Note: The Podcast Recording with Graphs is Here.
What would happen if Texas started turning away new wind and solar projects or even retiring some of the capacity we already have?
That’s not merely a hypothetical. It’s what several bills that have passed the Texas Senate aim to do. And according to new modeling from Aurora Energy Research, the consequences could be severe: higher prices, increased risk of rolling outages, and weaker reliability at the very moment we need the grid to do more.
I sat down with Olivier Beaufils, Head of US Central at Aurora, and a returning guest on the podcast, to walk through the results of their latest two studies. We discussed what it means for Texans: whether you’re a policymaker, utility, industrial operator, or just trying to keep your power bill manageable.
In the first study, Aurora modeled scenarios where Texas slows or blocks the addition of new solar and wind. The findings are stark:
By 2035, power prices rise 14% compared to a business-as-usual scenario
By 2050, prices jump 38%
A conservative forecast still shows 400,000–600,000 homes losing power in summer peak conditions due to 2–3 gigawatts of load shed
Delays in thermal generation supply chains mean replacement capacity can’t come online fast enough to close the gap
These results don’t even account for the possible retirement of existing renewables, which SB 715 would accelerate.
The second study focused on demand-side solutions like residential demand response, data center flexibility, and energy efficiency upgrades (especially replacing resistance heating with heat pumps). These are often overlooked, but for those looking to strengthen the grid they can dramatically reduce stress on the grid, while saving Texans money.
Key findings:
Demand response could save every Texan household $87 per year in conservative estimates, even if they don’t participate
Industrial customers could save $2 million annually under a moderate demand response program
Replacing resistance heat with heat pumps could save households over $400/year
In a Winter Storm Elliott-style event, widespread heat pump adoption could eliminate up to 6 GW of potential load shed; the impact would be bigger in a Winter Storm Uri-like event
Taken together, these studies paint a clear picture: A more diverse, more flexible, and more efficient grid is also a more affordable and more reliable one. We need more of everything, but especially the resources that are fast and scalable.
I’m grateful to Olivier and the Aurora team for bringing rigorous analysis to a debate too often shaped by ideology or misinformation. These are exactly the kinds of conversations we need to be having as Texas faces record-breaking load growth and increasingly extreme weather.
If you enjoy the podcast, consider subscribing and sharing with a friend or colleague. It helps more people find these conversations and stay informed on what’s really happening with Texas energy.
Timestamps
00:00 – Introduction & guest reintroduction (Olivier Brochu, Aurora)
02:04 – Start of Study 1: Limiting solar and wind in ERCOT
06:00 – Supply chain risk & reliability impact of throttling renewables
11:00 – Load shedding during heatwaves
13:30 – Wholesale power prices and affordability under constraint scenarios
16:30 – Impact on customer bills and total system costs
20:30 – Clarification: this is not a bill analysis, but a scenario study
21:30 – Start of Study 2: Demand-side flexibility & efficiency
24:30 – Demand response from residential vs. commercial/industrial users
29:00 – Data centers, system peaks & the synergy with residential flexibility
32:00 – Price suppression benefits of demand response
34:30 – Indirect savings: how even non-participants benefit
37:00 – Addressing resistance heating and the efficiency potential
41:00 – Heat pumps as a system reliability solution
44:30 – System-wide benefits & mitigating winter risk
46:30 – Efficiency = cost savings + resilience
50:00 – Final thoughts: all-of-the-above strategy & public/private collaboration
52:00 – Outro & where to find Aurora’s work
Resources
Aurora & Team
Olivier Beaufils - Linkedin
Reports & Modeling
YouTube Channel for DougLewinEnergy
Media Coverage
Utility Dive on Aurora Modeling - Texas Renewables Restriction
Related Posts on DougLewin.com
A Time for Choosing (on Senate Bill 819)
The Texas Senate Is a Threat to U.S. Energy Security (Senate Bill 388)
Another Anti-Energy, Anti-Growth Bill Looms (on Senate Bill 715 and House Bill 3356)
Transcript
Doug Lewin (00:04.45)
Welcome to the Energy Capital Podcast. I'm your host, Doug Lewin. This week I have with me Olivier Beaufils from Aurora Energy Research. Y'all have heard Olivier before on the podcast. I interviewed him last year when Aurora had done a fantastic study on battery energy storage in the ERCOT Market and some of the energy savings that battery storage brings during a winter storm. That was a great study. We'll put a link to that show and the study in the show notes.
But today we're going to talk about two additional studies Aurora has come out with that are very, very relevant to what is going on in Texas right now. One looked at the impact of renewable energy development, if it were to be throttled by some of the proposals that are out there, what that would mean for prices and reliability in the market. And another one looks at demand side management, energy efficiency and demand response and its impacts.
Doug Lewin (00:59.02)
I'm thrilled, Olivier, you could make time to be with us. Thanks so much for being on the Energy Capital Podcast once again.
Olivier Beaufils (01:05.4)
Doug, it's a pleasure to be back. Thanks for having me.
Doug Lewin (01:08.588)
All right, great. Well, let's dive right in. As I said, just a second ago, Aurora Energy Research is doing great work. You guys are global. Folks are interested in anything going on in the energy space. I encourage them to go look at your website. But you guys have a presence here in Texas with a big office here and just been doing great work, really helping people understand some of these market dynamics that are going on.
Doug Lewin (01:32.364)
Let me also just say on the outset for those that are listening on the podcast, we are at a few points in this conversation going to reference some slides from these reports that Aurora has done. This podcast will also be posted on our YouTube channel, @DougLewinEnergy, so you can go there and if you can watch, you'll be able to see the slides, but we will make sure to describe, Olivier, you check me on this and make sure I'm doing it, that we are describing it so somebody that isn't.
