But some of that is ending. So there's been a lot of conflict between Virginia and some of its neighboring states in the PJM system over the amount of money that's now going to have to be invested to upgrade transmission. So it kind of moves around. We also suspect a little bit of venue shopping. So the same data project is potentially appearing several places and then shows up on the breathless spreadsheet as several times what actually is there. But it's still a fairly nebulous thing.
Michelle Solomon
I would just add also that I think the reality is also that data centers are showing up in a lot of different places, but again, it's not every place that is sounding the alarm to the same extent on this.
David Roberts
Yeah, and we should say that, I mean, you both sort of alluded to this, but just to make it explicit, the areas that are loudest in demanding new gas do not have the most load growth in their forecasts. And also, getting back to something Eric was talking about a minute ago, have some of the worst records on energy efficiency. Do you know what I mean? So, like, of all the places where load growth ought to, you know, where load growth might call for new gas, the places that are asking for the gas don't actually seem like, ought to be the top candidates.
Eric Gimon
Exactly.
David Roberts
Let's talk a little bit about what utilities can do then in the short term. Let's talk about some of these short-term things. One of the arguments here is that, and I'd love to hear you guys expand on this a little bit, is that there's a little bit more room in the interconnection queue than people might think, and that there are like, spots opening up in the interconnection queues. Just that there are just, there's more room to interconnect than people might think. That was, I guess, came as a little bit of a surprise to me. Can you expand on that?
Eric Gimon
That was not one of the points we brought up in the report, but it's an interesting topic to bring up. In the Southeast, those are vertically integrated utilities, so they control the interconnection process. It's not like what we're seeing in PJM. And so they're making choices about what can connect and not connect. And I'm sure they could choose to connect a lot more. Like I pointed out with Duke, they're talking about smaller projects than are actually possible. What we did talk about also in the report is reconductoring. So there are ways to quickly upgrade a lot of the grid.
David Roberts
I'm going to get to that. This is number one on your list here. It says, "Build renewables and storage where you can."
Eric Gimon
All right.
David Roberts
Even though interconnection queues are clogged, there remain plenty of places to connect renewables. That's the point I'm trying to get at.
Michelle Solomon
I think one thing we are talking about there is actually using existing interconnection points. So, for instance, at a retired coal plant, or a coal plant that's soon to retire, there is existing interconnection infrastructure there and existing interconnection rights there that you could potentially reuse for new clean energy, new solar plus storage, and things like that. So we have a lot of existing sites that could go through faster interconnection processes that aren't being fully utilized.
David Roberts
Oh, that's interesting. So, if you have an interconnection at a coal plant, can you just transfer it? Like, if a solar power plant comes along and builds in the same spot, can they just take it over? They don't have to reapply.
Michelle Solomon
The rules depend. They're different a bit region by region. So, for instance, MISO, the kind of central, Midwest system operator, they have a separate interconnection process for reusing an existing interconnection, whereas PJM, I think their process is a little bit more difficult. But with a utility like a vertically integrated utility that has control over the interconnection process, you know, they may be able to. I think they may need to potentially apply to FERC. The rules are a bit different in different places, but there is certainly existing interconnection capacity. There was an RMI report that found that there could be, you know, around 250 gigawatts of potential existing interconnection space that could be used.
David Roberts
Oh, interesting. That's a lot.
Michelle Solomon
That is a lot, yes.
David Roberts
So, yeah, that's not a marginal thing.
Michelle Solomon
No.
David Roberts
All right, so there are spots in the interconnection queue that are opening up that could be reused.
Michelle Solomon
Well, yeah, it's not necessarily in the queue, but just on the grid, there are interesting areas. Yeah.
David Roberts
Right. Sorry. I've started saying "interconnection queue" as a single term now. This is just the interconnection, just the grid, right? That are already on the grid to be reused.
Yeah, also, you have to keep in mind the relative scale of things. So when you hear about the queue problems in PJM and MISO, you're talking about a traffic jam of like thousands of gigawatts of projects. If you look at what's actually getting built this year, it's 93% solar, wind, and battery, and it's in the hundreds of gigawatts. So why is it so foreign to some entities that we could build a little bit more of what's mostly being built right now?
