Will Interconnection Reform Unlock the Grid?

Electricity demand in the U.S. is rising fast, fueled by the rapid growth of AI data centers and other power-hungry technologies. At the same time, many fossil fuel power plants are retiring, putting added pressure on the grid to maintain reliability.

To meet this challenge, clean energy and battery storage projects are lining up to connect to the grid. The queue now holds more than twice the capacity of all power plants currently in operation.

But getting these projects online is proving difficult. The interconnection process, which evaluates and connects new power projects to the grid, has become a major bottleneck. It is overwhelmed by the sheer number of proposed projects and further slowed by permitting challenges, supply chain delays, and uncertainty around federal incentives. Reforms to the interconnection process are underway, but it is yet to be seen whether they will move quickly enough to make a difference.

RMI’s Sarah Toth Kotwis explores the technical and regulatory barriers to bringing new energy online, and what it will take connect new energy projects quickly and reliably.

Full Transcript

Andy Stone: Welcome to the Energy Policy Now podcast from the Kleinman Center for Energy Policy at the University of Pennsylvania. I’m Andy Stone. The US electric grid is under growing strain. Electricity demand is climbing fast, fueled in part by the rise of AI data centers. At the same time, many aging fossil fuel plants are heading for retirement. As a result, grid operators and regulators have warned that some areas of the US could face a shortfall between power supply and demand.

To meet this challenge, a massive wave of new clean energy and storage projects is lining up to connect to the grid. The interconnection queue, which is the waiting list for these projects, now includes more than twice the capacity of all power plants currently in operation. But the path from proposal to production is anything but easy. The sheer volume of projects has overwhelmed the interconnection process, slowing progress and raising doubts about whether new resources can be brought online quickly enough to keep the grid reliable. And even when projects get the green light, supply chain bottlenecks and potential changes to federal incentives threaten to delay construction and increase costs.

Reforms to the interconnection process are underway, aimed at speeding things up and clearing the backlog. But the question remains, will they work and will they work fast enough? Joining me to explore what’s at stake is Dr. Sarah Toth Kotwis, Senior Associate on the Clean Competitive Grids team at RMI. She focuses on the technical and regulatory barriers to connecting new energy resources to the grid. Sarah, welcome to the podcast.

Sarah Kotwis: Happy to be here.

Stone: So, it is the end of June as we record this episode. And this morning, as I was reading the news, I came across a comment from Mark Christie, who’s the chairman of the Federal Energy Regulatory Commission, which is this country’s electric grid regulator. And he was talking about the recent heat wave we’ve just gone through in this country, and its impact on the electric grid. And he said— and I’m quoting here— “Some of our systems really came close to the edge,” end quote. With that lead in, can you frame the current grid situation for us? How real is the risk that we’re running short on electricity supply relative to demand, and springing from that whole issue, how urgent is the need to add new sources of power?

Kotwis: Yeah. So, the recent heat wave definitely hit home in some parts of the US, especially the Carolinas region. There were a couple contributing factors to that. I heard that there was a nuclear plant that also went offline. I’m not sure what the reason was, but that also contribute-. You know, power plants unexpectedly going offline contributes to potentially not having enough supply during periods of peak demand.

It’s also a combination of not being able to get enough new resources online to be there just in case other power plants go offline. And so it’s really important to make sure that we have a steady new influx of generation. And making sure that we’re building enough generation to meet future demand, even during peak demand times. That’s a part of planning for resource adequacy. It is one of the three pillars of electric grid reliability to plan in advance and ensure electricity supply can always meet future demand.

Now, I will say that typically in the US, electric reliability is actually pretty high. According to a 2024 NREL report, we typically experience less than five hours of outages per year. That’s around 99.95% reliable. And when there are outages, overwhelmingly, those stem from the distribution system rather than generation or supply side outages.

But that being said, forecasts of increasing demand and all the reasons you mentioned— AI data center projections, electrification of buildings, transportation, industry and more, while constraining supply due to broken generator interconnection processes— causes these resource adequacy concerns and raises prices and brings us closer to the margin than we really should be on the supply side.

