Why Electrical Grid Governance Needs Reforming

The U.S. electrical grid faces declining reliability, often attributed to a rapidly evolving energy mix, surging demand, and more frequent severe weather. Yet a deeper issue lies in the fragmented governance of the grid, where conflicting visions from federal, state, and industry-level regulators hinder progress toward a clean and reliable energy future.

Shelley Welton of the Kleinman Center and Joshua Macey of Yale Law School examine the tangled web of grid governance in the U.S., and highlight inherent conflicts of interest and clashes between state and federal regulatory priorities. They also discuss the prospects for governance reform.

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 past few years have witnessed unprecedented challenges to the US electric grid. In 2021, power outages tied to winter storm Uri in Texas led to more than 240 deaths in the state. More recently, during Christmas 2022, another winter storm, Elliot, led to outages in the Eastern US and brought the PJM electricity market, the nation’s largest, perilously close to failure. These and other recent system problems are frequently attributed to the country’s rapidly changing mix of generating resources, including the growing reliance on natural-gas-fired power plants and renewable energy. The aging of the nation’s grid infrastructure and the increasing frequency of severe weather events have also been named as factors contributing to the grid’s reliability problems.

Yet there is a strong case to be made that the electricity system’s challenges are not at their root simply a matter of power resources and severe storms. Instead, these problems are symptomatic of a more fundamental failure of grid governance, or the institutions and rules that determine how the electric grid is planned and subsequently built and how it is operated. On today’s podcast, we’ll explore the byzantine reality of electricity grid governance in the United States, and why it makes developing solutions to ensure the current and future reliability of the grid so challenging.

Today’s guests are Shelley Welton, a professor of Law and Energy Policy at the Penn Carey Law School and the Kleinman Center for Energy Policy. Josh Macey is an associate professor at Yale Law School. The two will discuss how a lack of public oversight of grid reliability, inadequate coordination between electricity grid regulators, and a misreading of reliability challenges underlie the grid’s reliability concerns. They’ll also explore solutions to meet these challenges and provide their view on how a changing national political reality may complicate efforts to reform grid governance. Shelley and Josh. Welcome to the podcast.

Shelley Welton:  Thanks so much for having us.

Josh Macey: Thanks for having us.

Stone: So Shelley, earlier this year, you and Josh— along with co-authors— published a paper titled “The Key to Electric Grid Reliability: Modernizing Grid Governance.” Introduce the grid challenges that are driving reliability concerns today, if you don’t mind. And this is what your paper seeks to address.

Welton:  Yeah. So, I think you started to get at a lot of these dynamics in your excellent intro. But fundamentally, the electricity grid is changing. And there’s several things that are driving that. We are in the middle of a clean energy transition, and that is in part fueled by economics. It is in part fueled by the largest climate bill ever passed in the United States, the Inflation Reduction Act from 2022. It’s in part fueled by some state policies. And we’re fundamentally seeing a different set of resources connected to the grid, and lining up to connect to the grid.

I think the most staggering statistic that I’ve heard on this is that if you look at what’s in line to be built by developers, it’s almost all solar, storage and wind. And there’s about twice as much waiting to connect to the grid as there is total power in the US electric grid right now. And so we’re just witnessing a sea change in what developers want to build and put onto the grid. And you know, that creates a different set of challenges for the grid in terms of how you manage reliability under this set of resources versus the previous century set of resources. And we’ll talk more about those in a minute, I think.

Two other dynamics I thought I might talk about. One is, you brought up these worsening storms. And many of those that are putting severe stress on the grid are caused by climate change dynamics, or at least contributed to by climate change dynamics. And so at the same time as we see an energy transition underway to try and address some of these challenges, we are watching the ways in which climate havoc is really putting new stresses on the grid that need to be managed, and that, you know, in many cases, sort of weaken the ability of old resources to perform the functions that they used to perform on the grid.

And then the last thing I thought I might mention is just, we are also entering a period of load growth unlike what we’ve seen in a long time, driven by, you know, a confluence of electrification and increased domestic manufacturing, and everybody’s favorite darling to talk about— artificial intelligence and data centers as well, to a certain extent. And so as we see old resources retire and new resources line up but not get connected yet, and load growth increasing, you’re seeing more and more people worried about reliability under these changing system dynamics.

Stone: We’re also seeing a corresponding shift in the nature of the grid reliability challenge, which you started to talk about there. Traditionally, reliability was a, quote, “resource adequacy” issue in industry speak. More recently, however, concern has grown around the issue of what’s called “operating reliability”. Josh, what are these two concepts, resource adequacy, and operating reliability, and what are the implications of each four grid reliability?

Macey: Okay. So, I’m going to sort of softly question the premise. But first, I’ll answer the question, and then I’ll ask a follow up question. So one way of thinking about this is, resource adequacy is sort of the provenance of economists, and operating reliability is the provenance of engineers. So traditionally, resource adequacy refers to the idea that you need enough generating capacity to be able to meet peak demand. So when it’s negative 10 degrees in the winter or 110 degrees in the summer, you need to make sure you have enough generation to actually meet demand at those peak moments. You can sort of think of this as, how do you build a parking lot? You don’t build a parking lot so that there are enough parking spaces when nobody goes to the supermarket. You build a parking lot to make sure you have enough spaces during rush hour.

And there are a set of challenges about how to pay generating units, the generating units that operate a few times a year or only a few times every few years, but which are really, really needed to maintain reliability. And resource adequacy has typically been understood to be that problem. Now there are ongoing challenges about how to deal with that, and I think we’ll get back to those in a bit. But the resource adequacy issue is just, you need to have enough generating units to meet peak demand. And how do you develop markets, resource adequacy markets, to do so.

