Batch zero stops being theoretical on July 11. That is the day ERCOT’s rule for connecting large new customers takes effect. The new policy replaces a process that involved studying each giant load independently, then ordering restudies when new giant loads joined the queue, leaving projects stuck in a serial loop. Two prior episodes of this show traced how the new rule was designed. This one asks the people who connect the load what to fix before the next round.

Already projects are sitting in the interconnection queue as new regulatory deadlines loom. ERCOT figures show more than 445 gigawatts of large loads in the process, and the rule sorts them into base load, studied load, and excluded load. Developers have until July 10 and July 24 to meet certain filing deadlines, and the full batch study is targeted for early April. The policy also shifts more of the analysis from individual utilities onto ERCOT.

CenterPoint has been connecting large loads in Houston for decades. That experience drives a question the design phase mostly deferred: does a 75-megawatt cutoff for loads to participate in the program fit the manufacturing and industrial loads that move at the speed of business?

On this episode of the Energy Capital Podcast, Joshua Rhodes talks with Caitlin Smith, chair of ERCOT’s Technical Advisory Committee and senior vice president at Jupiter Power, and Jason Ryan, executive vice president of regulatory services and government affairs at CenterPoint Energy. Smith walks through how stakeholders developed the rules on a compressed timeline. Ryan presses the forward question of whether the 75-megawatt threshold and an annual batch process fit the loads Houston routinely connects.

Ryan’s concern is timing. When the batch becomes “the long pole in the tent,” he says, developers with real projects start to walk. The conversation works through:

* WL-PUN and PCLR, the withdrawal-limited private-use-network and provisional controllable-load resource programs ERCOT is repurposing to fit more load onto the current grid.

* The 75-megawatt cutoff, why Ryan questions whether mid-sized manufacturing loads belong in the batch at all, and the risk of projects sizing themselves at 74.9 to stay out.

* Non-firm service and reliability, how a load that agrees to curtail differs from the century-old obligation to serve, and what testing CenterPoint needs before it trusts the switch.

* What is permanent versus triage, which parts of batch zero survive into batch one and beyond as the Texas Legislature returns next year.

New to the batch zero mini-series? Start with Eric Goff on how batch zero took shape and Tiffany Wu on the mechanics.

How ERCOT sets the threshold and batch cadence will determine which loads get power on their own timeline and which wait for the next cycle.

Timestamps:

* 00:00 - Introductions: Caitlin Smith and Jason Ryan

* 02:43 - What Batch Zero is and why ERCOT needs it now

* 05:14 - Houston's diverse large loads, not just data centers

* 08:13 - Timeline: the July 11 effective date and key deadlines

* 10:44 - Base load, studied load, excluded load: winners and losers

* 12:55 - Inside TAC: compromises, new stakeholders, and fairness

* 16:10 - Does the queue mean a transmission build-out?

* 19:01 - The real number: CenterPoint's 40 to 50 GW prediction

* 23:18 - New constructs: WL-PUN and PCLR explained

* 28:13 - Non-firm service, reliability, and trusting curtailment

* 32:01 - Tracking success: what is permanent versus triage

* 36:07 - The 75-megawatt threshold and how often to run a batch

* 43:38 - Data centers, the final timeline, and what comes next

Resources:

People & Organizations

* Joshua Rhodes (LinkedIn)

* Webber Energy Group (Website - LinkedIn)

* IdeaSmiths (Website - LinkedIn)

* Caitlin Smith (LinkedIn)

* Jupiter Power (Website - LinkedIn)

* Jason Ryan (LinkedIn)

* CenterPoint Energy (Website - LinkedIn)

* ERCOT (Large Load Integration)

Company & Industry News

* ERCOT Again Revising Large Load Interconnection Process

Books & Articles Discussed

* Texas Senate Bill 6, 89th Legislature

* PGRR145, Batch Zero Process for Large Load Interconnections

Related Podcasts by Energy Capital

* Batch Zero, Explained with Tiffany Wu

* How Texas plans to serve ‘infinite demand’

* Open Season vs. Batch Zero with Travis Kavulla

Transcript:

Joshua Rhodes: Hey everyone, and welcome to another episode of the Energy Capital Podcast. I’m really excited today to have not one but two guests to talk about kind of what’s going on in the ERCOT Batch Zero process and kind of how that may continue to play out. So today on the podcast, we’ve got Caitlin Smith. Caitlin has a BA in econ from University of Texas and a JD Law from Penn State. She’s a policy consultant for CLEAResult for going on counsel at Jewell & Associates. She’s a vice president of AB Power Advisors and is currently the Senior Vice President for Federal and Regulatory Affairs at Jupiter Power, one of the largest pure play energy storage companies in the US. But she also is the current chair of ERCOT’s Technical Advisory Committee, the highest committee comprised of stakeholders, which makes recommendations to the ERCOT board. And that’s going to really come in handy today as we talk about one of the biggest policy shifts that’s working its way through the system. We also have Jason Ryan. Jason Ryan has his Bachelor’s of Business Administration and JD from the University of Texas. He was a Global Projects Associate at Baker Botts, managing partner at Ryan Glover LLP. And he’s also the information dominance warfare officer for the US Navy, which I kind of just want to stop and talk about that. If you can, you may not be able to talk about that. But now he’s the executive vice president of regulatory services and government affairs at CenterPoint Energy. Caitlin and Jason, welcome to the Energy Capital Podcast.

Jason Ryan: Thank you for having us.

Caitlin Smith: Thanks, Josh.