Stone: Talking about the supply side—and I guess on the bright side of that— there is a huge volume of mostly clean energy resources that are now waiting to connect to the grid. I wonder if you could introduce the concept of the interconnection queue, which is where this waiting is happening, and why there’s been a rush of new resources to enter it?

Kotwis:  Yeah. Largely, the reason for the interconnection queue is that grid operators have to study. There has to be some mechanism by which you get in line and you try to get studied to see if your generator is going to adversely impact the existing US grid. So there’s a lot of modeling and reliability modeling that happens to make sure that the existing grid can handle that additional generation. And so historically, you’re a prospective generator, you have to get in line to get studied, and then maybe you get to connect. So that’s what it’s like in most parts of the US.

The rush has actually been building for a while. It’s not just recently. You can see that over time, annual additions to queue requests have just gone up and up and up for the last few years. And that’s largely due to significant reductions in the cost of new renewable energy resources like solar energy resources, wind energy and battery energy storage resources. So those have all become the most cost-competitive resources out there available to build. And so project developers are seeing that, and they’re also seeing these price signals from our existing markets that, “Hey, we need more new energy entrants into the market to serve resource adequacy.” And so everyone is trying to get in line to be that next project that can come online, contribute to the grid, and also reduce costs.

Stone: So once a new generating project successfully goes through that interconnection queue process, it gets what’s generally known as an interconnection agreement, which is the green light to move forward. But one of the issues here is that, historically, just a fraction of the projects that actually enter the queue successfully get through. Give us an idea. What’s the success rate going through the queue?

Kotwis:  There’s a good stat from Lawrence Berkeley National Lab on this subject. It’s a nationwide stat that only about 20% of projects that requested interconnection during 2000 to 2018 reached commercial operations by the end of 2023. So that’s pretty low. And then, more specifically, in PJM— which is the grid operator that started as Pennsylvania, New Jersey and Maryland, hence PJM, and now includes 13 states plus DC— of the projects that entered their queue between 2014 and 2023, 67% of those withdrew, and 16% are operational, with the remaining in some stage of flux, primarily in engineering and procurement or temporarily suspended.

So there’s three primary drivers for this fairly low success rate. First, it’s about competition. Competition, this is an inherent feature, not a bug of competition, that not all prospective projects are successful. Poorly chosen points of interconnection or costlier build-outs are weeded out.

The second piece is declining headroom, or available space on the existing transmission grid. So ideally, market price signals incentivize savvy developers to build where it’s most beneficial to the market. But a lack of proactive transmission build-out historically means that available space on the existing system is in decline, which makes it even more difficult for developers to find good locations on the grid. And when they do find good locations, it’s even costlier to connect because it’s more constrained than before. So this has led to a lot of speculative projects getting put in the queues, because that’s the only way for them to find out what these potential really high costs at any location would be.

And finally, the third piece is the duration of time these projects are actually spending in queues trying to achieve the agreement. And then on top of that, post agreement, experiencing additional construction delays. So the bottom line, there is this gap between what we are hoping or thinking about building, and what actually gets built. And in order to improve that, we first need to build more of the right kinds of transmission to get power from where it’s generated to where people need it. And second, improve these queue processes to make the best use of the existing system that we have, via leveraging advanced technologies and software, and reduce time spent within the queue itself.

Stone: Well, as you just mentioned, there’s this rush of new projects. But the grid is basically what it has been for quite some time. So there are a limited number of areas where these new projects can actually connect to the grid without triggering necessary upgrades to the grid to handle that power from those new projects. Is that right?

Kotwis: That’s correct. Yeah. So that’s about trying to understand, okay. If you did want to connect here and this location is already full, what would you have to invest in the grid in order to make it safe for this project to push power onto the grid at this particular location? And that’s the theory behind the invest-and-connect strategy that happens in most of the US.

Stone: You also mentioned speculative interconnection requests, and you mentioned that the actual success rate, again, getting through the queue, is low. So those two are connected, right? So if I am a company that wants to build new generation, is the speculation, then, that I propose projects at several areas on the grid, hoping that I don’t get stuck with the upgrade costs for the grid to connect my resources, and eventually I just go ahead and connect where those upgrade costs are going to be relatively low?

