Aired Jul 21, 2026 · 40 min

Listen on: YouTube · Spotify · Apple Podcasts

The North American grid remains one of the most reliable in the world—but is that changing?

As electricity demand surges from AI data centers, electrification, and new manufacturing, utilities are racing to keep pace while balancing an increasingly complex mix of resources. So how reliable is the grid today? Where are the biggest risks emerging? And what needs to happen to keep the lights on over the next decade?

In this episode of Power Perspectives, Kinsey Grant Baker sits down with John Moura, Director of Reliability Assessment and System Analysis at NERC, to unpack the organization's latest long-term reliability assessment. They discuss why some regions face greater reliability risks than others, how the rapid growth of wind, solar, batteries, and data centers is reshaping grid planning, and why data centers may actually strengthen reliability if integrated thoughtfully. Moura also explains why traditional planning metrics are becoming outdated—and what utilities, regulators, and policymakers should do now to build a more reliable grid for the future.

Whether you're planning for the next wave of load growth or navigating the evolving generation mix, this conversation offers a timely look at the challenges—and opportunities—ahead for North America's power system.

John Moura: We're excelling in many ways, keeping the system reliable, but what looms out in front of us are tremendous challenges, but also opportunities.

Kinsey Grant Baker: This shift in the generation mix more toward wind, solar, batteries, when we're looking at enough supply to meet the demand, how does the generation mix factor into that equation?

John Moura: The risk factors are less about the mechanical failures and more about, well, what's the likelihood the wind will go down? What's the likelihood the sun will be cloudy? The data centers have the capacity to also amplify reliability. There's opportunity within its controllability to provide flexibility into the grid. Reliability really shouldn't be a real-time observation activity. Really is something that we ought to be planning in advance. Then we can kind of work on the affordability and security piece as well.

Kinsey Grant Baker: I'm Kinsey Grant Baker and welcome back to Power Perspectives. Three words every utility pro hears in their dreams. Affordability, resilience and reliability. Of that three legged stool, the last one reliability is perhaps the most crucial. It's reliability that ensures the lights go on when we flip the switch that keeps critical infrastructure like hospitals and airports and World Cup stadiums functioning safely without disruption, that makes sure all of this rampant demand has supply to back it up. Now, the U.S. electric grid is by all accounts highly functional. It works well most of the time, but reliability is trending in a strained direction with electrification and manufacturing and data centers and the rest of the demand-spiking laundry list that you know all too well. The truth is, Americans experience more outages now than they did a decade ago, with the average annual outage time exceeding 360 minutes. Still pretty great — 360 minutes is only about as long as it takes to watch The Matrix two and a half times — but that does mean that our grid is at times less reliable than that of other developed countries.

So, what are leaders doing to ensure that the trend heads back up? Because the grid gets more and more complex, that means our reliability needs are only increasing. Today, we have an energy pro whose job is squarely in that all-important niche. John Moura is Director of Reliability Assessment and System Analysis at NERC. His job? Figure out how reliable the North American bulk power system really is. And today, John is here to tell us what he's found, what it means and where we're headed. John, welcome to Power Perspectives. How are you today?

John Moura: Hi, Kinsey. Thank you for having me on.

Kinsey Grant Baker: Of course, I'm thrilled to have you. We're going to have a great conversation today. This is a topic that a lot of our listeners care very deeply about, as I'm sure you're not surprised to hear. But I want to start by really setting the scene and getting an understanding of that where we are, as I mentioned in the introduction. So if you don't mind, could you sum up the state of the power grid reliability in North America in maybe one sentence? I would challenge you to one sentence. Tell me what's going on right now with reliability.

John Moura: Yeah, I think you opened it up really great and the premise is right on track. Our whole grid has evolved so much in the past couple of decades. We're using electricity in so many different ways than we have in the past. We're excelling in many ways, keeping the system reliable. But what looms out in front of us are some tremendous challenges that but also opportunities. New data center growth can really amplify the grid's reliability, not challenge it if we plan it correctly. So I'm excited about the future and excited about talking a little bit more about the challenges we have in front of us.

Kinsey Grant Baker: That's a great way to put it, that these challenges and opportunities can really coexist in one specific example like data centers, but I'm sure across the board too. So if you had to give a grade to grid reliability these days, what would you give us here? What would you score us in the U.S. in terms of reliability right now?