I do — one does wonder. And so, the second item here is generating closer to demand, which I take to be a reference to both solar and batteries. I'd love to hear you expand on this a little bit because one of the things that comes up a lot is not just in this context, but generally, is the use of distributed energy to forestall just this problem, this sort of like queue, this interconnection queue jam up that we have, and this supposedly daunting new demand that's coming online and the lack of transmission capacity, and this whole sort of dilemma we're in, distributed energy helps solve all of that.
Talk about that just a little bit, how this could be brought to bear by these utilities.
Eric Gimon
Well, to prove that I'm a faithful listener, Dave, I think I can reference one of your recent shows on DERs, right? I think it's a twin story, right, of getting better mileage from the infrastructure we have now by generating closer. If we have the right incentives, right, we could do DERs the wrong way, we could do them the right way. But some of the things you can do with DERs, like integrating with VPPs (virtual power plants) or connecting them with storage, can really push back a lot of the need that the utility is talking about in terms of meeting peak demand on a cold winter day and so on.
The other element of the story, going to the long-term picture, right, these data centers, you're talking about maybe 100, 150 terawatt hours a year of extra load in your kind of worst scenario right now, if you were just going to replace all the aviation fuel in the US with something from, like hydrogen and biostock, with a lot of electricity coming in, you're talking about maybe 600 terawatt hours, right. Now, where are we seeing green sustainable aviation fuels projects kind of starting to germinate? Intersect Power has a project in the panhandle of Texas. That's several hundred megawatts of wind, several hundred megawatts of solar, and several hundred megawatts of electrolyzer all co-located. So they don't need to be interconnecting to the grid.
And then we had the announcement from DOE about green steel in Mississippi, right — that's Hy Stor. They also have wind and solar and batteries and hydrogen storage all in one place. So a big part of the long-term growth story, especially, I think, with industry, will be co-locating closer to where the resources are.
David Roberts
Yes, I'm a big fan of this. Like, I mean, sort of intuitively, when I hear about data centers, you know, and the threat to peak, they're sort of driving the peaks up. My first thought is just like, why don't you just require data centers to put a big stack of batteries next to them and shave the peak? Like, it doesn't seem insoluble. It doesn't seem, you know, like, why isn't that obvious? Why isn't that happening?
Eric Gimon
That is an obvious answer. It is something that may happen as the rubber hits the road. I mean, a lot of what we're hearing about this is really kind of proxy messaging in the Washington Post and New York Times and so on. When I talked to one of the big IT companies just last fall, they didn't seem that concerned. I think it's starting to become more pressing for them, but they were more concerned about being able to run their machines all the time and being near where people would live, that would run their centers, than moving to someplace where the resource was more accessible.
But now, I'm hearing stories about how they want to break these projects down to 10 megawatts so they can bypass some of the load queues. And so, as they get a little bit more desperate to access, we may see more creative decision and more creative problem-solving. It would be nice if the utilities were a partner in that.
David Roberts
It really would be, Eric. Michelle, all right, we've put it off long enough. Let's talk about another favorite subject here in Volts-world, which is getting more out of the existing grid. Talk about how that could help address this specific problem.
Michelle Solomon
Yeah, so I guess there's kind of two components to getting more — well, there's probably many components to getting more out of the existing grid, but a couple that I'll just address now. One is related to the interconnection queue and being able to interconnect more new resources faster. And as has been talked about on this podcast, we're in a bit of a bind with transmission capacity, which is making it more difficult to connect these new projects to the grid because they need transmission capacity to be able to move the energy that's generated to where it's needed. Unfortunately, it's very difficult to build new transmission in this country.
It can take five to ten years, even up to 15 years, if it happens at all, to build new, large transmission lines. But luckily, there are a lot of ways to reuse the existing transmission rights of way. And I know you already talked about it at length, reconductoring in one of your previous podcasts. So that's a really great way to increase transmission capacity. You can up to double the transmission capacity on an existing line, and you can do it in only 18 to 36 months, compared to the five to up to 15 years even.