Kotwis: Yeah, that’s the thinking. Yeah. And unfortunately, that’s the behavior that we’re seeing. But it is just rational market actors trying to make the best of a difficult situation, right? So it never used to be this way, that it took so long and it was so costly. And really, the only way to get a certain price discovery is to go through the queue process itself. And so that has compounded the issue, which is that now we have so many backlogged projects trying to get studied. And some of those— not all of them, but some of them— surely are speculative. And it results from this lack of available data on, where is the best location to connect to the queue? That’s hard to find, and also declining over time.

Stone: You mentioned the long wait time once you’re into the queue. And that obviously has to be a turn-off for project developers as well. How long are those wait times?

Kotwis: Yeah. Another good stat from LBNL is that pre-2008, it took less than two years to connect to the grid. But recently, this has grown to more than five years, and seven-plus years in some places. Now that’s a huge time delay, during which a lot can change. So there’s two distinct processes within that request to operation. Or I think previously, you called it “proposal to production.”

So within that, there’s two main pieces. First, the request to agreement piece. So that’s where in invest and connect in regions in the US, those queued projects are studied in power flow models, and you identify the necessary reinforcements, if needed. And then the second piece is from agreement to commercial operation. So now they’ve had an agreement— and maybe we’ll come back to this later— but they have an agreement, and there’s still additional project hurdles to face once you have a connection agreement. You still have to go through permitting and procurement and all that sort of stuff.

In terms of how long it’s been taking, there’s some data transparency and completeness issues here. Because a lot of times, the signed agreement date isn’t listed or is overwritten in publicly- available data sets. So for example, for projects that entered PJM’s queue during 2014 through 2018, we can see in the data that there are some of those projects still in the engineering and procurement phase or under construction,

Stone: But these are projects that go back seven years or so? That are still— 

Kotwis: Yeah, or more. Yeah, but we don’t know how long ago their agreements were signed. And then if you fast forward a little bit, some projects in PJM’s— they’re calling it “the expedited process,” that entered the queue between 2018 and 2020— so that’s five to seven years ago now— they just received interconnection agreements in April of this year. Five to seven years later. And so that’s a really long time to languish in the queue. And you don’t really know, are you even going to get an agreement out of this? You can’t start your permitting or procurement processes until you know. You can’t order equipment until you think you can connect to the grid.

Stone: So we’ve got two areas of delay. We’ve got the interconnection process, which is amazing. Seven years, some of these projects have been waiting in the queue just to get their interconnection agreement. And then additional delays, which we’ll talk about later, after that queue. Downstream from that queue. Just interesting, the costs of staying in the queue, there must be some cost to that as well. I mean, if somebody has entered to queue in 2018, this cost for maintaining that, I would suppose all this time. Too, the world in 2018 looked very different than the world that we’re seeing today. And AI wasn’t even in discussion seven years ago. Are the projects that entered the queue seven years ago, are they even relevant today?

Kotwis: Yeah. It’s an interesting question. I mean, I would say that they’re definitely relevant because we haven’t been adding enough new energy resources over the last few years. And we can see the negative consequences of that in the price spikes in the capacity market and energy market. That’s a part of a functioning electricity market, is the ability of new market entrants to be competitive and increase competition in the market and try to get prices to come down. So we really do need additional resources, and the load is another piece. It’s another piece adding to the urgency, that we need that for resource adequacy as well, not just to ensure competitive market prices.

Stone: Well, it’s interesting. So we have all these resources that are coming through, and I just want to bring this into the very current context. We’re still waiting to see what Congress does with the budget bill, which it’s working through. Again, this is the end of June right now. A couple episodes ago on this podcast, or maybe it was last episode, we looked at the House bill and what its impact may be on the pace of clean energy development going forward. And one of the things that came out of that conversation was, due to the House’s proposed end of 2028 in- service requirements for new resources to receive IRA tax credits, we may actually see an acceleration of clean energy projects over the next three years as those projects try to get in before that deadline expires. So what we could potentially see is we have pressures of so many resources trying to get through the queue, and the processes downstream of that. That sounds like that could potentially intensify. Would that clog up the system even more?