John Moura: Yeah, it's a tough one. You know, different parts of the country are, you know, I would say different risk levels. We don't have kind of traditional grades in our assessments. But when we do look at reliability and try to communicate those risks, you know, there's not really any one metric that tells it all. Reliability is a pretty complicated thing to comprehend. You can think about it around, you know, how much capacity areas have, or you can think about it around how stable certain areas are, making sure they have enough transmission. So there's kind of many ways to look at it. And so when you look at different parts of the country, you know, some areas really have a lot less risk factors. It's kind of how I would put it than other areas, right? We can't really predict their performance, but we can look at the risk factors.

Some areas have, for example, retired a lot of generation over the years. A lot of that generation was highly certain, highly reliable, right? Kind of, you know, you're able to look, you know, in your backyard, look at the coal piles and say, OK, one month, two months, three months of coal. I'm good, you know, under almost any emergency condition. But now that we've really retired some of those units, we've brought on more wind and solar, which is good during real good times of wind and sun, but might not be as good during. For example, winter peaks when the sun's not really shining and the wind's not blowing. So there's challenges that certain areas have to overcome in the future. But what we know for certain is that this is pretty much the riskiest the power grid has ever been.

For many decades, we've had ample reserves. And over the past couple of years, we've retired and retired the generation. So there are challenges looming. But I remain opportunistic about the future because the large loads, those data centers that are coming online, while they can be challenging to meet, really represent a huge boon for the industry and also really helps address the affordability challenge I think you mentioned in your opening.

Kinsey Grant Baker: What would you say are some of the biggest factors that create that reality that you mentioned that we are at a risky time for the grid? Tell me a little bit more about what what informs that risk.

John Moura: Great follow-up. So, you know, let's take a look at an area like Texas, and let's also compare that to California. Two standout areas that I think highlight a lot of the challenges that we're seeing. So in California and in Texas, both areas have been able to build a lot of solar and storage resources. That's really been helpful for meeting summer peak conditions. Sun shines during summer peak, and as the sun goes down, this four-hour storage can jump in and help smooth out some of those peaks.

Now, in California's case, that has reduced risk and it pretty much works all year round. They don't really have those harsh winters every now and then. Texas, very similar strategy, and they built out an incredible amount of solar energy and storage has been very helpful for meeting summer peaks but in the winter time solar and storage isn't very helpful some winter peak happens early morning and early evening when the sun's already down and storage really only lasts about four hours so not very helpful so the risk factor right you know they again they both had the same build out strategy. But the risk factors in Texas show increasing risks that are happening during those winter conditions because of some of the generation choices that have been built out.

Another risk factor is increasing amount of natural gas that's being utilized with maybe not as many pipeline connections to increase that resilience of pipelines. Again, we're counting on natural gas to help us be resilient against winter conditions. But time and time again, we've had some challenges in both the production, real-time production of natural gas, you know, winter freeze-offs of the fuel sources. But also some pipeline constraints where not enough gas gets to the generators that need that gas, may have not procured the gas, but still need it. And so that is an increasing factor that we see as well.

So those are just a couple highlights. You can go on and on about, you know, where the concentration of large loads and data centers are, where wildfires are, where GMD, geomagnetic disturbance risks are, where drought conditions prevail. So, you know, when we do our assessments, kind of look at all these risk factors. And our grading is more of a risk factor. Which is a normal risk, elevated risk and high risk and really the culmination of all those risks come into how we how we risk rank different areas across the country.

Kinsey Grant Baker: Yeah, and it's an important point, John, that you make that no two territories, no two regions of this country are the same when we're looking at the risk that they're facing, when we're looking at the reliability realities that they're facing. And I think that we'll dig more into kind of what that looks like. But, you know, across the country we have seen and across the industry we have seen this shift in the generation mix for energy, broadly speaking. And this was a big part of what you were just mentioning. Right. Do you see this shift in the generation mix, you know, more toward wind, solar, batteries, knowing that they're cheaper to add to the grid, knowing that they can serve certain purposes for power generators here? Does that impact resource adequacy in any way? You know, when we're looking at enough supply to meet the demand, how does the generation mix factor into that equation?

John Moura: Almost everything in the power system contributes to that resource adequacy. The way we integrated into our assessments as we think about resource adequacy is trying to understand all the different constraints on the system. You know, I like I like an analogy that I think might be useful here. Well, let's try it out. So if I'm a casino owner, bear with me. I might have to do some analysis to determine, well, what's the rare possibility and off chance that I'll lose a million dollars in a day, right? They might have to do that type of a rare analysis. And so the way they do that analysis is they look at all the different games that they have. They have craps. They have blackjack. They have the slot machines, and they total up all the different probabilities and statistics about those machines. And they run what we call a Monte Carlo analysis that runs thousands and hundreds of different simulations and calculates the different potential and outputs of probability distribution of the odds that that area, that casino might lose a million dollars in a day. Right. So so what are all the factors that go into that analysis? Right. Which is really, you know, what's the probability, you know, players might win at blackjack? What's the probability of players might win at slots? So all that goes into that analysis.