The operating reliability, you can sort of think of it as, can the system respond to sudden, rapid changes in demand, or withstand disturbances, without major disruptions? These disruptions can occur for a variety of reasons. A transmission or distribution line can trip. You don’t want the whole system to go down. You want to be able to contain it to that sort of small area. And so operating reliability, we think of as maybe an engineering question about, how can you make sure electricity continues to get where it’s needed, and how do we have backup systems or ancillary services to make sure that we can do that? I guess the final way you can think of it is just, how do we manage power during disturbances that we don’t anticipate? Because we know that things will go amiss, and so we need to have a way to keep the lights on.

So, as I said, operating reliability— you know, one way you can think of it is, it’s, can the grid maintain a stable flow of electricity during unexpected events and disturbances? And so unlike resource adequacy, it’s not about making sure there’s enough generation in the market. It’s about, how do you respond to sudden unexpected shifts? That can be, a transmission line goes out. That can be the grid, as many of you know, must maintain supply and demand in perfect balance in real time. And so if a generating unit is suddenly unavailable for some reason, or if consumers start increasing, rapidly increasing demand for electricity, you have to be able to balance that. Otherwise, you have system-wide challenges. And operating reliability is typically understood to be about, how do you address those challenges?

And when I mentioned that I was going to sort of gently push back against the premise of your question, what I meant was that, you know, you suggested that operating reliability is an increasing issue as the grid changes. And I think my own view is that both resource adequacy and operating reliability at least have become increasingly controversial as the resource mix changes and as climate change leads to different weather patterns.

And just to give a few obvious examples. To the extent that wind and solar are perhaps more intermittent than traditional base load resources, that can lead to both resource adequacy and operating reliability challenges. It means there may be more uncertainty about how much backup or firm power you need— though there’s excellent research on this question in terms of operating reliability. It means that you might see, both because resources are intermittent, but also because you see more extreme weather events, more instances in which grid operators have to suddenly respond to unexpected shifting conditions.

Stone: So Shelley, you and Josh and your co-authors argue in the paper that the way in which the grid is governed is ill-suited to enabling the necessary change, or driving the necessary change that is necessary to address the reliability issues we’ve begun to discuss. To start us out on this, can you define for us clearly, what is grid governance?

Welton:  When we use the term grid governance, we mean the landscape of institutions and rules that basically create the conditions under which we try to manage the grid. Right? So, the players in the system, and the powers that they’re given by law to try and build a good grid that delivers all of us power whenever we want it, at the times that we want it, from the sources that we collectively decide that we want it from.

Stone: So grid governance today is highly siloed, which in turn makes effective governance very difficult. Can you talk about the many different jurisdictional silos that exist and how they make responding to grid challenges much more difficult?

Welton:  This requires a little bit of a whirlwind tour, and I think maybe the easiest way to do this is to start from the top down and talk a little bit about the different players in the system. So, you know, at least putatively sitting at the top of all of this is the Federal Energy Regulatory Commission, which is the major federal agency that’s charged with managing the transmission system and the wholesale electricity system. That is, like, sales between generators and utilities, not to end-use customers.

And you know, FERC has some legal authority over several of the entities I’m about to discuss. But part of our argument is that perhaps it’s not enough. So one of the entities that FERC oversees, in part, is the North American Electric Reliability Corporation, which, as its name suggests, actually operates somewhat beyond the borders of the United States. A little bit into Mexico, a little bit into Canada. And NERC is in charge of setting standards for the reliability of the bulk power system, which you could basically think of as like generation and transmission, right? The big stuff. And it’s somewhat jurisdictionally cabined, so it can’t set any standards that require new things to be built. It can just set standards for the performance of the things that are already on the bulk power system. And it operates under fairly lenient FERC oversight.

Also operating below FERC are what are called regional transmission organizations, or RTOs, they’re often known as. And these are— I think of them as, like, a utility of utilities. It’s a bunch of utilities that get together in a region, and they’ve agreed to basically jointly manage their systems, at least to a certain extent. So RTOs control the dispatch of electricity across the system. They run whatever markets the region decides to run, and they do future transmission planning for the region. And again, these are overseen by FERC, but under, you know, pretty lenient standards.

And then I’ll say just a word about a couple other entities that have some roles in this space. One is that there are a bunch of other regional players in this system, right? So if you’re not in an RTO, you still have to be a part of regional transmission planning efforts. All utilities belong to some regional entity that sits below NERC that does sort of regional level planning, and provides like regional data for the big assessments that NERC does.

And then, of course, there’s another huge player in this system, which is the states, right? So the way that jurisdiction is set up in the US, FERC, as I mentioned, controls transmission and wholesale sales. But states control their distribution lines. So the smaller lines within their territories, they control retail sales—so, how much you and I pay for electricity. And then they control one other really big thing, which is, they decide what kinds of electricity gets built within their state. They control generation. And some of them do this through, still sort of very comprehensive planning, and some of them have ceded some of this authority to RTOs, to essentially decide what gets built through regional markets.

Macey: If I could jump in. Because I think Shelley’s answer is completely right. It is clear, and it is complicated, and it also covers up how complicated some of it gets. The reality is that the specific entities Shelley mentioned, the governance is even more complicated than just that we have these layers within governance entities. So for example, NERC is a member-owned organization that promulgates reliability standards, but it also does reliability assessments. Those reliability assessments are typically conducted by sub-NERC regional reliability entities. Those regional reliability entities typically consist of utilities that operate in the region.