Kotwis: This is my understanding of the current situation, is not only does the House version have this in-service by end of 2028, but it also includes a two-month deadline for projects to start construction after IRA guidance is finalized. Now the Senate version doesn’t include those deadlines, and I can’t speculate on what’s going to happen with the bill. But, you know, we just spoke about how projects have been waiting many years to potentially receive an interconnection agreement. To suddenly have an expectation of immediate construction isn’t very realistic.

That being said, of course they’re likely going to try to do as much as they can to get these projects constructed. I mean, that’s in everyone’s best interest. That’s in the developer’s best interest, the grid’s best interest, consumers’ best interests. But many factors contribute to whether a prospective project pencils financially and can access the necessary permits and equipment needed to be constructed. So those are all open questions. And yeah, I’m not sure if that necessarily will result in an acceleration, or if it won’t accelerate anything, and actually cause things to be much more chaotic.

Stone: So the grid operators and the regulators are quite aware of this interconnection backlog issue that we’ve been talking about. And PJM, which you introduced a few minutes ago, reformed its interconnection process in 2022. And FERC, the grid regulator, issued Order 2023 the following year with very similar reforms. What are the key elements of these reforms? Tell us about that.

Kotwis: PJM did initially develop a stakeholder process and identified some interconnection and reforms in the 2020-to-2022 time period. And those reforms were submitted to FERC, approved by FERC in 2022. And so they have been, since then, dutifully working on implementing those. What’s happening since Order 2023 is they’re trying to largely build those 2022 reforms as sufficient.

Now there are some things that those two have in common. For example, the transition from a first-come, first-served serial approach to a first-ready first-served cluster approach. So that’s common to both the 2022 and Order 2023 reforms. But it remains to be seen how this plays out. The first cycle implementing this approach in PJM is still being studied. They are ongoing in that process.

Stone: Could you tell us more about, what is that first-come, first-served approach that is being replaced by this new first-ready, first-served approach? What does that exactly mean?

Kotwis: The first-come, first-serve approach is more like a one-at-a-time thing, no matter who has gotten into the queue. Just like how you line up at a grocery store to check out. You are first in line, you get served first, right? So that’s that serial, one-at-a-time project approach. So the thinking was that that isn’t necessarily the most efficient way to go about studying projects when there’s such a huge backlog. That’s how it was done before we had this huge backlog. Now, with the huge backlog, maybe we could garner some efficiency if actually we studied projects together in a group of projects based on how ready they are to move forward, rather than just the project that’s first in line. It’s kind of like, akin to, “Well, everyone who has their pocketbook out, ready, you come first. And we’ll study you first and see you through.”

And so to implement that, for example, there’s new readiness requirements and study deposit requirements. That’s the other piece that both of these have in common. These are financial deposits that scale with project size, though the specific value may differ depending on if you’re talking about PJM or Order 2023. These are trying to make sure that the projects coming through the queue and actually being studied— it takes a lot of time and engineering effort to study these projects. It tries to make sure that those projects are ready to get built and more likely to be successful.

Stone: Again, PJM introduced its interconnection reform in 2022. FERC introduced its own in 2023. Again, I want to get to that issue. Are they basically the same reforms?

Kotwis: There are definitely some differences. So one of the differences in FERC Order 2023, it requires a 150-day maximum cluster study duration. PJM’s is longer. Order 2023 also imposes study delay penalties. PJM wants to use reasonable efforts and not impose penalties.

Stone: Is that a result of the stakeholder process in PJM, in which many of the companies that have skin in this game would not want to be subject to any kind of penalties?

Kotwis: I’m not sure about you know, their intention behind it. All I know is, comparing what they’ve written in their Order 2023 compliance filing submission versus what is being asked for in the order. So, yeah, they are largely trying to bill their 2022 reforms as sufficient. But in many ways, those are different than what is in Order 2023. So the duration of cluster studies is one. Study delay penalties is another. Using customer-provided information regarding how battery energy storage resources operate, and aligning the operation assumptions with how they’re studied in the model to ensure that they’re studied in a manner commensurate with how they’ll operate in reality. This is not something that PJM has wanted to comply with in their compliance filing.