Doug Lewin (01:59.052)
watching the podcast, but listening only will be able to follow along. Let's go ahead and start with the one that is called impact of limiting solar and wind development in the ERCOT market. So this is getting a lot of attention. Obviously it's being talked about a whole lot. News articles in one of those articles, or actually a couple of those articles, ERCOT CEO Pablo Vegas talked about how some of these bills that particularly those that would remove current generation or restrict
Doug Lewin (02:28.174)
future renewable energy development, throttle sort of the future development, particularly Senate Bill 819, Senate Bill 388, Senate Bill 715, and 3356. We'll talk about all of those as we go through. And you can find more of this at the Texas Energy and Power Newsletter and on previous podcasts. Any of those bills, the outcome would be less reliability and higher costs. Can you just, from the highest level, Olivier,
Doug Lewin (02:54.99)
Let's talk about why that is because it's counterintuitive to some people. They think that if you restricted renewables, you would automatically have some kind of increase in reliability. Just from the highest level, why would these different, why would the outcome of these proposals be lower reliability and higher costs?
Olivier Beaufils (03:13.13)
Absolutely. Yeah. I can talk about that. Maybe just a quick disclaimer before I dive into it. so, Aura is a power market analytics firm. and we don't have any specific agenda, right? Our job is to run power models and do analysis. we work with everyone. We worked with a lot of the larger thermal generators and air cot and other markets. we work with batteries. We worked with renewables. We work a lot with banks and investors. And so
Olivier Beaufils (03:40.03)
So our job is to do the analysis and figure out what it tells us and offer it to policymakers to, you know, to the public so that they can make more informed decisions. Right. So I think that's important to say before we dive in here, but on your question more specifically, why is this counterintuitive? Well, there's two aspects to it. I think there's one very simple aspect of, you know, we're seeing a lot of demand. And if we're about to remove some of the supply.
Olivier Beaufils (04:08.61)
that could come through either through solar generation, so wind generation, we're going to be in a position where there's just gonna be less supply on the supply curve and generally higher pricing because basic supply demand economics for anyone that's taken an Econ 101 class, you're gonna be higher on the supply curve that is going to result in more expensive generation. So that's the sort of very overarching point. Right now we need the megawatts we can get because of all this demand growth. And so when you remove that,
Olivier Beaufils (04:38.55)
It makes things more expensive. But the second piece that I think is maybe less well understood of renewables is that, you know, the cost of renewables is mostly in building the infrastructure. Once you've built it and in Texas, we do this with private capital, right? Ratepayers do not have to pay for any of this infrastructure being built. It's private capital taking risk. So once it's built, the wind is free. The sun is free, right? We know this. So operating it is virtually nothing.
Olivier Beaufils (05:08.384)
And so what happens is all these renewables, they sit at the very bottom of the supply curve. And so that means they're the one they're dispatched first, they're the least costly for the system. And then after those are dispatched, we then turn on or keep on more expensive generation, nuclear, coal, gas, and then increasing by order of more and more expensive. If you remove a big chunk of those less expensive renewables,
Olivier Beaufils (05:35.608)
then the system is going to become more costly because you will have to turn on older, less efficient plants more often. And so on average, we'll have higher costs for your system.
Doug Lewin (05:47.266)
Yeah. And then add to that and you guys reading directly from the report, restricting renewables while thermal supply chains are constrained leads to load shedding under summer heat wave conditions. So the other key piece of this is you're not restricting renewables or in the case of 715 and 3356 actually retiring existing renewables. You're doing it in a context of it's very hard to build
Doug Lewin (06:15.95)
replacement capacity right now because of supply chains. Can you talk about that? And then the second part of that leads to load shedding under summer heat wave conditions. You know, we're, recording on the afternoon of May 13th. It's 97 degrees right now. We don't know what the summer is going to be like, but it's Texas. know it's going to be hot. You guys calculated two or three gigawatts of potential rolling outages just based on restricting future growth. So maybe talk about those.
Doug Lewin (06:44.898)
those two aspects, the difficulties in the supply chain and what your modeling found that would mean for summer heat waves.
Olivier Beaufils (06:50.946)
Right, no, I think that's a key point. So we're not in a world where you can just decide that you're not going to build those solar or wind projects that are in the queue and suddenly you want to build gas generation and that's just going to show up tomorrow. Those projects take years to develop. We have some slides we put out. Typically a thermal project is three, three and a half years. A wind or solar project can be four or five years. And so if...
Olivier Beaufils (07:17.454)
Today, you decide that you suddenly want five gigawatts or so of more capacity, you're not necessarily going to be able to put that online until 2029, 2030, probably even later. Because also, the other part for people to understand is, five years ago, gas was in a really tough spot. You had gas turbine manufacturers, GE Mitsubishi scaling down capacity because demand for gas turbine was way down.
Olivier Beaufils (07:45.654)
And so they reduce their production capacity like any sensible, rational manufacturer. And so today suddenly we're asking them to double, triple their production and they can't do it instantaneously. Obviously they will adjust some of that production, but it will take time. And they're also wary of an oversupply situation. So you have limitations of how much you can build, how fast.
Doug Lewin (08:05.153)
Exactly.
Olivier Beaufils (08:10.732)
And so that means every gigawatt or megawatt you take out of the market between now and the moment where you can build that new generation effectively is a megawatt that you will not have in the summer when temperatures are really high. Because majority of those projects have been in development for years already. So they are getting close to ready to build. And so if you prevent them from building or if you make it more expensive, more difficult so that people have to make a different decisions, decide not to proceed.
Olivier Beaufils (08:39.404)
then effectively you will have less capacity to generate. And as much as renewables are intermittent, think something that's also really important to understand is the sun doesn't always shine, but it shines pretty predictably. It generally rises at known time and sets at known time as well. And there's a pretty predictable curve. And even if we have a set of clouds that go over the project at some point, generally we sort of know when they're to generate.
Olivier Beaufils (09:08.294)
And the same with wind. You you will have weather events where wind drops, but you also have a lot of weather events where wind production is high. And so those projects do help the system. They generate power, except obviously solar at night. And so they help the system during those hours. Particularly solar in the middle of the day, which is when the demand is highest in aircott, is extremely helpful to the system. And combined that with batteries, we're able to shift a lot of the energy when we need it.
Doug Lewin (09:37.516)
Yeah. I mean, all, all resources contribute to reliability. Some contribute more than others. Obviously batteries or gas peakers have certain flexible attributes. Demand response has that as well, but all resources contribute to reliability and they all have other things they contribute to, including some cost impacts. let's do go ahead for those that are watching on YouTube, I'm going to go ahead and pull up part of this presentation.