One way it’s complicated is we have NERC layered under NERC, which does some reliability standards which are different from the resource adequacy standards that RTOs do. And that’s sort of a horizontal complication. You have RTOs that do resource adequacy, and you do NERC that does reliability standards. But then there’s a vertical complexity, which is that NERC itself delegates a lot of work to sub- regional reliability entities, which delegate a lot of the work to the utilities themselves.

And so in terms of lines of accountability, our state and federal regulators don’t just look to NERC or the RTOs. They look to NERC, the RTOs, they look to the sub-regional entities within NERC. Within RTOS, there are— it’s usually a complex array of actors on subcommittees that develop specific resource adequacy standards. And time and again, it’s, A, difficult to keep track of these lines of accountability— but the result is, time and again, utilities themselves are in decision-making roles. Though often opaque ones that are very difficult to see, not just as a casual observer, but even for people who spend most of their days trying to track these things.

Stone: So what you’ve described here is an absolutely byzantine system of governance for the electricity system, where there are so many different entities that are responsible, and each of them has has a role to play. And they are not necessarily coordinated, which is one level of complexity we’re going to get a little bit more into in just a few more moments. But I want to bring in another issue that you highlight with a very interesting quote from the paper. Shelley, you and your co-authors write that “Those outside the energy industry are often surprised to learn that core decisions about the US electricity grid are made by functionally private entities.” Could you tell us a little bit more about that?

Welton:  Yeah. And I think Josh was starting to get at this thicket of ways in which it turns out that electricity governance is largely privatized, right? And I’ll say, I think everyone’s dream is that they wouldn’t ever have to know about this, because the system just works. And they don’t have to understand NERC or RTOs or any of the ways that this is comprised. And I think that that was true for a long time, but the dynamics that we’re talking about here that are shifting the system, we argue, are making it much more difficult to trust that a system that is run by private entities can actually do the job.

So let me talk a little bit about how these entities actually make their decisions, and why that’s giving us real pause in the clean energy transition. So maybe I’ll start with NERC, which, at least to me, has a really fascinating history.

So utilities, originally, as they were starting to develop their service territories— late 1800s, early 1900s— were working in individual, vertically integrated territories, where they all built all the equipment that they needed to serve their customers in their territory. They pretty quickly realized that there were gains to be made from coordination, and so began to share power. Began to form some cooperative pools where they would back each other up and have ties across their systems.

And by around the 1960s, they realized that they created this interdependent system where because they were connected, if one of them failed, it could have these cascading consequences across many other entities. And so they decided that they would voluntarily, as a matter of self preservation and their own interests, get together and form a group that would set standards, basically, so that one utilities’ generators couldn’t screw the whole system by not being up to snuff.

And so they got together, they created, the original NERC. And things persisted this way for a long time. And they basically said, you know, “Congress, don’t worry about this. We’ve got it under control. We care about having a reliable system. And so, of course, we’re going to set the right standards, and you can just be hands off about this, because we’re self interested in having a reliable system.”

This all began to fall apart when we switched from having vertically integrated utilities to a system where generators are competing to sell power into centralized markets, and utilities are buying that power. These generators don’t necessarily have incentives to maintain the same standards. Because, you know, they’re not running an integrated system where they could each cause the other’s demise, right?

And so in the 1990s people begin to question, does it still make sense to have industry self-regulation of electric reliability? This gets batted around Congress for a little while. And then in the early 2000s the big northeastern blackout happens. And this really brought reliability back to the top of Congress’s agenda. And a lot of people said, “It’s time to make a public reliability organization. Right? We need a public entity that’s setting standards, that’s making sure the grid is reliable.” And NERC basically went to Congress and said, “No, you don’t.” Right? “What you need is a little bit of federal oversight, but you should still have the industry doing this, because we understand the system the best. We know the technical details here, we’re best poised to basically figure out what we need to have a reliable system.”

And so we settled for kind of a weird compromise, where Congress said, “We’re going to create an electric reliability organization. FERC is going to oversee it, and FERC can decide who they certify.” But everyone always knew that it was going to be NERC that got certified. So they are now certified, but you still have this group of industry incumbents that sets standards for grid reliability through what is functionally like a membership club. Right? So all of the generators and other bulk power participants get together in committees. They propose standards. They work through the standards in the committees. They vote the standards out of their committees. They send it on to the board, and the board proposes these standards to FERC, which basically has to, pretty deferentially, accept most standards.

Stone: So, Shelley, you’ve just described a situation where private entities are responsible for the reliability of the grid, that a person who I work with in this industry has described as a situation where the inmates run the asylum. And there’s a very interesting quote that I actually wanted to ask you about, Josh, that also comes from the paper, that I think gets directly to this problem. The paper states, “NERC struggles to produce sufficiently stringent standards in instances where these standards would impose substantial costs on generators.” Talk about that reality.

Macey: I think one way to answer that is to go back to one thing Shelley was emphasizing, and to maybe put it in kind of more obnoxious academic terms. Which is, there is a set of conceptual reasons why academics think self regulatory organizations work in some circumstances. And one reason is, self regulatory organizations are typically thought to be effective when there are clear lines of accountability, such that some regulatory body can actually monitor what’s going on and sanction misconduct. And the second interesting reason is self regulatory organizations often are thought to really require a kind of harmonization of interest. So every now and then, the nuclear industry is cited on the idea that if a single nuclear reactor has a problem, the popular response is panic. And so it will negatively affect every nuclear reactor.