Joshua Rhodes: I’m really excited. So I’ve got two lawyers and two government affairs folks here today. So we’ll see how bad I do at managing this. Great. It’s gonna go great. But I know we’re under a bit of a time crunch, so we’ll get started because Caitlin, you’ve got a date for Elmo’s Got Moves. Is that right? Elmo’s Got The Moves. Okay.

Caitlin Smith: Almost got the moves. Almost got the moves. I’m seeing it tonight. I don’t know when this will air, but it’s in Austin Friday in San Antonio Sunday.

Joshua Rhodes: That’s some free advertisement there for almost got moves. But anyway, so we’ll go ahead and get started. And so the arc of this podcast is really I want to kind of catch up with what’s going on with the batch zero process. We’ve done two episodes, which we’ll link in the show notes, one with Eric Goff and one with Tiffany Wu, where we kind of looked at the overall kind of structure of the batch zero process and then with Tiffany got into kind of some of the details. But one of the things that it really was brought out, at least for my knowledge, during those podcasts, was We really had a framework for how things were going, but we hadn’t filled in all the details. And so I was just curious, Caitlin, if you could kind of refresh us on what batch zero is, where it stands, and why do we need it right now?

Caitlin Smith: Sure. And the impetus for me coming on, or one of them was Eric Goff said that the demand for load is infinite. I don’t believe it’s infinite. So I wanted to come on and correct. But maybe Jason thinks it’s infinite. So we could debate that. So batch zero, previously in ERCOT, there was not a uniform process for load interconnect. You know, before, I don’t know, six, seven years ago, nobody was really thinking about. Connecting these large loads when I was consulting, you know, we would have a call about a gigawatt hydrogen load or something that wanted to come online. This was starting in 2020. ERCOT hadn’t really heard about it or contemplated it at that point. So in the last five or six years, we had a major change to the system, which was ERCOT was actually seeing these applications for very large loads to connect and a lot of them What happened then was for the utilities, it was just either too much to process or they didn’t really know how to process it. Jason can correct me. And I think the other thing was there was not a uniform way amongst utilities on how to process these studies. And so they would study a load, another load would come on in their area, or maybe not even in their area. And ERCOT would say, No, we have to restudy. So people were getting caught in this infinite loop. So we are changing from that serialized process to a cluster or a batch, as we’re calling it, process where you can study a whole group of loads to make sure the system can accommodate the whole amount or the whole allocated amount of it at once and you can have a clear study. The other thing that does is really shift more responsibility to ERCOT. Before this was really each TSP. Was performing these studies and now there is a much larger ERCO component. And so batch zero is our way from transitioning from the status quo to the batch

Joshua Rhodes: process. Got it. And Jason, I guess the old process, like large loads were coming to transmission service providers like CenterPoint. Can you give us a feel for like when did the problem start to feel intractable in terms of like you going from having maybe one load at a time to dozens or hundreds of loads at a time? Can you give us a feel for when that started to come along to push this new process or to push talking about getting a new process?

Jason Ryan: Yeah, and so maybe I can answer it from a general perspective and then I can answer it from my company’s perspective. Sure. Because I think those timelines are a little bit different or the experiences are a little bit different. Between two years, you know, eighteen months, two years ago, I would say it started to become clear in many parts of ERCOT that something had to change. Okay. And yeah, we started talking about the batch process towards the end of last year and obviously the process has played out, got built this year and we now have more certainty on exactly what that looks like. And so I think the batch process has come together relatively quickly. Yeah. Once that problem was identified. I’ll speak to my company’s perspective. Yeah, in Houston we’ve been connecting large loads to the grid since before it was cool before everybody was talking about it. So we didn’t have the same challenges that you saw in other parts of the state. And in fact the current ERCOT study process for us continues to be very efficient, even through this transition into a batch way of doing things. Okay. Because like I said, we’ve had large customers on the grid down here in Houston for a long time, both transmission and distribution system, even we have large loads on our distribution system down here. The pace has obviously picked up. The size has expanded and gotten bigger. But for us, it’s business as usual at a faster pace. Okay.

Caitlin Smith: I think there’s a big some policy questions come into play for CenterPoint’s territory. I think a lot of those are things that need to be in the territory, right? Close to the shift channel or things that are kind of integral manufacturing businesses. That’s right. And so you also have these data centers or Bitcoin mines that come on maybe in other areas of the state that Some people maybe think don’t need to be there. Or maybe they can locate in a bunch of different places, right? They don’t necessarily need to be in the one spot. So I think not all of these loads have the same characterizations. And I think we tend to right now be solving for the data centers, which is something that’s new for us.

Jason Ryan: Yeah, I think that’s right. I’m glad you raised that too, Caitlin, because you we’ve got the largest petrochemical complex in the world here at Houston, the largest medical center on the planet, a significant amount of advanced manufacturing. And so it’s important that I share that view of a utility in a area that has a very diverse set of drivers of growth with large loads. We also have data centers, but I think you’re right. We shouldn’t over rotate on data centers and have unintended consequences to manufacturing jobs and energy and simply growth of populations that we have here in Houston.

Joshua Rhodes: totally. And I and I want to get to this new concept of well, a borrowed concept of the WL-PUN in terms of like getting through the batch zero process. CenterPoints obviously has connects a lot of the private use networks already. So you already have a lot of experience with that. I want to get to that in a little bit later, but first I want to like setting the timeline. So we’re recording this in kind of like late June. And at this point, the batch zero process has made it its way through TAC. It’s been approved by the ERCOT board. The Public Utility Commission accelerated its approval. Of batch zero and I think I got an email from ERCOT yesterday saying that the process is gonna start on July eleventh, if I’m correct, or something to that effect. Did I get that timerine right, Caitlin?