Doug Lewin (10:06.594)
that you guys put out. is the one that's the impact of renewables. And I did want to make one more point about that, Olivier. It's interesting because you mentioned like, it does take a while for the manufacturers, the OEMs to ramp up production. On GE Vernova's earnings call, it may not have been the most recent one, it may have been two ago, but they talked about, they were talking like a lot of the oil and gas producers talk about capital discipline, right? About like, they don't necessarily want to just ramp up production because like you said a minute ago, oversupply is then a risk. They actually kind of like these high prices, right? So anybody that thinks that like, hey, well, this is just a temporary thing and in two, three, four years, we're gonna have plenty of gas turbines. A, that next two to three to four years is really important and we have to be careful. And two, that probably isn't gonna happen. We're probably still gonna be in a constrained environment. So you guys referred to that in this slide.
Doug Lewin (11:00.972)
When accounting for thermal supply chain constraints, restricting renewables causes load shed between 1.8 and 3.1 gigawatts across cases. So anything you want to say about this slide in particular, or it just kind of speaks for itself.
Olivier Beaufils (11:14.03)
Well, I think it's a reflection of a point you just made and having less megawatt coming from solar and wind. So essentially at the times when you need more energy because demand is increasing and just for our listeners, our demand forecast is conservative. So we are not taking the ERCOT forecast at face value. We think there's obviously a lot that can be discussed on how much of it is real and how much of it will come through. We put together our own forecast that's a lot less aggressive.
Olivier Beaufils (11:42.05)
I think we're right above a hundred gigawatts of forecast by 2030 and then climbing from there. But even in this sort of more conservative forecast, if you take away a significant amount of wind and solar generation, you will see periods in the middle of the summer during those tightest hours, generally when the sun sets before the wind generation has picked up and you have effectively less solar, less wind generation.
Olivier Beaufils (12:11.278)
do see essentially two to three gigawatts and even a little bit more so if you look at extreme weather events, but even in sort of a standard weather year, two to three gigawatts of load shed, which represents about 400 to 600,000 homes that would lose power. obviously, know, aircraft would try to do rotating outages and for those volumes they could, but would lose power over the summer when, you know, it's 105.
Olivier Beaufils (12:40.91)
plus outside. Yep.
Doug Lewin (12:42.134)
And
Doug Lewin (12:42.294)
I appreciate the point you're making about the load forecast, because obviously if your load forecast is higher, and ERCOT thinks by 2030 we're going to be in a 135 gigawatt or so scenario, you guys were right around 100. If you're at 135 or even you're somewhere in between, then obviously these load shed numbers would be much higher. And I think you guys, correct me if I'm wrong, for this study, you guys looked at throttling future development, but didn't look at
Olivier Beaufils (13:01.644)
The problem is even bigger. Right.
Doug Lewin (13:10.926)
the impact that how some of the bills, particularly 715 and 3356 could cause retirement of existing assets, right? This is just looking, this is just forward looking, correct? Yeah. So again,
Olivier Beaufils (13:20.792)
That's right. That's
Olivier Beaufils (13:22.295)
right. So we didn't even include that at that impact of potential retirements, which would make the problem even worse.
Doug Lewin (13:27.628)
So yeah, as extreme as this is, 400,000- 600,000 homes losing power in a heat wave, that sounds really bad and it is really bad. It actually could be much worse depending on some of the assumptions. Okay. So the next slide I want to pull up, and again, we'll talk through this, is the slide that shows some of the cost savings. Cause that is while renewables contribute to reliability, they contribute even in a more kind of a fulsome, mighty way to
Doug Lewin (13:56.654)
cost reductions. So the title of this slide, limiting renewables cause power prices to rise. Prices in the fully restrictable renewable scenario increased by 38 % by 2050. So you've got kind of a midline here, a midline projection, and then a higher scenario, and you're looking out over 20, 25 years. Can you kind of talk through the findings of the cost impact of restricting renewables?
Olivier Beaufils (14:22.414)
Yes, definitely. as you restrict more and more renewables development, and so as you mentioned, we have these two scenarios, one that essentially is sort of a 50 % is a halfway and one that is more fully restricting development past the next few years. So we're still accounting that some of the developments that's very advanced late stage would go through, but then after that, a very severe slowdown in the amount of wind and solar that would come into the system. so, you know, comparatively, cause you still need to meet load, we do model.
Olivier Beaufils (14:52.078)
additional thermal generation coming online to support the system. And so as you see over time, as you have to add more and more thermal generation to support the system, but already you can see the effect pretty significantly by 2035, you see a divergence between our central case, which is sort of view of the future that we think is most realistic based on load growth, based on market economics, right? Under the current set of rule. So you could
Olivier Beaufils (15:21.326)
call this a business as usual scenario, if you will, maybe a little bit conservative on the load side, but business as usual. And then those restricted scenarios, by the time you hit 2035, which is very important timing in terms of planning horizon for a lot of investments, you're 14 % higher. Your power prices are 14 % higher, which is very significant for residential and industrial consumers as I think we'll dive into in a second. And that is primarily
Olivier Beaufils (15:51.05)
a function of that very cheap part of the supply curve that renewables would fulfill having shrunk or not having grown as much as it would have under a business as your scenario, which means more expensive units have to be turned on more often. And so on average, you see just higher power prices.
Doug Lewin (16:11.374)
Yeah, mean, we're, this is a pretty dramatic, mean, 14%, uh, it, it's hard to, it'd be hard to overstate like what that would actually mean to the business environment of Texas. I mean, that mean, that is the difference between a whole lot of businesses choosing to locate here or deciding to go somewhere else. So that is, that's a pretty dramatic. again, depending on how all of this actually plays out and how much load growth there actually is.
Doug Lewin (16:38.092)
with the actual price of gases, a whole lot of other variables like that, we could see an even more dramatic increase. So I want to talk next about what that actually means. You started to allude to this to industrial and to residential customers. Can you talk through the impact on households and on large manufacturers and the like in the state of Texas?
Olivier Beaufils (17:00.93)
That's right. So we wanted to translate that into more, I would say, tangible terms for people, right? Like, why do I care? And so we looked at what I would call a large or medium sized industrial consumers with a load of about hundred megawatts, sort of base load consumption in aircotts, and looked at how its retail rates would evolve based on the relationship between wholesale prices and retail rates. Obviously there's some assumptions around how that evolves over time.