And one of the things Shelley was getting at is that in the early days of the electricity industry, it is possible that these conditions applied. Individual utilities controlled their service territories. They would interconnect with utilities nearby, but they were sort of understood to be the entity responsible for keeping the lights on in that service territory. But as the market has changed and as the resource mix has changed, none of those conditions apply anymore. What you see after every single reliability failure, is everyone points fingers at everyone else, right? We now have this cavalcade of entities that participate in grid governance, in developing standards. And they can all blame everyone else.

And second, people seem to buy that other people are responsible. So after winter storm Uri, many politicians in Texas, along with certain industry participants, pointed at wind and solar. Wind and solar are certainly not perfect. They don’t solve our reliability issues in every instance. But during winter storm Yuri, the biggest culprit was gas. But because there was this narrative, it was reasonably easy to say, “Well, it’s not our fault.” And in the wake of winter storm Uri, we’ve seen interventions that increase revenues available to gas.

And so we have a situation where there’s no longer clear lines of accountability. It’s easy to engage in finger pointing, and we don’t have an alignment of interests. And that brings us to your actual question, which is, we don’t see penalties. And one of the interesting things about NERC is it actually has authority to bring enforcement actions. But as Shelley was mentioning, it’s a member-owned organization. And one thing we’ve observed, though causal inference is tricky here, is despite the increasing frequency of reliability challenges, NERC has been reluctant to bring enforcement actions. And when it has done so, it’s been very light slaps on the wrist.

And this is going to tie into something that I think we’ll get into later, about, why are resource adequacy markets not functioning? Once again, all of our regulatory entities are highly reluctant to actually force the utilities that they regulate to pay penalties when things go wrong. And one explanation for that that draws a great deal on Shelley’s work is, it’s really, really hard to penalize the entity that governs you. There’s just an obvious misalignment of incentives. And so, you know, while we can’t prove that causally, we think it is consistent with the incentives of NERC and RTOs, that they would be reluctant to really rely heavily on sticks rather than carrots when trying to keep the grid reliable.

Welton:  So I think, you know, one way in which you saw this manifest in truly, probably ways that you know caused death, was that after— I can’t remember what the 2011 storm was that got FERC thinking about this. But there was a big storm around that time where FERC said, “You know, what we really need to be doing is weatherizing a lot of these plantswomen” And suggested that NERC take up a mandatory weatherization standard, which would have required all plants to be able to withstand more extreme cold. And NERC started to run it through its process, and it just didn’t get traction. And what they ended up saying was, “We think that plants, in their own self interest, will weatherize, and that NERC doesn’t need to do anything to make them do so.” Right? And, you know, sort of the theory governing Texas was, “We’re going to have an energy-only market. And plants that do weatherize will see such upsides for being available at peak market prices that that’s going to drive them to weatherize. And so we don’t need to do anything.”

And it turns out it didn’t drive them to weatherize, right? Natural gas plants experienced extreme crisis under winter storm Uri conditions. And if NERC had been more in line with what FERC thought should have been done at that time, our public regulator, a lot of those plants probably would have been weatherized already.

So I talked a little bit about NERC and NERC’s governance. And I just want to point out that you have a very similar dynamic at work in RTOs, right? So, RTOs are voluntary regional organizations that utilities join if they want to, and they turn control of their transmission, though not ownership, over to the RTO to run the system. And you have a very similar structure to RTOs. These are membership organizations in which entities get together in committees to discuss— in this case, it’s not reliability standards. It’s really important things like, what are the rules of transmission planning going to be? What are the rules of market entry going to be? What are the rules governing how you interconnect to our grid going to be? So, really foundational rules about what kinds of resources can participate in the system and in electricity markets are decided by committees in RTOs.

Similarly, there’s a board at the top. You know, these committees pass their recommendations onto the board. And, you know, in many instances, the board does have independent authority to override these decisions. But rarely does it do so, because these are fundamentally membership organizations where they can back out if they don’t like the way that it’s being run.

And one other point that I’ll make on this is just— and this gets to what Josh was saying earlier. These are two different entities governing two different parts of the grid, but their membership is largely overlapping, right? So you have membership organizations of the same industry incumbents that are setting both all of the market rules, basic resource adequacy principles, and all the reliability standards. So it’s a very entangled, privatized system.

Stone: On the NERC side, you’ve also written that when it comes to solving some of the reliability problems, NERC has tended to look towards traditional fossil-fuel-based solutions rather than other solutions that may, for example, imply the use of more renewables or storage. Could you talk about that?

Macey: One thing that NERC does is it publishes these reliability assessments. And one thing that we were surprised to learn when writing this paper was NERC doesn’t actually write the reliability assessments. It outsources— the regional reliability entities that sit underneath NERC write them. And typically, they delegate considerable work— at the very least data collection— to the utilities themselves that operate in that service territory. And usually these reliability assessments contain recommendations. “We don’t have enough gas on the system” is a very frequent one, especially in the southeast and in New England.

And these reliability assessments have the NERC stamp on them, and it looks like they’re coming from this neutral body. But it actually—one of the claims we made is that they’re effectively written by the utilities themselves. And their recommendations typically align with the financial interests of those utilities that participated in drafting. So when you see NERC reports that call for more gas, they’re very often in regions where utilities have proposed additional gas investments in their IRPs. And there are often many reliability solutions to these issues. Additional investment in transmission, additional investment in storage. And it is extremely rare to see a reliability assessment that considers a menu of options about how to meet the reliability challenges that that region faces. What you typically see is a proposal that aligns with the proposal that the utilities in the region are making in other domains. So that’s one way we see what we think of as a kind of predilection, presumably even a bias, towards fossil resources. But the fossil resources that align with the utility zone investment preferences.