Caitlin Smith: Yes, that’s correct. I believe the rule becomes effective the eleventh, but there’s a lot of deadlines between the eleventh and the twenty fourth. Or the tenth and the twenty fourth.

Joshua Rhodes: Okay. Got it. All right. So whenever the rule becomes effective, well now that the batch zero is approved, what’s no longer theoretical here? Like what’s the process going to start to look like on July eleventh or maybe the twelfth the next day?

Caitlin Smith: That may be a better question for Jason. You know, I think a really different thing about this as opposed to some of the other policies stakeholders pass is you’re kind of immediately left with winners, kind of winners and losers. Or maybe not losers, but maybe you have to wait a little while. And so we have people who will be firm load in batch zero, right? They’re getting their studies are valid. They don’t need to be redone. They’re getting The allocation of everything they wanted, things that need to be studied in batch zero. So they’re good to go into batch zero. And then they will find out what their allocation is. And then we have people who are out of the batch who will have to wait till batch one. And so we know the criteria for that. I think by August, we will know who’s in those. The requirements are due. I think it’s mostly. The loads have to get it to the TSP by the tenth and then the TSP has to kind of affirm that to ERCOT by the twenty fourth. So we’ll know who those people are or I don’t know if that will be disclosed, but we’ll know the amount and they’ll know who they are in August.

Joshua Rhodes: So is the process, Jason, is it like are the large loads still gonna come to you first? Are they still gonna come to the TSPs first and then there’ll be a handoff to ERCOT or something like that for the big study? Yeah.

Caitlin Smith: Yeah.

Jason Ryan: That’s right. So, you know, what we’ve got now is we’ve got certainty over timeline and we’ve got certainty over requirements. In a couple of weeks you’ll know which loads are fall into what bucket. You’re either base load or your studied load, or you’re not either one. Then from there it’s the new part is ERCOT spap study, which we the utility would have provided the studies to ERCOT. ERCOT will continue to ask questions back and forth, right? We’ll provide answers, they’ll ask questions. That’s very typical even today. So the different thing is the batch study that Caitlin has laid out for us and you know, that is likely to take the balance of the year into next year. Early April is the target for the batch to be complete. And so there’s a lot of certainty now that we’ve got these rules in place, as opposed to January, February, March of this year when it was we were making the airplane as we were flying it. So a lot of uncertainty. You now have that certainty.

Caitlin Smith: Agree with that. We didn’t really start using the word batch, as Jason said, until the end of last year. And even though we did this on, I would say it was a very aggressive and kind of phenomenal timeline, it caused a lot of confusion, right? If you’re sitting there in January and you have some load projects and you just heard this word batch two months ago, but you’ve had loads in process for a couple of years. That really freaks you out, right? Even from January to today is sort of a long time to wait and have that certainty.

Joshua Rhodes: When we talked to Pablo Vegas on the podcast, I you know, he mentioned that this process was ongoing and we probably weren’t gonna get it exactly right, but it was gonna be basically building the plane kind of as we were going. I guess like the first step of that process, Caitlin, I guess went through TAC, right? Went through the technical advisory committee that you’re chairing, or at least that was kind of the process before it got handed off to the ERCOT board. And you had to take a pretty big contentious, you know, large load problem with a bunch of different stakeholders with a bunch of different wants and needs. And turn it into like a streamlined or a process. In that process, what do you felt like was the biggest compromise like y’all had to make or that everyone had to make?

Caitlin Smith: Yeah, I’ve been thinking about this because I think I may have used the word contentious or contested before. I don’t know that I would characterize this as extremely contentious, but I would characterize it as extremely important. Yeah. There was a lot of money on the line and extremely large, right? We’re sort of changing the entire system of load integration onto the grid. And we had a lot of new players. So as you said at the beginning, I chair what’s called the technical advisory committee. I can’t speak on behalf of a group here, but I have a lot of experience chairing that committee. And that committee is comprised of stakeholders, you know, from every segment, generators and Jason segment and municipals and everything. So I sit on the Committee for Jupiter Power. I happen to chair it. So I facilitate the meetings. These low developers have not been part of the process to date. So there was a lot of learning for them, right? They’re not even really voting members yet. And so they were sort of coming in from the outside to a process like this. So there’s a lot of learning. As I said, this was a little bit unusual and they’re kind of immediate commercial winners and losers. And so that gets contentious, but I don’t know necessarily that the policy things we discussed were contentious, but it was hard to differentiate sort of a policy cut versus what is a fairness issue. Is it appropriate to pick a winner and a loser? Because I think there were some issues that came up that were sort of fairness issues that people were very sympathetic to. You know, we talked a little bit with Jason already about the differences between the loads that an IOU might have, right? They might have a lot of data centers in their territory or they might have a lot of critical industry and manufacturing in their territory. And so I think there are a lot of specific to them things that load developers brought to the table. And it was sort of hard to kind of pick what needs to be in batch zero, what’s fair or not, that kind of thing.