And then finally, requiring an evaluation of alternative transmission technologies. That’s also required by Order 2023. And this is an even trickier piece, where PJM says they already do this, although it’s unclear to what extent and how this happens. You know, I haven’t seen grid- enhancing technologies discussed or recommended in any recent study report that PJM has published. And, you know, in comparison, in our Getting Interconnected, a PJM report from 2024, we found that more than 1500 megawatts of queued projects in Pennsylvania could connect more quickly and cheaply with grid-enhancing technologies. But right now, only 450 megawatts remain in the last study phase of transition cycle one. So, what happened? Were GETs considered? How rigorous was the study? We have no real way of knowing. And the problem is, there’s a lot of potential customer savings being left on the table if they were applicable and weren’t studied to be found to be applicable.

So we actually conducted that study across five states in PJM and dumped the results into an economic model of the future system, and found that as early as 2027, consumers could start seeing $1 billion in annual energy market savings. But it doesn’t look like that’s going to happen. And it doesn’t look like grid-enhancing technologies have been chosen, and we don’t really know why.

Stone: Well, what are the grid-enhancing technologies that you’re talking about here that would accelerate this process? Give us a little overview.

Kotwis: Yeah. So there’s three primary grid-enhancing technologies. The first is dynamic line ratings. The second is advanced power flow controllers, and the third is topology optimization. And these aren’t exactly the same as those identified in Order 2023. But Order 2023 does identify some of those. And those are the three that we studied in our report, because they’re comparatively, compared to reconducting a line or rebuilding an entirely new transmission line, they’re very cost effective, very cheap, and they are also modular. They could be moved from one place to another. And they can be deployed very quickly, on the order of months, instead of the years that it takes to reconduct or rebuild transmission lines. So in this way, we could leverage these like cheap, low-cost technologies to make the most out of the existing grid that we have even as it nears capacity, because these devices help make our use of the transmission system more efficient.

Stone: So basically, what you’re saying here is that, again, with the grid that we have right now, more new resources could come onto the grid without having to invest in the more expensive new lines of reconductoring, because these grid-enhancing technologies could make the most of the grid that we have right now.

Kotwis: Yeah, exactly.

Stone: So it’s interesting, you said that PJM does not require these in its reforms, but FERC does, if I understood that correctly.

Kotwis: Yeah. So PJM says that they already do this. But again, it’s unclear to what extent and how that happens. And the result being, we aren’t seeing them being offered in the interconnection studies as network upgrades. And so I would love to know more about why that is. You know, these aren’t applicable everywhere. But we found that they were applicable in some places. And so would love to know more about what the discrepancy is there, because Order 2023 is really trying to make sure we’re using flexible and readily-available technologies to get the most out of the transmission system. So you know, we have to see whether we’re doing that.

Stone: To make sure, also, I understand this. If FERC is saying that it’s a requirement that you at least consider these grid-enhancing technologies, wouldn’t PJM ultimately be required to do the same? Because ultimately it would have to comply, I would believe, not only with its own reforms, but with the 2023 reforms from FERC.

Kotwis: What can happen is that PJM and any RTO or any any FERC jurisdictional entity can request an independent entity variation. So it’s called an independent entity variation, and it basically means that, for whatever reason— there are the unique particulars of how they operate their system, and their system— they think that this shouldn’t apply to them. So that’s one of the ways that they could say, “Well, this doesn’t really apply to us.”

Stone: Okay, so these reforms have been out for a couple of years at least. What results are we seeing from them so far? Are they starting to make a dent in the backlog or timelines in the interconnection queue?

Kotwis: There’s a good Interconnection.FYI blog on this subject that came out recently. And for the first time in years, queue backlogs almost nationwide have decreased in size rather than increased, the one exception being the Midcontinent Independent System Operator, or MISO. So far, PJM’s transition cycle one and transition cycle two, which are the first cycles that they’re applying their 2022 reforms, they have proceeded faster than their prior serial processes. So this increased speed will go a long way in helping. And using automation and software-based automation to speed up study processes still hasn’t been implemented in a variety of regions, but could also help to speed things even further. PJM has announced that they’re considering doing something like that in collaboration with Tapestry. So that’s really exciting. And it remains to be seen when and how that’s implemented, but there could be a lot of further improvement based on that going forward.