Point is, this is the same analysis of resource adequacy and resource planners do. Instead of the games, it's the different generators. And so what the resource adequacy planners are doing is looking at all the forced outage rates of the generators, right? So every generator's got a force. So what's the probability that each generator fails is essentially the analysis. And it used to be, when we were mostly conventional generation, it used to be a statistical operation where you look at the mechanical failure rates of a steam turbine or a boiler or all these components of generators. As you build more renewable, the risk factors are less about the mechanical failures and more about, well, what's the likelihood the wind will go down? Right. What's the likelihood the sun, you know, be cloudy. Right. And so those are much more likely than mechanical failures of generators. And so as you get more, the wind is the point of the wind and the sun, depending if it's windy or sunny, becomes a large contributors to whether or not areas have energy. Because the final point is at low levels, it's almost insignificant. Right.

You know, you can integrate, you know, the first 5% is pretty easy to integrate renewables. But as you get to 10, 30, 40% renewables, you know, you can have these days. You know, it happens so often in Europe. They call it the Dunkelflaute. It's a German word. They call it the Dunkelflaute, which means a renewable droughts. And so, you know, it happens so often they had to create a name for it. And of course, Germans create names for everything. But this is a challenge where the wind dies down for maybe two or three days, right? And the sun is in its cloudy conditions for several days, kind of at the same time. And so those are conditions that system operators and planners have to now plan for.

Kinsey Grant Baker: Yeah, you know, I think this is such an important part, too, of the calculus of renewables in the United States and across our grid. I think it illustrates the reality that so many operators and leaders are facing today, which is a whole litany of factors and circumstances and impacts that have a force that they bring to the table when you're thinking about what you build for the future. I think the other element that feels really significant here is the pace of change. A lot has changed really, really quickly, at least is what it seems like to me when we're looking at the generation mix, at resource adequacy, when we're thinking about what that future looks like. How do you see the sort of impacts having changed over the years? Can you put into context at all the pace of change when we're thinking about reliability and the big picture for the U.S. grid right now?

John Moura: Yeah, it's kind of almost everything. I got a story for that. You know, it used to be, you know, it used to be where, because we've been as an industry integrating large loads and, you know, evolving and changing over time. But it used to be where, you know, you have a large load, maybe 100 megawatts, a large paper mill, you call the local power company and you say, Hey, I want to integrate, you know, 100 megawatt paper mill. And they say, yeah, when do you need it? And they the paper mill would say, well, take me about, you know, seven to ten years to build this thing. And then the utility guys would say, Oh, well, it'll take me about seven to 10 years to build the generation and transmission. And so the pace of change were kind of congruent between the industrial and commercial needs and the industry.

And right now they're incongruent. We can build these loads much quicker than we ever dreamed possible. And so industry is really kind of playing catch up, evolving. I think we're changing pretty rapidly, but we're evolving our processes that were very linear, very not batch work like we're doing now, soundness in the engineering. And in fact, not only the pace is more rapid, but compounded by, you know, the previous analytics I just described, which is kind of compounding these complex challenges. So that pace is much different than it was in the past, even though we've brought on large loads much quicker. And I think everything is kind of really constrained. The reason why we're worried about the pace of change is we're constrained by the the pace that we can bring on generation and transmission. A lot of that has to do with policies at the state level, policies at the federal level, all these challenges in siting and permitting new infrastructure, both on the electric and gas side. So, you know, we're incredibly challenged by that. But my hope is a lot of these challenges have been not only identified, but magnified most recently. And we want to see areas, especially market areas, be able to enhance their structures to advance the most capacity and the transmission build outs.

Kinsey Grant Baker: Yeah, you know, I think we'll dig more into what it looks like to be more productive when we're thinking about the generation and transmission build out and, you know, figuring out that concern, which is significant. But before we do, you mentioned at the very beginning of this conversation that data centers have the capacity to also amplify reliability. Yeah. When we're talking about the pace of change over the last, let's call it, decade, obviously the influx of data centers is one of the biggest parts of that conversation. Tell me a little bit more about what you mean by that. When we're looking at these huge hyperscalers, large loads coming on in concentrated areas, how can that potentially positively impact reliability?