Stone: Josh, you just mentioned transmission as being one of the possible solutions to the grid reliability challenge. And it’s pretty clear, and it’s pretty well established, that a bigger grid is going to be needed to maximize the potential of clean energy. But a reliable gas supply system is also needed to ensure reliable gas-fired generation. And this highlights another reality that we’re increasingly seeing, that the electric system has become increasingly reliant on a natural gas supply system that lacks its own reliability oversight. Can you tell us more about this, and the operational vulnerabilities that this might create?

Macey: I do think that electric gas harmonization is a really important issue, and one that we’re not dealing with. And it’s something that Shelley and I didn’t actually address directly in our paper, though perhaps we should have. So, to give a few examples. As you mentioned in the question, the reliability of the electricity system often relies on a reliable gas system, but the two are not harmonized.

So during winter storm Uri, one problem— one reason, it seems, that that gas-fired generators could not actually get gas, is that the while the ERCOT, which is the grid operator in much of Texas, regulates the electricity system, the Texas Railroad Commission regulates the gas system. And the Railroad Commission had mandatory curtailment rules that explain exactly who pipelines will serve when they can’t serve the needs of all of their customers. And what the Railroad Commission says is, you need to first provide gas to hospitals, schools and churches. And don’t get me wrong, I fully support, certainly, that hospitals and schools get gas before many other consumers. But I— you know, we can quibble about churches. But the way that this curtailment rule works is in tension with the way that Texas’s electricity market had worked. And the reason for that is it means, if you are a gas-fired generator, the incentive you have to make investments to make sure you can operate during extreme weather events, is you make an ungodly amount of money during extreme weather events. Right? If a one gigawatt gas fired generator had operated for two and a half days in Texas during winter storm Uri, it would have made close to $950 million. To give you some perspective about that, the entire fixed costs of building a one gigawatt gas-fired power plant are about $950 million. I think $960, but I’m not sure about that. So you would have collected the whole fixed cost of a 40-year asset by operating for two and a half days.

And one mystery that a lot of— you know, Shelley and I don’t always agree. I’m a little bit more of a markets person than she is. And one of the things that people said— and I don’t entirely disagree— is, “Oh, the Texas model obviously didn’t work.” So in some work that— you know, promising $950-odd million dollars is not working if the generators don’t make investments. And I can’t prove this. But a challenge, I think, is that these curtailment rules really diminish a generator’s incentives to make these investments. And the reason is that if a generator can’t enter into a contract for firm supply of gas, it is unlikely to make investments in preparedness for extreme weather events, even though it could get $950 million, because it doesn’t know that it will be able to get the gas.

So that’s just one example of how we have a set of rules about how the gas system works in Texas, and we have a set of rules about how the electricity system works, and they don’t align. And so that means that the resource adequacy markets we create in the electricity system don’t actually work well with the set of rules about reliability in the gas system.

Another example is that we’ve often allowed force majeure contracts with gas suppliers that say, “In an act of God, you can get out of this contract.” And, you know, I’ve struggled to get good data on this. If anyone has has suggestions, I’d be extremely grateful for my own research interest. But it does seem, at least anecdotally, that in extreme weather events, gas suppliers declare force majeure such that that creates a difficulty. It makes it difficult to actually guarantee you can get gas when, at the very moment we most need it, suppliers can get out of their contracts. And there’s even some suggestion I’ve heard— but I want to be very clear, I have not been able to prove this— that suppliers will declare force majeure and then sell it to another buyer at a higher price.

And that would be, I think, an illegal anti-competitive practice that we should carefully bring enforcement actions against. But the point being, to bring it back to the animating thesis of the work Shelley and I have done, across all of these domains you have a need to think about how these rules interact. But you have regulators that look at the gas system, the electricity system, the resource adequacy system. And that creates tensions that lead different reliability rules to operate at cross purposes with each other.

Stone: So Josh, we’ve just talked about the complexity, the silos here. So many hands in the pot when it comes to governance for the electric grid. And that fundamentally makes change difficult because there’s a lot of touch points, to keep things very simple at this point. But I also want to point out that the grid operators themselves have looked into solutions to address the reliability challenges that are becoming increasingly worrisome. And one of the biggest— or the biggest grid operator in this country, PJM, which is here in the Eastern United States— recently pushed through a few reforms to its capacity market. That’s the market that it uses to ensure that they have enough generators available at any given time to theoretically, and hopefully in reality, meet demand at those times. Could you talk about PJM reforms, and are they on the right track?

Macey: What PJM has done is it’s modified how it accredits resources that participate in the capacity market. So just to go back a second, a capacity market, as you said, is a market that compensates generators for being available to sell electricity, regardless of whether they sell it or not. It’s a solution to that peak demand problem we talked about earlier when we talked about resource adequacy.

In order to participate in a capacity market, you need to be accredited, which is, we basically use some measure to say how much will you contribute to meeting the system’s reliant resource adequacy needs? And so you might have 100 megawatts of nameplate capacity. Meaning, if you everything is going right, you can provide 100 megawatt hours of electric energy.

Stone: And that nameplate is literally the nameplate on the side of the power plant that says, “This power plant can output 100 megawatts of capacity.”