David Roberts
And for listeners who missed that, it's just replacing power lines with new, better, more conductive material, basically just upgrading the line itself.
Michelle Solomon
Yes, exactly. And then, I think, the Department of Energy today just released their report, their liftoff report on grid modernization, which I'm very excited to read. And they cover a wide, wide range of technologies, including things like dynamic line ratings, which kind of monitor the grid in real time and let you know where you can run some more electricity through. And those take only months to kind of get going. And they can increase capacity, I think, at least up to 30%, if not more. So, there's a lot of ways to solve this interconnection queue issue by using the existing transmission rights of way, instead of waiting up to a decade for a new transmission.
And then, the other aspect is sharing resources more. So, we have a lot of resources spread out across the country, and not every region is experiencing their peak in electricity demand at different times. So, if we can share our generation resources a bit better, that means we need to overall build fewer new generators.
David Roberts
Yes, it's almost like operating grid regions, like little disconnected islands, is very stupid.
Michelle Solomon
Yes, it's certainly not the best option in terms of reliability or price. So you can save a lot of money, and you can improve reliability by sharing resources between regions better and even within regions better.
David Roberts
Right. And so, just for listeners, you know, all these utilities, all these vertically integrated utilities down in the Southeast, they all have, every utility has a reserve margin, some capacity in reserve, just in case. But as you say in the report, like this grid region has a bunch of reserve capacity. The region next door has a bunch of reserve capacity. They all have a bunch of reserve capacity. If they could just share reserve capacity, you could slice so much of that off and save so much money, yes. It's the same logic as interconnecting transmission regions, right?
It's just like, why are we not sharing? Volt's listeners are also familiar with the work that's happening out west to try to create — there, famously, is not an electricity market, there's not a central grid operator out west, and they're in the midst of trying to — fumbling their way toward one precisely to do more of that sharing. Is there anything like that underway in the Southeast? Or are these hidebound utilities just opposed to that down the line?
Eric Gimon
A few years ago, kind of right in the COVID time, we did some modeling of the Southeast with Vibrant Clean Energy — that's Chris Clack's modeling outfit, he's now with Pattern — just to see what could be done. And it came out like a sore thumb how badly the current operation is serving its customers in terms of what we talked about. So there was operational savings just from sharing reserves and things like that. We also found that just the way if you added up all their plans, it gave you something much less rational than if you try to kind of centrally plan the whole region.
Now we're not advocating for some central planner in our mind, maybe a market or something with discipline, the build out, but there was an even bigger difference there. Those utilities came under pressure and so they came up with something called SEEM.
David Roberts
Oh, right, I forgot about that.
Eric Gimon
I always thought this was kind of blowing smoke.
David Roberts
This is the "Southeastern Energy Market..." something. Some one quasi version of a market.
Eric Gimon
Right. So, when we were writing this up, I contacted one of our contacts in the Southeast that we used to kind of truth things when we worked on our paper and asked him, "Well, how's it going with this SEEM thing?" And he said, "You know, the traded volumes are like tens of millions of dollars," which in electricity business is nothing. So, yeah, like you said, it's kind of a version of what's going on in the West. So again, the Southeast is not looking like a pathfinder in progressing forward.
David Roberts
Just like for listeners' benefits, describe some of the reasons why utilities might be resistant to the kinds of things we're talking about and leaning instead to gas. I mean, this is an old story again on Volts, the sort of perverse incentives facing utilities and their kind of hidebound ways and their conservatism, all that. Just talk a little bit about why utilities with their current staff and incentives are not well-suited to embracing this future we're talking about.
Eric Gimon
Well, I'll start from like nice-ish to less nice about it. So, in 2012, I worked as a technical expert representing boat solar in California long-term procurement. And it was interesting to hear the California system operator, which doesn't own any plants, right, doesn't have an economic incentive.
David Roberts
Right.
Eric Gimon
They really pushed back on some of these planning documents. They were like, "Well, we're happy to compete any resource you want for our reliability needs, as long as it quacks exactly like a gas plant." If it looks exactly like a gas plant, they will let you do what you want to do. And it's taken some work. That organization has really made a lot of progress since 2012 in terms of thinking about being more creative with solutions. So that's the hidebound engineering culture. We also see things said in these IRP documents in the Southeast about "batteries being kind of an unproven technology."