Olivier Beaufils (17:30.606)
but there's a very strong correlation between the two. And so we found that their bill would increase by about $6 million, over $6 million actually, with that power price increase. And then for an average Texas household, and we took really the average consumption of two bedroom or three bedroom family home, that represents about $225 a year in terms of their power bill. So a very significant increase that's really just linked to this decision to not
Olivier Beaufils (18:00.626)
add or discourage additional investment in wind and solar generation, which is obviously indirectly what those bills are going to do. Maybe one thing to add to that, right? So you could look at the bills and say, well, the modeling that Aurora did is not really reflective of what the bills are doing. And in a way you would be right. We decided to not model the language of the bills specifically. One, because there is a number of those. And so you need, you would need to do things very differently from each.
Olivier Beaufils (18:29.71)
because they're still very much in flux, right? They're being discussed and it's very hard to pinpoint a specific version. And they're all at the end of the day going to do a similar thing, which is discourage investment, make it costlier to invest in wind and solar. And so we wanted to already look at the impact of this lower investment in wind and solar rather than spend too much time discussing the specific language of the bill in this case.
Doug Lewin (18:32.661)
over time.
Olivier Beaufils (18:57.154)
And obviously we can go back to the language of the bill and discuss more of that. But in a way, if you want to think about 388 to comply with 388, you would have to be in that fully restricted renewable scenario. That is the only scenario by 2035 that has more than 50 % dispatchable capacity. So you're already in that sort of more extreme scenario to comply with that SB 388.
Doug Lewin (19:20.782)
And if I remember from your footnotes and details on this study, cause I did get in the weeds on this one, obviously, I think you guys modeled like an 1800 square foot house if memory serves. So like if you had a bigger house, which a lot of people in Texas do, there's a lot of 2,500 and 3000 square foot houses. Frankly, like I don't think there's too many people that are only paying $2,200 a year, which is what the slide shows right now for their electric bills.
Doug Lewin (19:45.782)
If I could pay that, I'd take it in a heartbeat. So I would just make the point that, it's not to obviously, Olivier, you I think the world of what Aurora does, I read your studies about what's going on around the world. It's not to dog the study. It's just to say that like, I think you guys have done something very responsible here, but very measured and very conservative in the way you're analyzing it. I think there's a strong argument to be made $225 would be on the low end of the range of what households would see, but still it's a pretty eye popping number. There's not a lot of.
Doug Lewin (20:14.702)
folks that would love to pay $225 more for electricity every year.
Olivier Beaufils (20:18.968)
I mean, particularly if you have a home that's not very well insulated, maybe with an older AC that's consuming a lot of power, you could easily see much higher than that.
Doug Lewin (20:27.554)
Which is probably a great segue because you guys did a second study and I do not want to give short shrift to that study. Renewables are phenomenal in Texas. We've obviously seen tremendous growth, you know, up to 70 gigawatts of wind and solar and over 10 gigawatts of storage. Texas is still lagging on the demand side. And so you guys did a great study there too. There's, there's a whole lot more I want to say about the renewable study, but I, but I want to make sure we leap.
Doug Lewin (20:55.982)
plenty of time to talk about this too, because I think a lot of times Olivier, know, folks are like, look, you can't possibly say, Hey, the grid is the most reliable. It could possibly be like we've done everything we needed to fix after winter storm, Yuri. have minimized the risk to the system. There's a lot more to be done. And I think one of those areas is on the demand side. So before we get into some of the charts and the figures and all that, can you just from a high level talk about.
Doug Lewin (21:24.268)
demand side strategies, the difference between efficiency and demand response. and again, Aurora does work all over the world. know you're, mainly focused on the U S but you have background experience working other places, just a little bit of kind of where Texas is on the spectrum of different states or regions around the world in terms of our demand side achievements.
Olivier Beaufils (21:48.854)
Definitely. And maybe I'll start with the last part of your question first. You know, let's recognize this. I mean, I've lived in Texas for probably 12 years now. Texas is a supply state, right? We like to produce energy and focused on that more so than anything else. There's a lot of good historical reasons for it. And that's sort of, that's where we are on the spectrum of demand side views is we're sort of the extreme on the supply side. I'm sure you could find examples are even more extreme, right? So I'm not.
Olivier Beaufils (22:18.158)
I'm not claiming we're the most extreme, but we're definitely on one end. And then you have countries, historically, that haven't been energy producers, right? A lot of countries that have to import a lot of energy. Think about island nations in particular, but other countries in Europe that don't have a lot of oil and gas productions, for example, that are very much demand oriented and are trying to make sure they're not consuming more than they need because they have to pay for every single.
Olivier Beaufils (22:45.73)
gallon of oil or megawatt hour of electricity they consume. I think that's the overall spectrum. So what that means is there's a lot we could do that's already been done elsewhere that obviously we haven't really put in place in Texas. Now, why haven't we done it? Why should we do it? Well, energy has been cheap in Texas. And so it's obviously less of an incentive to consume less of it. But we're now facing a situation where demand's growing pretty rapidly from industrial load, right?
Olivier Beaufils (23:15.098)
We're building LNG on the coast. The oil and gas activities in the Permian Basin are electrifying. It's not just data centers, but it is also data centers. And so we need to bring a lot of supply on the grid. And right now we have a concern that we've had since Yuri that clearly highlighted that concerns that maybe we don't have enough, especially when the system is tight. And so think it's a great time to start looking at how much progress we can do on the demand side in a way to unlock some of the supply and
Olivier Beaufils (23:42.892)
that is going to be challenging to build when all this demand is going to show up. So now what can you do on the demand side? There's sort of two main approaches. One is what if we have more efficient appliances, more efficient homes, so we don't need as much energy to cool them or as much energy to heat them up, depending on which season you're in. And so the demand doesn't grow as much with population. Every time someone moves to Texas, maybe it doesn't add as much electricity demand.