Macey: Pretty much. When everything is going right, when, when nothing is going wrong, if it’s producing as much electric energy as it possibly can, that’s its nameplate capacity. And accreditation will say, “But we know you’re not always available. So in some sense, we’ll say you can operate, you know, 20% or 30% or 60% of the time. And so we’ll basically treat you as though your capacity is 60 megawatts, not 100 megawatts, because we know you’re not always going to be around.”

And so what PJM has done recently, it has shifted accreditation to something called “effective load carrying capability”, or ELCC. ELCC is a measurement of how much that resource will be able to produce energy when the grid is likely to experience electricity shortfalls.

Stone: That 30% or 40% that you just mentioned, right?

Macey: Exactly, yep. And often, we’ve used a historical approach. And ELCC basically moves to more sophisticated probabilistic modeling about how to calculate this. Ideally, it happens on the unit-specific level, rather than every resource of that type. But basically, grid operators run all sorts of simulations where they look at when will there be extreme weather events? What is the expected load or demand at that time? How much will intermittency lead to swings in supply? So that we can look at each individual’s resource contribution to reliability.

And one thing that we do when looking at this is we say, “If we took this resource off the market, how much perfect capacity would be required to replace that unit?” And so if you take 100- megawatt solar facility off of the market, if you have a hypothetical mythical generator that doesn’t actually exist that operates 365 days a year, 24 hours a day, you might find you only need a 30-megawatt generator to replace the solar facility. And in that case, we say, “Okay, the solar’s ELCC is 30, not 100.” And so it’s compensated as though it contributes 30 megawatts of capacity, not it’s 100 megawatt nameplate capacity. So that is the accreditation reform that PJM and other grid operators have been thinking about.

Stone: So it’s basically trying to understand, really, what is the actual capability of this power plant to provide power in some of the most critical hours of the year? And, you know, if you take a coal-fired power plant, traditional fossil fuel plant, you have control over the fuel. And that is generally going to be a higher rating. If you take a renewable, because you can’t depend on it for the intermittency reasons, right, you’re not going to have that high of a number. And also, if you have a bunch of wind and solar plants or a bunch of natural gas plants, the individual capacity value of each of those is lower because it’s likely that they are going to correlate in their outages. If one goes down for a reason, like there’s a cut-off in a gas pipeline and there are multiple gas plants tied to that pipeline, multiple gas plants will, in a correlated fashion, go out. So the more you have of any type of resource in general, the less reliability contribution it makes to the grid. Is that accurate?

Macey: Exactly. I mean, accuracy is sort of the relevant word here. ELCC is trying to make our capacity markets more accurately predict whether resources will be available when needed. And so my own view is that getting resource adequacy markets right requires two things. One is, you have to get the accuracy right. You want to know that resources will have a reasonably accurate way of predicting whether you have the types and portfolio of resources needed to meet peak demand, so that we can maintain a reliable system. The other thing you need is incentives so that the resources actually do what they’re supposed to do.

So one of the things I’ve been worried about is that while ELCC, I think, reflects an improvement in the accreditation process, we’ve been less successful at improving the incentives that those resources have to perform. And in capacity markets, one reason for that is that non-performance penalties have historically been far too small.

The way that you’re compensated in a capacity market is you sell capacity, and you receive a great deal of revenue for selling that capacity. Then in real time, if you’re unable to meet your capacity obligation, if the grid is under stress and you are supposed to be able to operate, you pay a non-performance penalty because you were not meeting your obligation. And until recently, non-performance penalties have been much too small. And what this means is, if you’re a resource, you have some incentive to over-promise your capacity so that you’re paid more upfront, because the penalties for non performance don’t actually give you an incentive to operate when needed.

And recently, in both PJM and in ISO New England, we’ve seen reforms to increase the stringency of non-performance penalties. And this is also a welcome development. And in order to get a reliable system, we should accurately predict whether resources will be able to operate, and then we should have some amount of carrots or sticks to make sure that they do.

Interestingly, though, as non-performance penalties have gone up, we’ve seen two problems with getting incentives right. The first is that when grid operators have tried to actually enforce non- performance penalties, there’s been considerable pushback. So during winter storm Elliot, many, many resources were unable to operate. They didn’t meet their capacity obligations. They faced a very, very high non-performance penalties, and then they threaten to sue. They reach a settlement agreement, and the non-performance penalties go down a great deal. Now, you could think of this as aligning with a lot of the governance problems Shelley and I have talked about. If a non-performance penalty is levied against a generator that participates in the governance of the grid, what we’re asking RTOs to do is bring penalties against the members that own them. And so it is perhaps unsurprising that grid operators have seemingly, thus far, been reluctant to bring significant no- performance penalties, even when they’ve had rules to do so.

The second problem is that we’ve seen, during both winter storm Uri and winter storm Elliot, bankruptcy by the unit responsible, in a capacity market, for paying a non-enforcement penalty in Texas’ energy-only market— the retail supplier that has to pay astronomically high prices. And what this suggests is that we have what we might think of as a counterparty credit risk issue. That in a capacity market, a generator, rather than paying its non-performance penalty, could simply default on its obligation. And in doing so, not pay for it.

And so what you need if you really want to get resource adequacy markets right, is you need high non-performance penalties, good accreditation, and you need to make sure that generators can pay-non performance penalties. And that probably means they need to be able to hedge. They need to— there’s a requirement that they get insurance such that they can pay. Or the grid operators themselves force generators to keep enough capital on hand that we know they have cash available to pay high non-performance penalties.