Joshua Rhodes: Yeah, I think another word that maybe got overused a bit was when we were talking about the large load queue, we I started talking about it like there was a process. Like, you know, there was no real queue. It’s kind of more like a list, the thing, you know, and with the individual TSPs where things would kind of work their way through. That makes sense. I mean, it’s like if you’re staring down kind of the barrel of like, I think the latest number is like four hundred and forty-five gigawatts of large loads in the process and like ninety percent of that being data centers, it’s like it’s either create a process or just get paralyzed by the fact that you know, you can’t move forward. So it’s like, yeah, hats off to y’all for coming up with something so quick. That was pretty impressive. And I think the rest of the country is watching, particularly other other grids like maybe PJM and stuff. So Jason, in this process, ERCOT’s gonna allocate transmission capacity. It’s like, so we’re gonna go through this process, projects are gonna get either their full allocation or they’re gonna get like a staggered allocation. But that means that utilities actually have to build the stuff and connect. Given how much is going through the queue, is that gonna mean a like a substantial acceleration of transmission expansion in the next five or six years?

Jason Ryan: Well, not to connect to these batch zero loads, most likely. Okay. But you then have to replenish the capacity on the grid. Right. And so if you assume that we will continue to do batches, you know, if we run out of capacity on the current grid in batch zero or batch one, then obviously you have to build more capacity, which gets to maybe Eric’s point of the infinite growth. Whether it’s infinite or not, it certainly is relentless right now. And you see it’s a fair amount of transmission projects being constructed today with early twenty thirties in service dates. Yeah. I think you will continue to need that infrastructure for the future batches. But if you’re in batch zero and you want energy in twenty eight, twenty nine, you’re gonna have the existing grid that you’re counting on. There obviously will be some upgrades that are needed, but you’re not talking the hundreds of miles of transmission line. That’s gonna serve growth end of this decade, early next decade. You know, it is interesting. I think I mentioned this to you, that things will slow down for us so that it’s an unintended consequence of the batch process is that we won’t be able to move as quickly as we otherwise would. I think that will work itself out of the batch process over time, I suspect.

Joshua Rhodes: But the new classification framework, it creates base load, study load, and excluded load. Do we know how much is going to s at least start out in each one of these buckets? And do we have a feel for like what might make it out the other end of the batch zero process?

Caitlin Smith: I don’t have one I’m willing to commit to. I think we’ve heard various numbers from ERPA along the way, but sort of big ranges. And I think a big part of this process is for baseload and for studied load, there are requirements, eligibility requirements. So I think we know based on by all accounts ERCOD and the TSPs have been doing a really good job on their end of what studies are done to say who’s firm load, who studied load. But there are these criteria now for the loads to meet, right? Certain development criteria, a significant financial security posting. So I think we don’t know how many of those loads are going to meet those criteria or choose to meet those criteria. And if they get allocated less than what their project is, do they then drop out? I think what’s been hard about load and the data center load in particular is just the lack of historical context. So certainly for batch zero, we don’t have any historical precedent to say, well, about 80% of the loads, you know, end up hosting their financial security. And with the load forecast, we don’t have kind of historical data to say, well, this is a huge number, but about 60% of these aren’t real or 30% are real. We just don’t have that yet.

Joshua Rhodes: Yeah, I guess we had that for the generator interconnection queue. I remember doing a kind of a study, kind of a longitudinal, we could figure out what percentage would make it between to each stage and f at the other end. But yeah, like I said, I don’t think we even really had a queue. We called it a queue, but we didn’t really have one for large loads. Jason, are you able to offer up any numbers for CenterPoint? Like what numbers are y’all looking at when it comes to kind of what’s trying to get in bash zero at these tranches?

Jason Ryan: Yeah, so I you know, the number if you unpack what ERCOT released earlier this year informs my view of what the ultimate number will be. Okay. So of the more than four hundred gigawatt number, we were a little more than four gigawatts of that. And we have since then had additional load that I believe is real come into the batch process. So, you know, that is not Currently my number that I’m predicting. Okay. But yeah, if you want my entire queue, I can see how you can get to hundreds of gigawatts. But yeah, we meet literally every day to go through our list of potential customers. Yeah. And have only submitted the projects that we believe are real. It’s one of the reasons why our review with ERCOT continues to be relatively straightforward and on historic timelines, and we’re not getting bogged down with a lot of unreal projects. As an aside, I do think that to some extent this is a creature of the utilities making. I’m not being critical of the utilities in saying that because we didn’t have any tools to help figure out who’s real or not beyond just our judgment. Yeah. But we do have to exercise our judgment in terms of what we bring forward. And so that’s why you’re not seeing eye popping numbers out of the Houston area because we’re not turning in, you know, our entire queue. That said, I think that we’re about twenty five percent of ERCOT’s load down in Houston today. Yeah. If you assume that I’m growing at roughly the same pace as the rest of the state and that I have about the same amount of available capacity on the system today as other parts of the state, maybe I have a little bit more. I think that drives me to I think realistic number coming out of that zero base load is forty, fifty, something like that. Again, it depends on what the timeline for those projects are. If those projects are needing power in the twenty thirties, then sure, perhaps that’s not an issue. But that’s my prediction. By the time this airs, I’ll probably be proven super wrong. Because we’ll know. But you may be right. But that’s my prediction just based on what we think is real among the customers that we interact with. You know, the utilities are the closest to these customers. Right. We have to do our job. Of vetting these before we just throw them into the machine of ERCOT. And so I do think that I’m not critical of how we got to the batch process. I think it’s needed across the state, but I think when you throw around numbers that are hundreds of gigawatts, we know that that’s not going to happen in this decade, right? But like I said, the utilities were kind of without a process to make the decision on well, who does get to go forward and who doesn’t. So that’s why I think batch is good. But we also have to be more realistic on the numbers that we put out there.