Olivier Beaufils (24:12.716)
And so in this specific study, we analyze heat pumps in the winter and replacing resistance heat with heat pumps that are much more efficient. And we can dive into this in a second. The other side is demand response. So that's totally different. It's not about efficiency. It's about when the system is really tight or some, you know, demand some customers able to turn off, save on energy to reduce the demand of the overall system. And especially when you have large demands, you
Olivier Beaufils (24:42.638)
For example, data centers is obviously a big part of the discussion, but industrial has been doing that for decades in Texas, right? This is not new. have industrials that are able to disconnect from the grid when the prices are really high. But also why not residential customers, right? That would be able to turn up their thermostat or turn down their thermostat depending on the season again, when demand is high and so reduce their consumption. And so those are really the two mechanism. And so we've explored both on the study.
Doug Lewin (25:10.126)
Yeah, something I've been talking about for a long time is like the large customers in the state have the ability to save on their bills from demand response. We need to make those opportunities available to customers. And this is an area where ERCOT is really taking some strides. They're really, they've made in their own words, residential demand response, a corporate priority for 2025, which is good to see. So this is getting talked about a lot more and Olivia, think this is a really, really important point.
Doug Lewin (25:38.126)
I recorded a podcast a couple of weeks ago with Tyler Norris from Duke University. He and his colleagues at Duke had done a study that looked at data center flexibility. If 1 % of the energy use of data centers were to be flexible, then 15 gigawatts of data center capacity could come online, or demand rather, without adding additional supply capacity. And that's just looking at the data center flexibility.
Doug Lewin (26:04.216)
Then you start looking at if you could bring in residential flexibility as well, because for those that didn't listen to that Tyler Norris podcast, and you should go listen to it, it was a really, really good one. But 90 % of the hours of the year, we've got a 30 % buffer before you get to the ancillary services and the various backups that are there. So there's a lot of extra room in the system to increase the megawatt hours, the gigawatt hours, the actual energy consumption.
Doug Lewin (26:33.524)
If we can control that peak, that's really where like the cost of the system start to spiral is when you drive that peak number up. We could get into talking about the slides, but if you want to add anything or correct anything or whatever I just said, feel free.
Olivier Beaufils (26:49.056)
No, that study was very timely. I'd say it's the heart of the debate now of how we can accommodate all of these data centers' demand growth, right? And I think what we're hearing is data center developers, operators don't necessarily want to get in the business of having all this backup generation firing it on or the disruption that it would take to turn off the data centers so they generally don't want that. And so it's a complicated problem, but it is a big part of the...
Olivier Beaufils (27:17.262)
equation here, because if you cannot have any sort of response on that data center load, then the amount of capacity you have to build to serve that is really large and is probably not something we can do quickly. Right? That's the issue. Timing is of the essence here. And there is a big uncertainty about our ability to serve that much demand that quickly. So I say very timely conversation and so very, very helpful study to be part of the debate.
Doug Lewin (27:45.75)
Yeah. And so for those that are watching on YouTube, you're seeing a slide. Well, again, we'll describe it for the listeners, but what we're looking at is the growth rate. And the first thing I want to point out on this slide, Texas peak demand grew at about 1 % from 2010 through a little beyond 2020. think Olivia, that goes like 2021 maybe. And then the last three or four years, we've seen 5.2 % growth rate. Now, if that were to be sustained over a decade,
Doug Lewin (28:15.33)
That would make the 2020s the largest growth we have had since the 1960s. That's based on national. I should probably go back and look at the Texas, but nationally there hasn't been growth on the electric system of 5 % or higher since the 1960s. And again, that 5.2 % is historical. That is the last three or four years. So there's a lot of discussion around, my goodness, this load growth is coming or is it coming? There's a lot of debate about how much is coming and how much isn't.
Doug Lewin (28:44.334)
But we've already seen load growth within Texas, as you said, driven by economic development, increasing population, data centers, heat, a lot of different factors. You guys put a map on here that shows the top five areas in the country for data center load. And Texas is what, fourth of those. But as far as actual energy consumption.
Doug Lewin (29:11.822)
We're second in the country. And if I'm doing my math right, and I think I am, that's roughly 5 % of our energy consumption in the state already. Because we used in 2024, I think it was 460 terawatt hours. This is 2023 data you have on here. But whatever, call it something like 440, 450, and 22 terawatt hours used by data centers already. So Texas already has a lot of data center load. We know there is the potential of a lot more
Doug Lewin (29:41.23)
coming. This is kind of a context slide and discussion we're having right now to kind of set the stage for why this demand side management on the residential side is so important. Anything else you want to say as far as that contextual scene setting here as we get into talking about the demand side?
Olivier Beaufils (29:57.804)
No, no, I think maybe just a note, right? Like I don't want to repeat myself too much, but the Aura Central I would qualify maybe a bit conservative, but probably closer to reality than the blue line, which is the very bullish aircraft forecast. Reality is probably in between the two, but if I had to guess, I'll probably be closer to the orange line.
Doug Lewin (30:16.066)
I think I agree with that. I think part of that is because, and this is why it's like we get into these, it's a certain point, becomes a little bit angels dancing on it ahead of a pin discussion, you know, debate, but it's not because you have to make planning decisions. I understand the importance of it, but I think so much of this is it's impossible to answer because it depends on the decisions we make, right? Right. It depends on the decisions the policy makers make. It depends on program design of demand response and energy efficiency. If we actually
Olivier Beaufils (30:36.579)
That's
Doug Lewin (30:44.888)
do the things that you're talking about in this study and implement, and what Tyler Norris was talking about, if we have demand flexibility for data centers and we really go big on residential demand response, we might actually see a line much closer to the Aurora, but we will also see a lot more energy sales, which is kind of interesting because a lot of the folks in the Senate and a lot of other places that are so worried about thermal generation, the best thing you could probably do would be to increase
Doug Lewin (31:14.798)
energy sales from data centers and reduce the peaks. So they're actually selling more power throughout the year. And I think that which of these lines are we going to go all the way up to 150 gigawatts, we're going to stay around 100, really kind of depends on some of the policy decisions that are made from here, right?
Olivier Beaufils (31:33.27)
That's right. I think that's a really important component of this equation is all the decisions we make, all the decisions the legislature makes are going to impact the demand forecast, the supply, the economics of those projects. so, all of these are moving targets.