Batteries are not some technology from Mars. I mean, we're seeing tens of gigawatts of deployment now, especially in the West.
David Roberts
Yeah, I'm pretty sure they work.
Eric Gimon
Yeah, the second part is siloing. Right. So, even though most of these companies in the Southeast are owned by Southern Company, they're all individual companies. And so, they each have their own kind of dynamic in terms of wanting to, like, build out reserves, like you said, and also just kind of build out the so-called rate base. Right. What they're allowed to charge their customers for and anything they do can add on there is good for business. And then there's also Florida, which is kind of this separate entity, and there's not a lot of love lost in terms of trading back and forth between Florida and the rest of the Southeast, where that would be of great benefit to both.
David Roberts
I just want to pause there because it's my duty as host of Volts to take my shoe off and pound it on the table whenever this point comes up. If you confront one of these utilities and say, "Hey, instead of building something of your own, you could share with your neighbors, save a ton of money," that is directly bad for the utility's bottom line. The utility makes its returns by building stuff. It wants to build stuff. So, if you're offering it opportunities, "Hey, you can avoid building stuff," right on the front end of it like that, they're hostile to that just by their built-in incentives. I have to reinforce that point as often as I can.
Eric Gimon
Exactly. And a few months ago, there was a Brattle report. It kind of modeled some kind of putative utility, not a specific one, mid-size. And it looked at dealing with a peak need issue, either with a gas speaker, with a battery, or with a virtual power plant. And the cost benefit got better and better as you went from left to right. The virtual power plant was clearly the option to go. And I remember thinking of this graph and thinking, "Yes, but for the utility, you reverse the flow." Because the amount of capital involved in the utility needs to go up and up for them to be interested.
The other part, and this part's a little harder to completely parse. But some of these utilities are involved on the gas side, too. And in order to build new gas pipelines, you need to show that you have a customer for it. And there have been cases, law cases about this where one hand of the octopus is saying, "I want to build this gas pipeline." And then another hand, the tentacle, the octopus is saying, "Oh, we're a ready customer right here that wants this gas pipeline."
David Roberts
Works out well.
Eric Gimon
And we're not sure exactly how much money they're making on the capital there, or if they make extra money on the flow of fuel. I think it's a complicated situation from place to place. It's also a bit of a source of skepticism when you have a business that can double-dip both on the electricity side, on the gas side when they start favoring gas solutions.
David Roberts
Yeah, that's interesting. This is one of those things that when I first started looking into and studying utilities, kind of blew me away. So, we did this restructuring where we're going to separate these companies because if they're co-owned, there's obvious conflicts of interest, but we'll allow a third company to own both of them. And I just like, well, that's like, you just haven't separated them, have you? The separation is entirely notional. Of course, they're going to be like, of course the one hand is going to be serving the other hand. It's just crazy to me that utility holding companies are a thing at all.
So, that's the utility incentives. They are askew in a variety of ways. And the culture is a little bit hidebound and backward-looking. But Michelle, maybe you can speak to this. What about regulators? This is why we have regulators. This is the whole point of regulators, is if you're a regulator and you hear, "I want new gas plants," and then you also read in the docket, "Oh, but by the way, for much less money, we could use interconnection better. We could reconductor, we could use gets, we could etcetera, etcetera." All these things, all these other solutions are available for cheaper.
That's what the regulator is for. So, why are regulators, what are their incentives? Why aren't they putting a stop to this?
Michelle Solomon
That's a great question. And absolutely, the regulator is the first and best line of defense against the building of all of these new gas plants. One issue is that there is quite a bit of information asymmetry between the utilities and their regulators. You know, to some extent, the regulators only really know what the utilities are telling them, and a lot of these commissions are not, you know, they don't necessarily have the staff to go and vet every single one. These integrated resource plans are long, and it can be difficult for the commission to be able to independently verify each line item there.
So, that's one aspect of it. Another aspect that's not as related to the regulator, but in some of these states, the state legislature actually is kind of trying to go around the normal order of regulatory process and fast-track approval of these gas plants. So, there is kind of some political involvement on that side as well.