Joshua Rhodes: Well Kaylin, hopefully that helps tack there to figure out kinda maybe what’s going on.

Caitlin Smith: to the number. I’m not figuring out the numbers.

Jason Ryan: Yep. Forty or fifty gigawatts of baseload growth is mind boggling. But I think the grid could actually digest that in a relatively modest period of time. Then we should talk at some point about potential unintended consequences of the batch process because I think there are some that are worth talking about, especially down in Houston where all of our growth is not data center growth.

Caitlin Smith: Is a lot. Yeah.

Joshua Rhodes: totally. Yeah, and we’ll get to that here in a little bit. I did want to so you’ve talked thrown around a few numbers. I mean I think all of them are kind of wags at this point. Yeah, maybe but okay, between zero and in less than infinity, which mathematically is still inf whatever you

Caitlin Smith: They’re not infinity. Should do Twitter game like Russell Gold used to do. You should have people

Joshua Rhodes: we have to bet the price of oil. Yeah, next year. Okay.

Caitlin Smith: Yep, you should have people bet the gigawatts and batch zero.

Joshua Rhodes: Okay, I didn’t do it for batch zero, but I think I did this years ago. But when the numbers were like 100 gigawatts, not 400 gigawatts. And so it’s I’ll be honest. The last study I was hoping for 500, just you know, why not? Okay, but there’s a couple different to get a lot of these large loads in there, ERCOT has come up with a couple new constructs. Well, not new. They’re borrowing existing constructs to kind of help some of these. There’s the PCLR, the provisional controllable load resource, which, you know, essentially will have some firm service, but if the grid can support it, might be able to go above that during certain times, but they may be curtailed down to their firm. And then kind of this WL-PUN, which is a withdrawal limited kind of borrowing from the private use network that we kind of already have in a lot of like the high industrial loads in the CenterPoint region in the Gulf Coast region. So Jason, starting with WL-PUN, can you kind of explain just how a private use network works, like how it interacts with ERCOT and kind of how y’all see that when you’re doing your planning?

Jason Ryan: Sure. So, you know, as you mentioned, it’s not maybe the acronym is new. The concept isn’t super new of having generation sided with load, either literally right next to it or in close proximity to it. And you know, you look at a lot of our historic large load customers, many of them have their own generation. And sometimes they are using that generation for their processes and sometimes they’re selling that back onto the group. So we’re super familiar down here with how that works and how that’s engineered and those customers are super familiar with the economics of it as well. And you know, as it relates to you know, the move to more of that, it doesn’t cause us great concern because of that historic precedent. And as you talk about the controllable load resource, it’s again not super different. Then what we already see, again, the acronyms may be different and maybe everybody doesn’t have the same kind of experience that we have with these large loads that, you know, for various reasons might have to change what they’re doing, change their use of the grid in the moment, or even put extra power back on the grid for a moment. We understand how to build the infrastructure for that and how to take that into account when we’re interconnecting on

Joshua Rhodes: And Caitlin, I is my understanding that some of the acronyms are kind of new, but we’re borrowing from existing constructs that exist. Is my understanding, and please correct me if I’m wrong, that a lot of these new larger loads were actually wanting a new construct, like a point of interconnection netting, like with generation kind of behind the meter. If I got that right or it got it partially right, can you explain like what they were asking for and kind of maybe why we didn’t want to tackle it here with the batch zero process?

Caitlin Smith: I think that’s right. I think these are old concepts. You know, certainly the private use network. CLR, same thing as what I was talking about in the beginning with these loads, us not having seen this really until six years ago. I think the concept of CLR existed, but it wasn’t really being used until these data centers, maybe, I don’t know, four or five years ago. It’s basically a software solution to be able to respond to the grid as fast as a or faster than a generator would be, right? You’re a SCADA-dispatchable load. So it’s still a new concept. And I’ll want Jason to weigh in. I think what we’re talking about maybe since SB6 is more a concept of non firm service. And so these loads say, well actually I don’t need firm service all the time because I’m a CLR. Or because I’m a generator and I’d actually prefer not to have firm service if it increases my speed to market. And I think that that is a new thing, right?

Jason Ryan: Yeah, certainly newer in the electric space. We’ve been doing that forever on the gas utility side of our business though, right? That is how large users of natural gas get connected to a system that has limited capacity only in certain times. Right? So think about in the wintertime when all of us at home are turning the heater on using natural gas, we’re using more capacity. Especially here in Texas, that’s a limited period of time. Right? There are a couple of days in the year. Where we’re consuming a lot of natural gas, that capacity is there. And on a normal day, you know, the other three hundred and sixty days of the year, that gas is available to large users, but they know that they’re curtailable in the tails of the probability curve and they curtail their usage of the system. So it’s perhaps being applied in a different way, but the concept has been around to consume energy for a hundred years, right?

Caitlin Smith: Yeah.

Joshua Rhodes: Yeah, no, totally. I guess like but I guess maybe for like the electricity sector, it is like a bit of a different, you’ve always kind of had the obligation to serve, like four or five nines, you know, reliability. Whether it’s like a PCLR or just a different reliability class of something, I know that’s something I didn’t wasn’t fully aware of that we were having that particular conversation outside of the PCLR construct here in ERCOT. I know they’re doing that in PJM. They’re talking about different levels of reliability. For different customer classes. But like, I guess, Jason, say you got a project going through the batch zero process, they’re signing up as a PCLR, a provisional controllable load resource. Do you trust them enough? I guess right now? Do we have the solutions in place? Like Caitlin said, CLRs have only been around for a little while, and I think mostly have been used by Bitcoin mines, if I’m correct. But what do you need to see to be able to trust that, okay, you need to go down to your hundred megawatt firm limit? I mean, what do you need to see to make sure that your system stays stable?