Doug Lewin (31:48.876)
All right, I want to look at the impact of demand response on prices. So similarly to the way renewables put downward pressure on prices, demand response can too. Do you want to just talk through the slide I've got on the screen right now about the downward pressure demand response can put on prices?
Olivier Beaufils (32:08.568)
So we ran the model similar to what we've done on the solar and wind side, but instead this time we focused on demand responding to high prices. And we made, again, we made conservative assumptions on how much residential response we would get, how much data center sort of large industrial load response we would get. And all the assumptions are in previous slides in the study. And so look that when those assets would decide to
Olivier Beaufils (32:38.166)
essentially turn off, which is not really what happens, right? People don't just turn off facilities. They're generally firing up their backup generation. Or in the case of residential consumers, they might be not running their dishwasher or increasing their thermostat by a few degrees. So we looked at these responses and then we rerun the model with those price responsive demand and looked at how prices would settle according to me. And what happens is because you are saving
Olivier Beaufils (33:08.014)
hundreds of dollars by demand being lower during the tightest hours of the year, mostly over the summer. You are in effect saving a significant amount of money on average. And you can see on the slide, you know, two to $3 per megawatt hour, which is significant, right? And in a market that's clearing at 30, $40 per megawatt hour on average. most of the savings happen when prices are four or $5,000 per megawatt hour. So you're.
Olivier Beaufils (33:37.898)
essentially hedging against those most expensive hours.
Doug Lewin (33:41.12)
And then again, just like you did on the renewables, when you looked at what does this actually mean for individual customers? And one of my favorite things about this, I know you're going to talk about it, but I'm just going to call it out because I really, think this result is kind of amazing. And I talk about this all the time, but I think it's hard for people to understand. It's just like adding any resource to the market. Demand response is a resource. And even if you as a consumer are like, don't touch my thermostat, I'm not volunteering for any of this, I'm out. I don't want to.
Doug Lewin (34:10.19)
deal with this in the summertime, even if you're one of those customers and that's okay, I don't judge you, right? It's fine. It's okay. If even a subset of customers participates and you don't, your bills go down too. Now, the person that participates is gonna get more bill savings because they're participating, but you found based on the level of demand response you modeled that even if you don't participate, you would get $87 a year in savings on your electric bill
Doug Lewin (34:39.853)
from increased levels of demand response. That's kind of amazing. So speak to that, but talk about some of the broader results that you found as well from demand response.
Olivier Beaufils (34:48.237)
That's exactly right. And maybe we're not very used to collective, you know, sort of impact in this state sometimes. Right. Right. But yeah, exactly. Is the fact that other people are doing it is a benefit to the system. And so it saves on, on your bill, even if you like your house to stay at 68 throughout the summer and not increase your term with that at all. Right. But if you do, if you do increase it, maybe from 68 to, to 74 or 76, then you're going to save even more money.
Olivier Beaufils (35:15.654)
So there's still very much a component of individual choice. Some people just don't want to do it. But if enough people do it, the overall system benefits.
Doug Lewin (35:24.96)
And you found the cost for industrial customers reduce $2 million, roughly 5%. That's again using that example of kind of 100 megabyte industrial customers. So savings across all kinds of different classes of consumers as well.
Olivier Beaufils (35:40.372)
Exactly. And that would be true for a small business as well that's running their AC in a shopping mall.
Doug Lewin (35:46.36)
So let's shift and talk about energy efficiency, something near and dear to my heart. I talk about it all the time. I have, for those that might be confused of what resistance heating is, I'll give the most basic definition right now. But the very first article I ever wrote on Substack, was during or maybe just after, I think it was actually during Winter Storm Elliot, was about resistance heat. I've done a couple of different podcasts on it as well. We'll link to all those in the show notes.
Doug Lewin (36:11.726)
But resistance heat is literally the same technology as what is in a hairdryer or a toaster oven. It's a coil that heats up and it is sized for a home or an apartment. And so you guys looked at this heat pumps require 64 % less energy to heat a home during the winter relative to traditional resistance heat. So heat pumps are technology that can replace resistance heat and have really large impacts for grid reliability and for savings for customers.
Doug Lewin (36:40.984)
So before we get into the reliability and savings, do you want to just talk through this slide of what the actual impact in demand and load are when you shift from resistance heat to heat pumps?
Olivier Beaufils (36:53.484)
Yeah, absolutely. essentially you are requiring less electricity to heat up your home at different temperatures and the curves sort of the results are not exactly the same depending on what is the outside temperature, right? And I think maybe this is a good opportunity to dispel a sort of a myth that people have about heat pumps. think a lot of people will tell you, yeah, heat pumps are great, but you know, but when it's too cold outside, it doesn't really work really well and, and resistance heat is better.
Olivier Beaufils (37:23.32)
I think that might've been true at some point in time. This is definitely not true anymore. We've done a lot of research on it. These heat pumps are being installed in New England where it definitely gets cold for an extended period of time. Colder than in Texas for sure. And so those heat pumps do perform well, including when it's cold outside, including when it's 10 degrees outside and they perform better than resistance heat. There's a little bit of a curve, but essentially what we're showing on the slide here is that
Olivier Beaufils (37:52.462)
heat pump are about 64 % more efficient than resistance heat when it's 30 degrees outside, which is most of the time in Texas, and the gap's even bigger when the temperature is lower.
Doug Lewin (38:05.518)
That is the key here, right? Because at 30 degrees, like our system probably, we're not going to have big problems on the system. At 11 degrees, you might. And of course, during winter storm Uri in the Dallas-Fort Worth area, the temperature was like negative 2. And so what you start to see is this really, really big gap as the temperature decreases. The percent might not be as much, but the gross amount is still extremely large. And so at a 50 building apartment, you guys were looking at something like an
Doug Lewin (38:33.998)
80 kW reduction. And I can't emphasize this enough. I didn't realize this when I wrote that first column in late 2022. I was mostly talking about homes and you've got on your slide a home that has resistance heat will see a reduction of four or five kW. So I wasn't wrong to talk about homes, but I had a few different folks that work in the energy efficiency industry sort of like literally physically stopped me at a conference and say, you're missing this.