David Roberts
Interesting.
Michelle Solomon
Oh, I guess I will also just add on the point about why we're not doing as much reconductoring and GETs and things like that . Just a quick plug for a report that we released last week on some of the biggest barriers and policies to enable more reconductoring faster. And in this area, you have some of the same kind of disincentives to install these technologies that Eric just talked about. But particularly for the advanced conductors, they're a little bit caught in the middle because building a new transmission line is the big capital expenditure that the utility wants to do.
David Roberts
Yeah, that you can rate base.
Michelle Solomon
Exactly. And then, at the same time, while reconducting with an advanced conductor is much cheaper than building a new line, that advanced conductor is more expensive than the traditional conductor. So, to a regulator, there's a risk that it can look like gold plating the electricity system, which is what the regulator is there to stop. So, one thing that we're starting to see states do is kind of give the regulators a kind of mandate to look at these advanced conductors and do a cost-benefit analysis so that they are aware that this is not gold plating and that this is one of the best ways to expand the grid faster.
David Roberts
And a case study just came out — I shouldn't bring this up since I don't have it at hand — but I feel like a case study, just like the most involved case study thus far — LineVision was involved, the grid enhancing technology company — basically showing that, "Yeah, this works and it saves money over time." It's getting a little bit less abstract and more concrete that these things work.
Michelle Solomon
Yeah, it's gaining a lot of traction, and there are really fantastic options that we need to be implementing very quickly.
David Roberts
Can you rate base if you reconductor or is that maintenance?
Michelle Solomon
You can. It's just that there is some concern from the utilities that there is a risk that they won't be able to because it's not the cheapest option for installing a transmission line. So, just making sure that there's clarity around the benefits so that we remove that risk is really important.
David Roberts
Eric, the one other thing that was on your list of solutions that I forgot to hit on earlier, but I did want to ask about, because it's of immense interest to me is: pushing industrial customers to do more on demand response. You know, so like, I feel like when people are confronted by this rising load and then difficulty building transmission, they're like, "well, the only thing left is new power plants," but that leaves out the entire sort of family, the entire field of demand response. I've always thought that if you're electrifying a bunch of big industrial loads, or if you're bringing a bunch of new and big industrial loads onto the electricity system, they can either be a problem, which is sort of how they're being framed here, or if they can implement some flexibility, can be helpful, can be a help to the grid. So what could utilities do to try to juice that along?
Eric Gimon
It's a really interesting topic, Dave, and maybe a bigger topic than we have time for today. This is an area that we're actively investigating. One issue is that we would love to see flexible industry kind of cherry-picking the cheapest energy when people don't want it that much. Right?
David Roberts
Right.
Eric Gimon
But a lot of the T&D delivery cost is on a fixed basis, except for places like Texas where they have this 4CP system, this demand charge type system. And so, there's not a lot of incentive for flexibility because you're not going to get a discount on your bill. And if we want to electrify industry, we need them to be able to have cheap enough electricity that it's going to compete with fossil fuels unless we put some really strict standards on how much fossil fuel they can use, which I'm sure you and I would be okay with, but seems politically more difficult. So, that is going to be a big lift over the next few years, is how to think about rate reform, not just with the recalcitrant utilities, but even in the more progressive places, so that in a fair way you can offer a cheaper price to industry in return for flexibility and kind of nailing down the details of what kind of behavior leads to what kind of discount is important.
But just to zoom out a little bit here, this is part of a narrative war, right? Michelle alluded to the EPA rules coming on. There's a lot of walking around the halls of power here in DC where we're visiting this week by certain folks. And this narrative war kind of comes and goes as you've seen it so many times. Right. The overall narrative of "we can't do it." And then it kind of gets — they crank up the volume when they need to. And part of the way of framing this narrative is this idea, "Oh, I have this one load A, I need to match it with this supply B," right.
Or, back in the day, it was like every wind you install has to be, like, backed up one to one with some gas plant or something. But that completely sidesteps the fact that all this is connected to a system. Right. There's a lot of capacity in the system today to serve whatever need you have. And it may be that even if you add a 24/7 load, it only causes challenges a few hours a year, not thousands of hours a year. Yeah.