Jason Ryan: Yeah, so I think it’s fair to say that we’ll have to do a fair amount of testing to make sure that we understand how this works, what impact it may have on our system when we need it to work. Yeah. And to ensure that we engineer a solution that works in all kinds of scenarios, not knowing exactly what the scenario is that would require them to trigger that feature of their site. And so I have no doubt that we will be able to work through that. There are things to work through though, right? We don’t have all the answers as we’re sitting here today. Right. I trust that we will be able to figure out together with those customers. And I suspect the answer is gonna be that you test various scenarios along the way. Not super different, more complicated, but not super different than what we do today with our load management customers. We test it periodically to ensure that they’re able to drop their load on the timeframe that they need to. This is a bigger scale. But it’s not super different than what we do already with some of our other programs. Okay.

Joshua Rhodes: Are you also concerned about one of the things when we saw this happen in PJM a little while ago? So we had a large data center, like a gigawatt worth of like data center load, like trip offline. And that created a lot of like local instabilities in the system. And when I teach electricity markets, I generally used to brush past the frequency going too high because we have like too much generation and not enough demand. And like it’s easy to turn things off. Like this is no big deal. But apparently, maybe is a big deal. And so like as you’re seeing more of these. Larger loads, what are you thinking about in terms of making sure that, you know, that side of a trip is covered? Caitlin Smith (00:30:51) I think batteries can help with that, Josh. Just I love the plug there. That works great. It’s a softball right for you.

Jason Ryan: In addition to batteries, I think again, we’ve had to consider this for a long time down here in Houston. Now again, the so take a large L and G terminal tripping off or, you know, any other kind of re large refinery type load or even some of the large manufacturers that have a significant amount of load, especially if you’re talking on the distribution system, we will need to work through that as the size of these facilities start to become multiple gigawatts, not just a gigawatt. Yeah. And so it is a consideration that we’re working closely with our engineering teams and our customers to ensure that we think through all of those scenarios and think through how we need to design a system to withstand that. By the way, it’s not super different in concept to designing a system to take that into account in terms of loss of large generation. Right? Yeah. All of these large contingencies that happen, you have to plan for and design the system to withstand them.

Joshua Rhodes: Got it. So Caitlin, TAC has handed off like the batch zero process. It’s been again been voted. ERCOT board, public utility commission, it’s getting started. What are y’all looking to track between now and I guess kind of the fall of twenty twenty seven is when this process is like supposed to end up with a plan for the regional planning group to say, Okay, here go build this stuff. What are you tracking between now and then to know whether or not the batch zero process is working? What’s success or what’s an issue?

Caitlin Smith: That’s a good question. I haven’t really thought about what kind of reporting we’ll want. I’ll defer to the other members of TAC. I should have said at the beginning, this really started with ERCOT staff and with commission staff. We did our part as stakeholders. We did, I think, more than our partners as much as we could. This was a highly collaborative process, which is an achievement on a short timeline. But the commission staff and ERCOT staff to their credit has been very hands-on and involved. I don’t think that there’s a world in which we say this is a failure, right? It’s what we’re doing. We’re moving to a new process. I think it was necessary. We’ve harped on a couple of the Things that were maybe problems before. Maybe studies were working in some areas, but there’s sort of a lack of transparency, right? Because there was no process. There wasn’t a standard interconnection agreement for load. So you couldn’t go online and see what other people were doing. Okay. You couldn’t go to a dashboard anywhere and see what your status was or if ERCOT was gonna need a restudy because some other load came online. There’s just a lack of transparency that I think needed to be. Remedied. So I think we have a lot more transparency and certainty. I think there are things that we will need to change, but there are probably things we’ll continue to want to change. You know, after batch one and batch two, we sort of always keep working on our rules for the market and for interconnection.

Joshua Rhodes: Yeah, that was gonna be my next question. Is you know, batch zero has often been characterized as kind of a triage of this big large load list or queue or whatever you wanna call it, you know, trying to inject some discipline into this process and like get things moving and get a process. But like as batch zero is kind of a triage process, presumably as we’ve mentioned, there’ll be a batch one, a batch two, a batch three, a batch in, who knows how many of these batches we’ll need. Do you have a feel for like what parts of batch zero should be treated as permanent versus what part of it is just the triage right now, kind of the emergency scaffolding here?

Caitlin Smith: You know, I think it’s a really good framework. I think people are going to have more asks, right? You brought up two of the big ones, the WL Han and the PCLR. Yeah. I think people are going to have more asks on those things. I think more people will start to weigh in, right? The legislature’s back in town next year. I think we’ll hear more voices, as I mentioned before. One of the challenges, but it was Don’t get me wrong, it was very good they were participating, but these load of developers were new to the ERCOT process. Yeah. They’re not new anymore, right? So we’ll continue to hear from them, which I think is good. But with more time, we also know more. I think they’ll just be continue to be more and more asks as we do batch one, batch two, batch three.

Joshua Rhodes: Got it. So kind of similar question to you, Jason. Kind of like we’ve got batch zeros, we’re gonna build the plane as we’re going, but you know, before we get to batch one, like is there anything in particular that’s CenterPoint?

Caitlin Smith: We built the plane. It’s great.