And these kinds of reforms would be extremely costly to the generating units. And we have not seen— as non-performance penalties have increased— and again, ISO New England might be the exception here, where it does seem to be thinking about the counterparty credit risk issue. But you also need to make sure they can pay. And because that is so costly, it’s, I think, unsurprising, that that reform has— the RTOs have been reluctant to do this kind of thing. They would be basically saying, “We’re going to bring a stick against the some of the generators that participate in the governance.” And so just like NERC has been reluctant to bring enforcement actions against its own members, it seems unsurprising that the RTOs, if the two things they need to do are get incentives right and get more accurate, the getting incentives right is the one that is costlier for their members. And that’s the thing that they’ve been most reluctant to do.

Stone: Well, so that’s a very interesting point. So as you’ve just very clearly stated, the RTOs, which are member organizations, are essentially going to be very reluctant to levy additional costs or financial obligations upon themselves. And Shelley, this gets to the next point of the paper. How might governance, therefore, be de-privatized and consolidated to remove some of these conflicts of interest and lead to better governance?

Welton:  So, maybe I’ll try to zoom us back out a little bit and just point out, we were just in the weeds of one piece of how you ensure grid reliability. Right? Like, one theory is, you run a capacity market to try to meet resource adequacy, and you redesign it for changing conditions to try to get it to meet resource adequacy under a shifting grid. But in fact, as we started with, there’s a huge array of organizations that actually impact reliability beyond this question of capacity markets and resource adequacy. So, NERC is setting these standards that are supposed to ensure that these generators show up when they’re supposed to. RTOs are designing these markets, and they’re also doing transmission planning. We really haven’t touched that much on transmission’s role in reliability in this podcast yet. But as we’re trying to solve everything on the generator side, transmission is an alternative or additional solution that gets at a lot of these reliability challenges, often at a much cheaper way if we can build the kind of interregional, high-voltage, long distance lines that we need. We have states doing generation planning, so they can be getting new resources to show up on the system through the ways that they run their planning. Interconnection queues can help new resources show up so that you could lower capacity market prices. This is a very deeply interconnected system.

And part of our argument is that we have a real coordination challenge in the ways that we siloed these little pieces off to various entities, when what we really need is somebody that can kind of mind the store and do the whole thing. So I think there’s two pieces of the reforms that we suggest, and part of it is that somebody needs to have more authority to be able to look at this holistically and systematically, and not sort of apportion reliability off into these various little pieces of ways to solve the puzzle. But understand that this is a system that’s going to run as a whole, and we should be planning it as a whole, so that we don’t overpay for reliability and under-deliver on it. Okay. So there’s a coordination, matching the powers to the problem, kind of challenge.

What we’ve been focusing on the last few minutes is the privatization problem, right? And if you’re going to try to match the scale of the problem to the powers that you give an entity, then privatization becomes an even bigger problem. Because you don’t want to give that kind of power to an entity that has incumbent bias, self interest, all of the kind of pathologies that we’ve been tracking.

And so part of what we argue is that, especially if you’re going to consolidate power to really do reliability well, you want to hand that authority to somebody that’s going to do reliability well. And, you know, do it in a way that’s going to be most cost effective, most align all states’ priorities across the system as best possible. Be forward-looking, not backward-looking. Resource neutral, as best it can be, in terms of how you come up with solutions. And so to do that, our argument is, you’re going to have to stop relying so much on these membership clubs to do both your standard setting and your rule setting.

Macey: And just to put Shelley’s point in maybe dramatic terms, one way you can think about the effect of— we have essentially outsourced reliability planning to the firms themselves. Those firms often have significant, arguable market power in both generation and transmission. And holistic approach to planning would reduce costs. It would facilitate decarbonization. It would also reduce market power in wholesale power markets in the generation side. And it would reduce transmission owners’ ability to preserve their monopoly over transmission.

And so one way you think of what Shelley, I think, referred to as a reliability Fed, or a central coordinator for reliability, is that we really need some way of making sure that entities that want to use their authority to plan investments, to protect their own market power— we need some way to prevent that from happening. And that looks like a central coordinating entity.

Stone: And that’s the reliability Fed that you just referenced.

Macey: Yes.

Stone: Okay, so that reliability Fed. Where would that be housed?

Welton:  I will say, I think that we tried to be measured in our piece, and both lay out a grand vision, and then lay out some smaller steps that could move towards that vision. So I think the reliability Fed concept is our grand vision, right? If you wanted to really plan and run a system for maximum reliability, least cost, under decarbonization conditions, our proposal is that probably that should be housed in FERC. You should give the power to your federal regulator that already, at least nominally, is charged with ensuring reliability, but has a pretty scatter shot set of tools to do it. You should give them the tools that they need to actually do reliability and meet their mandates to run a just and reasonable system.

And we make this analogy to the Fed, because if you think about the the Fed’s role in the financial system, they’re given a real like coordinating set of tools. So they can set interest rates, and they can sort of monitor banks and make sure that they are doing what they need to do to be stable.

Macey: And it has emergency lending powers when needed. It sets capital requirements. So it essentially has some amount of authority to do bank-specific regulation, supervision and monitoring, to make sure that we have a resilient financial system. And it has authority to step in when there’s a financial crisis. And I think that’s a reasonably good model for the utility system. That you need to stress test utilities. You need to do scenario planning across a very large system to look at what set of investments would best address our reliability challenges. And you need to have a entity that’s not captured by incumbent interests, that can bring enforcement actions when things go wrong.