Going back to the original three reasons that commercial operation success rates are so low, speed is one of those aspects. But the other remains the lack of transmission. So Order 2023, that’s going back to this requirement of the consideration of transmission technologies that help make the most efficient use of the existing grid that we have. So this is one of those reforms that still holds a lot of promise, even if we are able to crunch through the queue and problems more quickly.

Stone: Well, I want to bring up another electricity market, which is ERCOT, which is the grid operator in the market for the majority of the state of Texas. And they’re a bit of an outlier in that they have been quite successful in getting new capacity online relatively quickly. And this ties into what is known as their energy-only market model. Can you explain that model and how ERCOT has managed to get resources onto the grid a little bit faster?

Kotwis: Yeah. ERCOT is unique in a couple of respects. ERCOT stands for the Electric Reliability Council of Texas, and so ERCOT is not subject to FERC jurisdiction. They don’t have a capacity market, and they leverage what’s called the connect-and-manage interconnection approach. So under connect-and-manage prospective, new generators are allowed to connect to the existing transmission system expeditiously, with the understanding that the grid operator will curtail them or limit their output even if they’re available of producing output, depending on the local transmission grid needs and congestion levels. On a quarterly basis, the grid operator will review the performance of the entire transmission system to identify the most optimal upgrades and then take action on those.

Conversely, the rest of the US operates under this invest-and-connect paradigm, where the prospective new generators must first be studied in a power flow model to assess whether the current grid can handle the additions. And if not, they’ll identify the necessary reinforcements. So those reinforcements are called network upgrades, and the prospective generator has to pay to build those, thereby investing in the grid. That’s where “invest” comes from in invest-and- connect, prior to be being allowed to connect. So in return, the developer earns the ability to participate in the capacity market.

Now, ERCOT doesn’t have a capacity market. As you said, they’re an energy-only model, so they can do things differently. And to be clear, there are pros and cons to each approach. And it’s not necessarily a binary, although current markets make it out to be. For example, energy prices in ERCOT can be more volatile than in other regions because it’s the only price signal available to incentivize new entry. Other than its ancillary service market, but that’s a discussion for another time.

But on the other hand, the pro to that is that they can bring new resources online very quickly and expeditiously, and so that’s helping them sail through these heat waves that we’re seeing, even with low prices. So on the other hand, in invest-and-connect regions, FERC stipulated in Order 2003— not to be confused with Order 2023— that there should be an option for developers, even in invest-and-connect regions, to request energy-only service, rather than network-to-resource service, which theoretically would offer quicker and cheaper interconnection to the existing system, subject to curtailment. This rings and sounds kind of a little bit similar to this energy-only-with-curtailment model that ERCOT is going for.

However, each region was given deference, through these independent entity variations that we previously talked about, in how it implemented this option. And in reality, in many cases, it’s just as costly and cumbersome to pursue as a network service. And so many developers choose to simply pursue network service given that they would additionally be allowed to count on earnings from the capacity market.

Stone: So to sum this one up, if I’m a new resource and I want to connect to the grid in ERCOT, which is the energy-only model, I connect, and I don’t necessarily have to engage in upgrading the grid in any way whatsoever. I just take it as it is and I connect my new resource. But there is, as you mentioned, the possibility of curtailment, which means that if it is a particular period where there is a lot of demand on the grid, the grid is under a lot of stress, and the energy from my generator is adding to that stress by putting more energy into the system, at a particular point, I may be shut off from the grid or asked to reduce my output so that I don’t destabilize the grid. And then, for that reason, I don’t enjoy the revenues that would have come from producing more energy at that time. So it sounds to me like the ERCOT example has worked. But if I am somebody who’s investing in these new resources, I do face that risk, that my revenue may be volatile in the future, because I may be curtailed once or multiple times at times of grid stress. Is that right?