David Roberts
Yeah. I want to stop and stress this, too, because this is an important conceptual point to understand. I think nobody's talking about this new demand, sort of like maxing out the system. There's plenty of spare capacity in the system. It's just the peaks. It's just the peak that is the concern. So if you could shave, if these things could shave their peak or somehow shift their demand away from peaks, we've got room on the system for plenty more demand.
Eric Gimon
Right. Or, you know, if some of the money they're willing to put up to get quick connected service went instead of to new generation to programs that shape load, not even necessarily their load. Right. I mean, the Duke IRP is also assuming a lot of EV's, which is nice, but it's assuming they're all charging in the worst times, which is not nice. Maybe a little time and money could be spent on adjusting that instead of building a whole new combined cycle gas plant to —
David Roberts
You could get so much more for so much less by just tweaking EV charging than you could by building that gas plant.
Michelle Solomon
Yeah, and just to add on that, you know, the peak that we're seeing is in the winter. It's a short peak in the winter. And as we know, gas plants are not always the most reliable resource during a winter peak.
David Roberts
Been in the news recently.
Michelle Solomon
That's an important point to make. And then again, in the Southeast, we know that there's quite a lot of electric heating, which is not the most efficient form of heating, particularly when it's electric resistance baseboard heating or something like that. So another way that's not exactly demand flexing, but more on the energy efficiency side, is shifting that electric resistance heating over to heat pumps. This would be a really great way to reduce your short duration winter peak.
David Roberts
Thank you for satisfying the legal requirement that heat pumps be mentioned at least once per episode on Volts, Michelle. I appreciate that.
Michelle Solomon
I'm happy to serve that role.
David Roberts
Yes, shifting resistance heat to heat pumps in and of itself could free up enormous amounts of room here. I'm sort of curious as a side thing, and speaking of issues that probably deserve their own podcast, but do either of you have a good sense, or does anyone yet have a very good sense of how flexible data centers can be? Like, what is the potential here for demand response?
Eric Gimon
I've heard different stories on the topic. Google has a white paper where they talk about flexing some of their computing power, and obviously, those are machines you can turn up and down very fast. Most of it's just like cooling loads and the computational loads. But then I've talked to others that say, "Look, Eric, these chips that we're buying are so expensive and they're dear to get your hands on that you have to kind of run them as much as you can while you have them, and the electricity is still not that expensive a part of the overall value proposition."
So, there's not a lot of incentive yet to modify your load, unless you're like Bitcoins or something like that. But for the big AI people, it's a question of incentives. On the other hand, these training models, they're just cranking away, learning. They're not necessarily responding to a real-time query from you about where to get the best taco in Oklahoma or something. So, that could be theoretically shifted around. It's just a question of incentives.
David Roberts
Yeah, yeah, it does seem like AI in particular, but there's a lot of that that's not sort of on demand, like a search query, like you say. It seems like AI in particular ought to be more shiftable than traditional compute. But maybe I'm wrong about that. That's definitely something I want to explore more, because if you could get all those data centers to be active participants, right, sort of like active demand shifting participants, that's a huge, potentially a huge ally instead of a huge drain. You know, we've gone over the incentives facing utilities. The reason they're sort of groping for gas when confronted.
I mean, they kind of grope for gas when they're confronted by anything. When the wind, when there's a gust of wind, they grope for gas. Once again, they're groping for gas and why they always do that. And we've gone through a couple of kind of short-term, immediate things that utilities can do to forestall that need instead of new gas plans. But I think we all agree, and we've discussed in the longer term, there is a really large amount of new load growth coming, and so these sort of short-term compensatory policies are not going to do the whole job.
I just love to hear a few reflections from either or both of you about what to do in the long term. What are longer-term solutions to this problem? And I want to say, raze the entire utility regulatory system to the ground and rebuild it from the ground up in a rational, national way, that's probably not on the table, but I'm curious, what are your thoughts about long-term solutions?