Stone: So this concept of a reliability Fed that would be housed— potentially, ideally, under the scenario that you’ve just been talking about— within the Federal Energy Regulatory Commission, the FERC, would imply that the FERC would have additional authority. And some of the authorities that are distributed throughout the industry, let’s say, would come more to the FERC directly. Now, that sounds— you know, great to have a one-stop shop. But the fact of the matter is that there has been some opposition to the FERC having additional authorities. And we saw this under the first Trump administration. We’re now one week after this fall’s election, and Donald Trump is coming back in. What does that potentially mean for having all of this authority concentrated in one area? And what is the argument that that administration used in the past, and I suppose will use in the future, against allowing the FERC to have additional authorities in these areas?

Macey: One thing I think we need to mention is that a FERC Fed can’t possibly do all of the reliability things, right? There are technical standard-setting issues where, A ,it may not have the competence to do so. But perhaps more importantly, much of the data is only housed in utilities themselves. So some of the information needed to do planning, as well as the technical expertise, is something that would be difficult for a federal entity to do.

I still think, and I think Shelley agrees with me here, that you want a central entity to develop best practices for resource adequacy markets without having to let utilities do so first. To do transmission planning, to sort out cost allocation. But it might be equally important to do things like mandate information sharing and transparency requirements, so that both FERC and third parties actually have an insight into some of the really technical criteria that would otherwise make it difficult to do this.

Now your question, I think, is basically, we just had an election. At least while campaigning, the Trump administration suggested it would limit federal energy regulatory authority. And the Trump administration has spoken out of two sides of its mouth. On the one hand, Project 2025, along with FERC Commissioner Mark Christie, have been highly critical of too much federal authority over the grid. And that would suggest a more limited role for FERC. On the other hand, during the first Trump administration, there was considerable pressure on FERC to use almost unprecedented tools to favor certain resources.

So, to give two examples. At one point— I think he was Commissioner then. He might have been Chair. But Chatterjee said, “Well, if too much coal and gas retires, we’ll RMR everything.” And what he meant by that is, “We’ll use ‘reliability must run’ agreements, which take a generating unit out of the market and guarantee it a return, and we’ll make sure that certain resources simply stay in the market because those are tools that we have.”

Stone: These are resources that would otherwise retire. Right?

Macey: Exactly, yep. And the other thing that got more press was, there was a proposal that was pushed on DOE and FERC to use FERC section 202c authority, which is its emergency authority, to essentially find out-of-market ways to guarantee coal and gas would get huge compensation for storing fuel onsite. This too, would have— you can think of it as— actually, there are many examples. But these are just two. We could go into MOPRs. All of these instances that seem to require a far more aggressive federal regulator to achieve various federal policies that seem to involve a central coordinator. Which means I’m uncertain about how to predict what the Trump administration is going to do here. Because at times they seem to want to give authority back to the states, but at times they seem to want to push for rules that favor gas and coal, and to do so by using FERC’s own authority.

Stone: Shelley, I want to ask you this question. Just kind of cutting to the chase, could governance move in the wrong direction going forward, in terms of addressing the reliability issues that we’ve been discussing?

Welton:  What I think I really worry about, and I think— you know, pretty much the starting animus for our paper— was that we saw entities using this concept of reliability to promote a particularly backwards-looking, stalling vision of what the grid should look like. And I have real concerns that this is going to rear its head again. That reliability is, in a way, used as like an excuse or a delay tactic against the clean energy transition. So you see the ways in which Texas is now leaning back into subsidizing gas as an answer to reliability challenges, even though gas was central to their problems in ways that we discussed. And you see entities pushing for changes in how resources get interconnected to the grid to prioritize gas resources, essentially. And so I think my big worry is that we lose sight of reliability as a governance challenge that has many solutions at the technical level, and it becomes this sort of buzzword that maintains a fossil fuel past that we absolutely don’t need, from a technical perspective of how we manage the grid.

Stone: Josh, I want to ask you just a final question here. You know, we’ve been talking this whole podcast about consolidating authorities in one place. And as we see an administration coming in that may be deregulatory. As a practical consideration, it may be advantageous that, at least for the next four years, these authorities remain dispersed. That the states maintain and exercise their own authority over grid development, generation, resource mix, things like that. What are your thoughts?

Macey: It is obviously hypocritical for me to say this. But Shelley and I wrote this paper when there was a federal government that wanted to reduce electricity bills, improve reliability and support decarbonization efforts. We’ve seen a shifting political landscape. One might say that that’s democratic. But I suspect that the Trump administration will continue what it did in 2016, of favoring certain fossil resources over other resources. And when Shelley and I wrote, “Here is what a empowered FERC would be able to do,” an assumption was that FERC would actually try to do things that improve rates and support decarbonization efforts.

Now that the political landscape has shifted, there are entities that still have a great deal of power to make decisions about how we operate the grid, and those are especially states’ utilities in areas of the country that have either corporate or state decarbonization goals that they need to meet. And those entities are now— you can think of them as acting counter-cyclically. There are things that they can do. Some of them are quite aggressive. They involve leaving a capacity market and entering into a joint forward capacity market. Some are simply using the authority that states do have under recent FERC regulations to do things. And those are in direct tension with what Shelley and I wrote about last year.

But it does suggest that to the extent that at a federal level, the political landscape starts to look something like something I don’t support, the siloed, parochial, terrible governance has a silver lining. Which is that some of the actors that do want to invest in transmission, invest in clean energy resources, might be able to do more than they otherwise would.

Stone: Shelley and Josh, thanks for talking.

Macey: Thanks, everybody.

Welton:  Thanks for having us.