Kotwis: Yeah, absolutely, that’s exactly right. And so it’s important to keep in mind that there’s trade offs to each approach, right? That is a long-term risk for projects. But in the near term, you know, they could connect really quickly. Whereas in the rest of the regions, you might have more financial certainty down the line, but you might not be able to connect for five to seven years, you know. And there’s a real discrepancy there.

Stone: And to get those capacity payments, like in PJM, which has a capacity market, it’s a companion market to the energy market that we’ve been talking about here. You have to prove yourself to be reliable, and your interconnection has to be reliable, before you can even connect to that grid. So that’s where that additional investment in buttressing the grid before you even connect comes in. But once you’re in, you are no longer subject to curtailment. Your revenues are going to be much more predictable after that point.

Kotwis: That’s true in theory.

Stone: Okay.

Kotwis: But in reality, what actually happens is the dispatch models that grid operators use to dispatch the grid does so on a cost-optimal basis. So whatever the generator is that is available to produce and is lowest cost will be dispatched first, regardless of whether they’ve paid to upgrade the network. That isn’t taken into account in existing economic dispatch models. And this is how we’re able to dispatch the grid and keep prices as low as possible, because we build from the ground up. What are the cheapest resources first, that can be deliverable, and then add on top of that.

So a lot of folks think that if they go for network resource interconnection service, then they won’t be curtailed. But that’s not necessarily the case. Curtailment happens based on whether your resource is cost-competitive at that time and can be delivered to load.

Stone: So let’s look at some of the solutions going forward. So again, we have this interconnection queue issue that we’ve been talking about, and some of the solutions to that. You already started to talk about grid-enhancing technologies. You also mentioned the importance of data transparency and key data about the grid that should be available to new project developers. How available is that data? How transparent is that data right now, and how would increased transparency potentially help to resolve these interconnection delays that we’ve been talking about?

Kotwis: As an organization, we’re doing research on what does the existing data tell us? And trying to make recommendations for improvements based on that. It’s really important to have access to high-quality, transparent data. Now, what we’re finding is that every region across the US implements their processes differently. They use different words and different keywords to talk about the same information, for example. Sometimes they’re updating their databases and they’re overwriting information that previously existed, and so that information is now lost. So there’s a lot of ways that the way that data practices are being handled right now results in a lot of holes and gaps in the data. And that prevents us from being able to really have a holistic picture of what’s happening.

So for example, like one thing that we would love to see is a simple data entry box that says, “Okay, which technologies were evaluated when trying to identify the appropriate network upgrade in this circumstance?” Well, that doesn’t exist. We don’t know what all was evaluated. It would be nice to see like, “Oh, we looked at dynamic line ratings. And those aren’t appropriate here for XYZ reason.” But that’s not something that we have access to.

Also, I mentioned the date at which interconnection agreements were signed. That’s really important to know, to really get an idea of how long projects are languishing in the queue. Now, we have some information for it, but it doesn’t cover 100% of the queue requesters, right? So just having more complete information that isn’t overwritten and can be compared across regions would really help us to gain a better understanding of not only how things are going, but how reforms are improving things.

Stone: It’s interesting as well, Department of Energy, in April of 2024, released a report that’s titled “The Transmission Interconnections Roadmap.” And it had some additional suggestions as well. Some issues such as surplus interconnection service, the energy-only interconnection that we’ve just talked about. And also something they refer to as market-based approaches to rationing interconnection access. What are these things, and what’s the potential behind them?

Kotwis: So I’ll focus on the SIS, because that piece, I think, is really important. And we’ve worked on it a little bit with PJM. We presented to PJM on this topic in the fall of last year. So what surplus interconnection service is, it’s enabled by a pre-existing FERC order, and it allows the full utilization of the capacity interconnection rights that are given to existing projects on the system. So for example, let’s say you have a natural gas plant and it operates a majority of the year, maybe 80% of the year. But that remaining 20% of the year, if it’s not operating during a time where some solar energy resources would be at full output, then you can use the surplus interconnection service avenue to add solar resources to that existing plant, and only up to the level of interconnection rights that have been given already to that existing gas plant.