Doug Lewin (39:03.018)
It's the apartments and we're still building apartments with resistance heat. Basically 80 % as my understanding of new apartments that are going in are built with resistance heat as the only heat as the primary heat. they are, they are just using a massive amount of power and the winter is the biggest risk to our system.
Doug Lewin (39:27.436)
So I really appreciate that you guys did this and that you've got this. encourage folks again, if you're watching on YouTube, you're seeing it. If you're listening to the podcast, do go to Aurora's website, download this research and look at it because it really does tell a story. Like we are not going to get out of our winter problems. This is my view until we address the demand side. There's just, you're not going to build your way out of the winter problems. Efficiency is not a nice to have. It's, it is essential. It is a prerequisite for reliability.
Doug Lewin (39:57.846)
Anything you want to add to that or disagree with me, you can.
Olivier Beaufils (40:00.782)
No, I think, I think that was, that was well said. mean, particularly in a system that's growing. think if you were in a situation where you have an oversupply and you have plenty of energy and everything is cheap, maybe you wouldn't be as hard pressed to do it. But in a system that's looking to add gigawatts of data center load, that is growing, where the population is growing quite rapidly and is, is having challenges, maintaining reliability. mean, we've seen under winter storm, Yuri, obviously aircocks.
Olivier Beaufils (40:30.914)
been a lot better since Yuri, we haven't had another episode of Load Shed, but the risk is there. And so I think you should use all of the resources at your disposal. You should build more generation of all the different technologies, but you should also explore what you can do on the demand side to reduce the risk for those tightest system hours.
Doug Lewin (40:48.59)
Yep. And you guys did find some significant cost savings. We're running out of time. We could talk about that briefly, but I do just want to just keeping with that point about winter and what we might see in winter. What you guys found was that if we got to 50 % heat pump adoption and to be clear, that is a big goal. That would take some time to achieve, but you've got this in 2030. So if we really got started in earnest four or five years to rep...
Doug Lewin (41:16.6)
place 50 % of these heat pumps, of the resistance heat with heat pumps, and to really stop digging ourselves deeper into a hole, to stop having apartments that have resistance heat. If we were to get to that 50 % where we might end up in a load shed situation, even in a winter storm Elliott type event, not even winter storm Uri, but in 2030, we could find ourselves if we had an Elliott style event, that was December, 2022.
Doug Lewin (41:44.386)
We could find ourselves in load shed and you found that you could actually reduce that load shed by what about about six gigawatts. Talk through the findings about heat pumps and what they would mean during a winter storm Elliott type of event in 2030.
Olivier Beaufils (42:02.008)
That's right. So it's very important for reliability for a couple of reasons. One is it gives you more flexibility on the demand side. And so this is what you show in what we show in the graph here where you have a situation and this is very much what happened during Elliott's the first day. There was a lot of wind on the system. mean, think about it. You you had a cold front coming in tons of wind works really well and helps the system a lot during the day. Now that we've built all this solar, you also have a lot of solar.
Olivier Beaufils (42:31.554)
You have the batteries, they're able to shift quite a large amount of the solar into the early evening hours. But at some point, you know, the sun sets, the battery capacity is exhausted, right? You've discharged everything you had. And then now the depression is set. So the wind is actually quite low. And so you're in a situation where you have to rely primarily on your thermal generation. And that's also why thermal generation is a key component, a key part of a reliability, right?
Olivier Beaufils (43:01.58)
Certainly I don't want people to take away from this conversation that this is all about renewables. Thermal generation is important and there's a whole conversation on how to make sure we get enough thermal generation on the system that we're having.
Doug Lewin (43:14.004)
It's a system, right? And all the different resources bring different attributes and they all, they all have their place for sure.
Olivier Beaufils (43:20.654)
Exactly. And that thermal generation needs to be there. But this additional lever you can use, right, is if the system is more efficient, if you've looked at more efficient heating appliances, right, so in this case, heat pumps, you can lower your demand to the point where you would avoid Lodgehead even in this really tight system situation. So here it would be overnight when solar generation is not present and wind generation is very low.
Olivier Beaufils (43:50.05)
in that specific event. And so you would really be able to avoid low chat.
Doug Lewin (43:55.052)
Yeah.
Doug Lewin (43:55.322)
And look, I think the other key piece of this is, depending on how cold those temperatures get, you also end up with freeze-offs in the gas fields, right? So gas supply can be a big problem, like during winter storm Uri. So with this scenario, looks like you guys are assuming you do have that thermal generation there. And definitely there have been improvements made on power plant weatherization. So we've seen improvements over the years as the PUC and ERCOT have implemented that.
Doug Lewin (44:24.622)
power plant weatherization, but still there's some level of power plants that drop off of all kinds. Some wind freezes up, gas supply freezes up. And then again, you have the problems in the Permian with freezing. This ends up acting as a kind of a hedge against that, right? It could be low wind. It could be thermal plants going offline. It could be gas supply freezing up. It could be a combination of all those things, but every increment of energy efficiency you have,
Doug Lewin (44:52.27)
either gets you closer to avoiding load shed altogether, or if God forbid, there is load shed minimizing the impact of how long that lasts and how many people it impacts. And this slide really does kind of tell that story. again, pictures worth a thousand words. I encourage folks to go check this one out. Anything else you want to say about this one, Olivier?
Olivier Beaufils (45:14.528)
I think the only thing I would add is just basically every technology has a role to play here and they all play, they all have different attributes and bring something different to the system. And I think something that's probably important to even from just an economic theory standpoint, right? Is that it is better to incentivize and reward the attributes we that are desirable in the system rather than try to punish certain technologies for having attributes that we don't like.
Olivier Beaufils (45:44.236)
Right. So it's, think this is how we create the incentives for a new generation to come in. so for example, if we want more demand response, we should find a way to reward industrial for having that demand response. Historically, it was for CP, right? What was rewarding demand from, from not showing up in the title system hours. You know, that program might need to be modified now to encourage the right kind of behavior. But again, you need to find a way to reward.