Joshua Rhodes: Well maybe we’re putting seats in the plane. We have engines, maybe. We’re taking off. We haven’t painted it yet, maybe? Something like that.

Caitlin Smith: I think putting seats on it is right.

Jason Ryan: We’re about to have passengers, right? So yeah.

Joshua Rhodes: There are about

Caitlin Smith: But not infinite passengers, a finite amount.

Jason Ryan: That’s fair, yeah, yeah.

Joshua Rhodes: Not infinite passengers. But Jason, so like before batch one gets going, like is there anything in particular like the CenterPoint is already looking to like get or caught the change or fix or to alter?

Jason Ryan: So I think I’ll talk maybe conceptually the things that we should be asking. Yeah. I feel like we should be asking ourselves is seventy-five megawatts the right cutoff to go into a batch? And that’s gonna depend on the answer to the other kind of high level question I think we should ask ourselves. But it’s not uncommon for us to add a hundred megawatt customer down here, especially on the manufacturing side of things, and the timeline to win a project like that whether it’s a new customer or an expanding customer, is not going to line up well with the batch process as we see it today. Okay. What I mean by that is we’ve got customers that have options. They could build a manufacturing facil they could expand their Houston facility or they can expand their Mexico facility. They’ve got contracts and obligations with customers to make stuff. And they are going to make that stuff wherever they can get the power quickest. And they’re not gonna be gigawatts of manufacturing facilities, right? So that’s why I say maybe that seventy five needs to be looked at and have some kind of stratification for what is still large load, but I’m not sure it’s the large load that’s causing the need for the batch. Okay. So I think that’s a question. And then the second question is how often are you going to run a batch? Yeah. And you could design a batch process that works even for manufacturing expansion where you can move at their speed of business, but maybe not if it’s only once a year. Okay. You probably could if it was twice a year. And so as you get past batch zero and you know, one through end, yeah, can we at some point get to the point where the batch process is not the long pole in the tent. If I have available capacity today There is no batch process that’s the long pole in the tent. It’s getting the studies through ERCOT, which is relatively efficient for us down here, and then connecting to the customer at their speed. Right. When you have the batch process be the long pole in the tent, not whether I have the capacity to serve them. Mm-hmm. That’s where I think you have unintended economic development losses. In Texas, we don’t accept those outcomes. I am confident we will figure this out. But I think those are the questions that we have to ask ourselves post batch zero. What do we want this to look like forever? And I think those are the top two considerations from my symbol.

Caitlin Smith: Is there a solve for that?

Jason Ryan: I think that if you don’t want to change the seventy five megawatt threshold and if you don’t want to increase the frequency of the batches or don’t want to or can’t, then you know, perhaps there’s a separate track where there’s clearly available capacity. Yeah. So again, if I’ve got a hundred megawatt facility that they just wanna employ a couple thousand Houstonians, I’ve got the capacity, everybody agrees there’s the capacity. Yep. Why should they wait? For a batch to be run. So you could maybe create that kind of exception. I don’t like having an exception to a brand new process. Exception. That’s why I think that we have to ask ourselves, are we concerned about the hundred megawatt loads? Because if we’re not and we’re concerned about you have a super large loads, then perhaps you create some kind of different process going forward.

Caitlin Smith: That’s interesting. You know, the seventy five megawatts, I’ve been wondering about that. Like, do we see a bunch of seventy four point nine? Like you see the nine point nine generation.

Jason Ryan: Hundred percent. We are seeing it today. Yeah. In terms of the distribution interconnection requests that we’re getting. So I think you are absolutely going to see that because when the process becomes the long pole in the tent, business is gonna wanna move at their speed still. And if the only option to move at their speed is to stay below that cap, I think you’re gonna see a lot of projects that stay below that cap in order to get speed to power. That’s why I raise the question if Is that the right threshold for a longer process? And the answer could be yes, right? I just think we need to ask that.

Joshua Rhodes: Yeah, no, that’s fair. I think that’s one of the questions I was wanting to ask is like, okay, how many 74.9 megawatt data centers are you seeing? Cause I think this’ll be get more clear as kind of like the needs of AI actually play out. So 90% of these large loads are data centers. Gonna presume, given the CapEx spin, that most of this is AI at this point. And we presume we need big data centers for the training of these models to create the new near next frontier models. But for the inference, the actual my students cheating on their homework or, you know, everyone asking kind of how things go, like you don’t necessarily need gigawatt scale data centers kind of for that. And there’s a lot of people talking about we may we’ll start to see a lot more inference data centers that are smaller that are popping up. I may have misunderstood you, Jason, there bit. It sounded like you were arguing for the cap to go up. I’ve heard most people argue it to go down. Like the original was 25 megawatts or something, but it sounds like you’re arguing for the cap. To go up for the batch process. Is that what I’m hearing?

Jason Ryan: I mean that’s the question that I would like to have a debate on. And again, maybe this is the exact right number. Maybe it should go down. No. I think the unintended consequence though of a batch process that is not more frequent than once a year is not going to be consistent with non-data center large loads business plans. Yeah. Especially if they just want to expand an existing site. I think the unintended consequence is that we could lose out on those projects. That’s what I think we need to have a debate about. You know, again, it is not uncommon for us to have a 75 or 100 megawatt facility dropped into our system. We’re quite used to that. It could be unusual in other parts of the state that aren’t used to that kind of large industrial manufacturing load. And I would dare to say that I don’t think those are the ones that are causing the need for the batch. If all we had was a lot of hundred megawatt load, not that that’s small. Right. But I don’t think that we would be in the So why are we scooping them up as well? Right. And again, we may decide that we need to. And I’m always happy for that to be the answer once we have the debate. Yeah.