Now, an even more beneficial use would be taking standalone wind and solar resources and adding batteries to them. That’s a really quick and cost-effective way to add capacity to the existing system. Right? So a solar plant that’s standalone will only use a certain portion of the interconnection rights that it’s been given. And adding a battery behind that point of interconnection to firm up that resource and store some energy when it’s not needed for the rest of the grid, and then sell that energy back to the grid when prices are high in order to reduce those prices—that’s another really beneficial way of extending the capacity contributions of existing resources in a really cost effective manner.

So this is a new thing in PJM. FERC recently approved their proposal to treat storage resources more appropriately when they’re being studied for surplus interconnection service, and hopefully we’ll see a lot more storage come online via that pathway in PJM,

Stone: You mentioned earlier that projects that do successfully get through the interconnection queue still face a number of hurdles, and I think those hurdles have become a little bit higher recently. They include siting and permitting, which are always kind of an issue, as well as supply chain issues that are impacting renewable energy at this point as well as natural gas projects. How big are these issues right now, especially the supply chain delays?

Kotwis: Beyond the queue, these additional hurdles that projects face— it’s primarily siting, permitting, and supply chain. So this siting piece, site selection is often led by transmission availability and market price signals, right? We talked about how ideally, the market can incentivize new entry in the right places. But because the limited transmission capacity is this big bottleneck, that means that the queue dictates the pace at which developers are able to proceed with these other pieces of the puzzle.

So as a reminder, the queue is still a big driving bottleneck, because it takes so long to get through it. And then once a project has gotten through it and they’ve obtained an interconnection agreement, there’s a lot of discretion in local permitting that increases uncertainty and often timelines for project construction. So this discretion at the local permitting level sometimes leaves decisions up to an individual’s personal judgment without clear guidelines. Just yesterday, actually, PJM published that these permitting delays caused more than 14 months of commercial operation delays on average.

We do think that predictable rule-based permitting systems could increase clarity and consistency through this process, and lower risks for projects going through it. But even once a permit is obtained, then supply chain constraints can add to the procurement dilemma. So having an interconnection agreement and permit gives developers the certainty to place an order on equipment. But COVID and more recent hurdles have strained supply chains, and especially high-voltage equipment has much longer lead times than before. So in PJM’s same presentation yesterday, transmission owner material procurement caused almost 20 months of delays. That’s a long time for just to be waiting on equipment that you’re purchasing.

Stone: So that’s like big transformers as well, right? That type of thing.

Kotwis: Exactly. Transformers are a really big piece of this puzzle. We need to increase the manufacturing of this critical high-voltage grid equipment. And transformers, especially. And it’s not only for new generation builds, but also to connect large loads. They also depend on transformers. And to help regions recover after damage due to natural disasters. You know, every time a big hurricane comes through, there’s sometimes hundreds of transformers that are down. And we really need to be manufacturing this really critical grid equipment to keep pace with everything that’s going on.

Stone: So let me ask you a final question here. And I guess this is kind of the takeaway of the conversation. We’ve talked about all these challenges with the interconnection queue, downstream with supply chains and permitting, et cetera. But we really need resources on the grid quickly right now. And it’s kind of a battle. We need the resources quickly, but we have these delays. What is your personal outlook on this? Do you think that these queue delays, the downstream delays in getting the projects actually under construction and built, do you think they will be resolved? And do you think we’re going to have enough resources coming on to the grid over the next few years to mollify our concerns about resource adequacy going forward?

Kotwis: It’s a big question. And I would say that we have to be realistic. This is a really tall task. We haven’t necessarily set ourselves up for success. But we can work to make the best of what we have. So we have both a lot of new resources trying to connect, and the enabling technology to get them connected. Study automation software, grid-enhancing technologies, high -temperature, low sag advanced conductors, advanced batteries, surplus interconnection service and power couples, all on the supply side. And then on the demand side, energy efficiency, demand flexibility, virtual power plants. All of these are low-cost tools in the toolbox that we can leverage here and now.

And if we focus on being an enabler of new generation and transmission, and implement more interconnection solutions than obstacles, consumers can reap the benefits of abundant, low cost, reliable energy.

Stone: So we have the tools. Hopefully, we’ll use them.

Kotwis: That’s right,

Stone: Sarah, thank you very much for talking.

Kotwis: Thank you so much for having me.