Michelle Solomon
Yeah, our paper is a bit more focused on the details of the short-term solutions, but long term, there's, again, quite a lot we need to do to build out our electricity system to be able to support this huge load growth due to electrification. So, a couple of the things that we mention in the paper are, you know, not just reconductoring and getting more out of the existing grid, but really investing in those new high voltage transmission lines, both on the regional and interregional level. And we also talk a little bit about the need for new technologies, this kind of class of technologies called clean firm technologies that we will need to be able to replace the existing gas fleet. So, that includes things like geothermal energy, long-duration energy storage, industrial thermal batteries, potentially advanced nuclear, other technologies.
David Roberts
You've now hit all my favorites, Michelle.
Michelle Solomon
Yeah. So it's really expanding our transmission system and getting these technologies that can completely get us off of fossil fuels ready to go.
Eric Gimon
Yeah, I would add two things. One is there's been a misapprehension from some people who heard about our report that we're somehow saying, "Let's turn off all the gas tomorrow." That is not what we're saying. What we're saying is we already have a lot of gas. We don't need to add more. Because the challenge is going to be figuring out how to kind of slowly wean ourselves off.
David Roberts
Right.
Eric Gimon
And adding more things to the rate base is just making things harder. So that's the first thing to keep in mind.
David Roberts
Can I ask real quick? Like Duke, I think, has a decarbonization target. Almost all the utilities do now. And this, just like building a new gas plant, as you say, is like a five-year operation that results in a 20 to 30-year asset, is just straightforwardly incompatible with those decarbonization goals. It seems like that ought to be the end of it. They're either serious about those goals or they're not. And they seem to be like, it seems like those goals don't even trouble them that much, that they don't even consider them that much. Is that accurate?
Eric Gimon
That's why they have to foster an environment of panic. They're saying, "Oh, yeah, those are like long — we want to end poverty and suffering in the world. But right now —"
David Roberts
Not today!
Eric Gimon
"My car across the road. Right. Not today." So, I think there's a little bit of that driving this kind of breathlessness about load growth.
David Roberts
Right.
Eric Gimon
The other thing that needs to happen, I think, is a shift in mentality, not just the utility, but like broadly through society, to associate new, clean energy with prosperity. Right. That the states that are going to thrive and do well in the next decades are the ones that can attract investment in clean energy technologies in their state, and development of wind and solar in their states, because those will be the engines, those will be the sources of low-cost power for your thermal batteries, for your green steel, for all these things. And the states that are the best at doing that are the ones that are going to have an advantage.
And I think once some states start seeing how much more advantaged other states are, it's going to be harder for utilities to make these arguments. And they may even come to a reckoning. There's a very amusing case of the most expensive coal plant in the US was called Dolet Hills — it's retired now — in Mississippi. But apparently there's talk of clawing back billions of dollars of money that was paid to that plant because it was uneconomic. So I don't know. I wouldn't want to put that specter in front of some of these Southeast utilities. But I think there'll be a point where people turn around and ask them, "Why are these other states doing so much better than us?"
David Roberts
Yeah. And it's just like, it's just a fact that if we're going to hit net zero by 2050, that these gas plants that they're proposing to build are necessarily not going to serve their entire lifetime. Right. I mean, it's just, they're incommensurate. It's funny, these southern states, you know, they have these sort of low wage, low labor standards, low tax corporate friendly sort of environments, and they'll give corporations billions of dollars in subsidies to site things there. But will they clean up their grids to attract business? It'll be, we'll see which force is stronger, their antipathy towards hippies and clean energy or their love of new corporate activity.
I appreciate this, guys. This is a great, seems to me like a great sort of first entry into a debate that is going to be with us for some time now. And thanks for coming on and walking through it with us.
Michelle Solomon
Yeah, thank you so much for having us. It's been really great.
Eric Gimon
You know, long-time listener, first-time caller, so it's great to be on.
Michelle Solomon
Seriously.
David Roberts
Thank you for listening to the Volts podcast. It is ad-free, powered entirely by listeners like you. If you value conversations like this, please consider becoming a paid Volts subscriber at volts.wtf. Yes, that's volts.wtf. So that I can continue doing this work. Thank you so much and I'll see you next time.
Share this post