Olivier Beaufils (46:11.608)
the behavior you want in the market and then let the market decide what is the best solution, what is the best way to respond to that. I think it's the same story for energy efficiency. And there's a personal incentive of you see of saving money, but right now, sometimes it's difficult to justify the investment. So there's additional levers that could be pulled, but it's also the same story for thermal generation. If we want to have thermal generation with certain characteristics, we should
Olivier Beaufils (46:37.09)
find a mechanism to reward that to the point where we have the system we want.
Doug Lewin (46:41.942)
Yep. Okay. Very last one, because I know we're over time here, but I just want to make the point that like one of the beautiful things about energy efficiency is that it is a way to increase system reliability and resiliency and all those things. There aren't many ways you can do that that also save consumers money. So we talked about how demand response does that. And again, if folks go to Aurora's website, get the study, you can see there's a very similar graph for the demand response impacts on reliability.
Doug Lewin (47:11.064)
check that out for sure. On the energy efficiency side, you found that heat pumps actually can save the average household just about what, 50 % on their heating costs over $400 a year. So not only, I mean, it's just, it's still just baffles me Olivier that we have not as a state four years after Yuri really leaned into this. again, it's a short list of things that you can implement that.
Doug Lewin (47:38.794)
increase reliability by this much and put money back into consumers' pockets. Just give us the takeaway on the dollar savings to consumers from energy efficiency, and then I'll give you a chance to share any final thoughts, and we'll end there.
Olivier Beaufils (47:52.974)
Yeah. I mean, it's the same logic as, what we've seen before, but I think the number here is pretty striking, right? It's $400 a year you're saving. Right. Before we were talking about $200 increase in the case of wind and solar or $200 decrease in the case of demand response. But here you were at double the impact. So this is very impactful. If you're switching from resistance heating to heat pumps.
Olivier Beaufils (48:19.0)
for an average house, your energy consumption will decrease significantly and you'll save a lot of money, especially during those winter storm conditions. And so I think it's something that it will pay off pretty quickly, especially if you can get some tax credits on installing the equipment. But even if we can't, there's really an economic proposition there. But then on top of that, you do generate indirect savings for everyone else and for other customers. So people will benefit from these.
Olivier Beaufils (48:46.872)
programs from higher adoption of these technologies in general.
Doug Lewin (48:50.702)
The 424 is for somebody that takes resistance heat out of their house and replaces it with a heat pump, right? This is for the family. That's not the indirect, right? Okay, okay. Yeah, yeah. So look, if you've got resistance heat in your house, particularly if your central AC, gas furnace, whatever kind of system you have is nearing the end of its life and you're looking at a replacement, look at these heat pumps because there are significant energy savings associated with that. And...
Doug Lewin (49:18.016)
If this is the kind of thing that makes you feel good, you'll also be doing a really good thing for the grid. Olivier, again, can't thank you enough for the research you guys have done. Like you said at the beginning, it's not about an agenda. It's making sure that people have the information so they can make good decisions. You've made a great contribution here. Is there anything else you want to add? Anything I didn't ask that I should have asked you?
Olivier Beaufils (49:40.558)
No, I mean, obviously there's a broader context with the number of bills that are being discussed in the legislature. And I think we could probably spend another entire podcast session discussing the merits and the pros and cons of each of these proposals. But I would say in general, what we tend to see in the modeling is that a solution with more of everything tends to be both cheaper and more reliable.
Olivier Beaufils (50:07.426)
than trying to favor certain technologies over others or impose costs on existing technologies to make them pay for certain attributes that we think are not desirable in general. And so, so I think it may be, it's a bit cliche, but like when we run the modeling, we tend to see that an all of the above approach is a beneficial for reliability and it is also a benefit to consumers. So there's a lot of work that still needs to be done around the Texas power market, particularly around
Olivier Beaufils (50:36.2)
Texas Energy Fund, some of the dispositions are already in DRRS is another one that needs to come in and ERCOT needs to design. I think thinking about it in that way is definitely yielding better outcomes when we run the modeling and the analysis than some of these bills are trying to impose more costs on certain technologies.
Doug Lewin (50:57.218)
Yeah, I think thinking about it from a frame of like, want to have abundant and reliable and affordable power. Texas is the economic engine of the United States and in many cases of the world, we have to have more power to fuel our economy. So I think that's well said, like more of everything. And I think for folks, there's some folks that don't want to see more gas peekers, we're going to get more gas peekers. There's some folks that don't want to see solar and storage. We need that solar and storage.
Doug Lewin (51:26.818)
wind as well. These are all resources that contribute to reliability. They keep costs lower than they otherwise would be. And I certainly hope that's the direction policymakers and stakeholders and regulators and grid operators and everybody wants to go. I agree. Sort of more of everything is I think we're hopefully where we're headed, including more energy efficiency and more demand response. It's a, you know, kind of an irony or something, more of less.
Doug Lewin (51:56.398)
If that makes any sense. All right, Olivia.
Olivier Beaufils (51:58.574)
Right, exactly.
Olivier Beaufils (52:01.174)
really appreciate you. for taking the time. I encourage everybody to check out.
Doug Lewin (52:03.246)
time, do encourage you to check out the
Doug Lewin (52:05.366)
studies. We'll have links to them in the show notes. And Aurora also has, by the way, a great podcast, which I've learned a lot from over the years. It's called, is it Aurora Unplugged or something?
Olivier Beaufils (52:15.064)
like that I think is the name. Energy and plagues almost. Yeah. Aura energy and plagues and thanks for mentioning it.
Doug Lewin (52:21.55)
put a
Doug Lewin (52:21.71)
link to that as well. I mean, you guys, again, you guys kind of cover the world. it's fascinating to compare and contrast. My head's always kind of in ERCOT, but to listen to some of the stuff you've done in Europe and Australia, different markets around the world has been a great learning experience for me. Thanks, Olivier. Appreciate you.
Olivier Beaufils (52:38.136)
For sure. Well, thank you, Doug, for having me and for being able to discuss these topics with you. And we're all big fan of your work in AirCut as well. And always great to read your posts and follow exactly what's happening and live in the market. Thank you.
Doug Lewin (52:51.47)
Appreciate it. Thanks, Olivier.
Doug Lewin (52:55.298)
Thank you for listening to the Energy Capital Podcast. I hope you enjoyed the episode. If you did, please like, rate and review wherever you listen to your podcasts. Until next time, have a great day.