Caitlin Smith: You know, the lower number, I think it’s confusing, but I don’t know that it’s a real problem. I think ERCOT, FERC, and NERC all have different numbers for what is the large load. So I think that will get confusing. But what Jason raised about raising the megawatt threshold, I think maybe makes sense. The exception point is something I’ve been thinking about a lot because we can get to a n great outcome. We can pass something through TAC that everybody loves policy wise. But what if something critical to Houston’s economy wants to interconnect? You know, what if yeah the governor has a press release about a hyperscaler load? You know, what about all these things that are really critical to our economy? How are we going to accommodate those? Or how are we going to say, well, now that’s on hold for eighteen months, even though we got everybody excited about it or we need it in our city. And it’s just really hard, I think, to provide for exceptions. So maybe the idea of raising the threshold is one that could help with that.

Joshua Rhodes: Yeah, I mean, I I wonder how politically salient something just putting us a particular customer class in the batch system, like the customer class maybe that’s kind of like maybe causing the need for the batch system to come around. I mean, I know that other regions are also looking at separate, either like we were talking about earlier, a little bit reliability standards or different rate classes or different transmission cost allocation mechanisms and things like that for particularly data centers right now that are kind of driving a lot. Of this. And so maybe that’s some of the debates and things like that that we’ll be having. But Jason, I have heard that concern from like non-data center loads about, you know, being kind of caught up in this kind of whole process. But we’re going after the same thing, electricity. Right. So it’s like a tough process. I guess like one final question is we kind of touched on this a little bit, but like the timeline is is we’re going to start the rule takes effect mid July. And, you know, there’s a roughly a five step process kind of coming out the other end. Are we still expecting that we’ll be able to have transmission plan handed to RPG at the late to the end of 2027 that would allow for the output of the batch zero process to then start to take effect, which then won’t get built out for the next like five or six years, kind of depending on how it kind of lands.

Jason Ryan: You raise a good point that I also think raises the question of unintended consequences because the timeline that you just laid out is quite long. Right. And if you think about it, this process is designed in part, maybe in large part, but at least in part, to kind of weed out speculative projects. Yeah. The question you’re raising that talks about all the steps, even once you get past the July tenth and July 24th dates of this year, you’re going well into the end of next year for even more process. I think that customers that are the most real are going to have a problem with that timeline. Okay. Right? So if I am a real customer with contracts with other real customers to deliver something to them, whether I’m a data center or I’m building something, and my contract with that customer has a timeline associated with it. That the more process and longer the timeline to get power, the more you are weeding out the most real projects. And I think that’s the reason why once we get past batch zero, batch zero kind of is what it is. Yeah. But once we get past batch zero, we have to start asking some of these questions of how do I make sure that we maintain the reliability of our system and affordability. Of the rates of that system, but also move at the speed of business. And I am confident we’re going to figure that out. But I think these are the critical questions we have to start asking. Because again, if you start then saying, well, what’s the timeline for batch one? Mm-hmm. You’re talking about timeline for batch zero that goes through the remainder of next year. If I get asked by a customer, what’s the timeline for batch one, when do you think I can get power? It’s very uncertain right now. Okay. And I think the more That the utilities who are kind of on the front line with those customers every day have to shrug their shoulders and say, I don’t know. You know, the more we have the possibility of losing out on development. And that’s why I’m encouraged that ERCOT’s going to turn their attention very quickly to batch one so that we’re not in a phase of having to shrug our shoulders because we don’t know. And again, I have great confidence that we’re going to figure this out and be able to meet this moment. ERCOT, do you see? All the stakeholders have worked super hard to get to this point. I know we’re going to work super hard to understand batch one. And the more we can have certainty over that future batch and what it’s going to look like, the more we as the folks that are talking to the customers on the front line can exude that confidence that Texas is open for business. We want their business. We want the benefits to existing customers of this growth that’s going to reduce costs. Not add to them. Right. But we have to get started on batch one and I’m excited that we’re gonna start those conversations soon.

Joshua Rhodes: Well it sounds like you’ll be there to ask a lot of questions of Caitlin in the intact and as soon as we get this photo. Someone will be there. So it sounds like we need more podcasts later on about this process as Infinite Podcast. That’s exactly right. Caitlin and Jason, thank you for coming on the Energy Capital Podcast.

Caitlin Smith: Infinite podcast.

Jason Ryan: Thanks for having me.

Caitlin Smith: Thank you.

Joshua Rhodes: Thanks for listening to the Energy Capital Podcast. If today’s conversation helped you make better sense of how the energy system actually works, share the episode with a colleague and hit follow on your podcast app. You can find us on Apple Podcasts, Spotify, and all the usual platforms. For deeper analysis and context each week, subscribe to the Texas Energy and Power at texasenergyandpower.com. That’s where you’ll find every episode, every article, and our latest updates. We’re also on LinkedIn, X, and YouTube. Where we share clips, insights, and ongoing commentary on energy policy, markets, and the grid. Before we go, a quick note. The views expressed on this podcast are my own and do not represent the official positions of the University of Texas, IdeaSmiths, Austin Energy, or Columbia University. A big thanks to Nate Peavey, our producer. I’m Joshua Rhodes. Thanks for listening, and we’ll see you next time.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.texasenergyandpower.com/subscribe