John Moura: So, you know, at the levels that we hear and we are already seeing data centers at, which are, you know, plus 300 megawatts, gigawatts, two gigawatts, these are become really, they must be grid-friendly assets. It can be even more impactful than many generators on your system because they're so large and they're so, they consume a lot of not only power, but voltage, frequency, stability within certain systems. These are pretty advanced loads, right? They're not that paper mill I describe, which essentially has an on and off button. Right. So the paper mill had an on and off on and off button, and it was either 100 megawatts or these loads are highly sophisticated. They're highly variable and there's opportunity within its controllability to be to provide flexibility into the grid. And those are the opportunities because they're a large source of both could be a large source of voltage and frequency and energy. They could actually be grid resource and reduce the amount of transmission costs there would otherwise need to be if you didn't have that as a helpful resource. The other reason that they're very helpful is almost all of them are coming with backup standby generation. So that not only... Adds to the flexibility offerings, but working in conjunction with the local utility can actually increase the capability of the system if there's an opportunity to use that backup generation that comes with it.

Kinsey Grant Baker: Does that positive impact change at all when we're looking at this trend of hyperscalers really opting more for behind the meter, often natural gas behind the meter? What does that kind of trend look like in terms of impacts reliability for the power grid, if anything at all?

John Moura: This term, we need to kind of help clarify a bit, because there's a couple of different options we see out there. And so, you know, there's the, you know, complete islanded system, right? They have the data centers that are completely islanded from the grid that are trying to self-power. My understanding about that is that is not the preferred approach, right? that it's very difficult to get the level of reliability that they need on their own, meaning the level of reliability they can get from the power grid is unparamounted and the most efficient option. And so since that's that, but there are those who are doing kind of, then there's kind of that behind the meter, more traditional sense that we think that they're connected to the grid. They are contracted for some certain amount

And they're also using self-generation to maybe arbitrage or provide flexibility. And what's important about that is grid is still planning to serve that load, but only up until that certain amount, right? And then they have to rely on their backup generation. And even if some of those data centers aren't using... You know, the are self generating on their own more often. You know, the grid is still reserving that capacity based on whatever contracted amount those data centers have.

Kinsey Grant Baker: Yeah, it's helpful context and I think important background to make sure that it's clear kind of how all of these data centers operate. Yeah. Much like no two regions in the US are the same when it comes to reliability, no two data centers are the same when it comes to their impacts on reliability. I do want to talk a little bit more about the data centers as part of this larger tapestry of where we are today when we're thinking about grid reliability.

NERC recently released a long-term reliability assessment. I know this is a project you spend quite a lot of time with. The idea here is kind of predict over a 10-year horizon what reliability looks like. A lot of uncertainty seemed to kind of be the theme of this most recent assessment that, you know, the system is changing really quickly. Infrastructure might not necessarily be keeping up. I would love to hear a little bit more in your own words about reliability. What with all that we've talked about today from the generation mix shifting data centers, resource adequacy in general, what that means to you when you're thinking about the holistic view of reliability in the U.S. in the coming decade? Tell me a little bit more at the high level of what that looks like.

John Moura: Yeah. Yeah. And thanks for the opportunity to talk about it. You know, certainly that report we've been doing for quite some time and we've mostly been doing it for the same way for quite some time. And the value in that is that we can kind of look at the historical record and see where we were 10, 20 years ago and kind of what shaped out over the next 10 years. And really in its full history that we've been doing this, more than about 60 years, we've been producing these long-term assessments. The risk map or our risk assessment of the next five to 10 years really hasn't been this risky really ever.

And I think that that is something that we really need to take hold of, take understanding of. And really, you know, NERC, we're kind of putting the message out there. We can't do anything. We can't really do it on our own. We really need the partners at the state, the federal level, and industry, particularly in market areas, to rectify that posture. Yeah. You know, there's certainly solutions that are being put in place. I like to kind of call them edge of your seat solutions. You know, you might be and listeners might be familiar with, you know, the FERC and the markets advancing resource positions and resource integrations in certain market areas. And those have been very helpful. Yeah.

You know, reliability, you know, really shouldn't be a real time observation activity. Right. It really is something that we ought to be planning in advance, making sure that we've got that reliability foundation, like you described, along with then we can kind of work on the affordability. And security piece as well. So those are kind of the challenges we have in front of us is really this risk posture we're in. And if we don't rectify that, you're kind of always digging, you're trying to always dig yourself out, right? Really try to get ahead of the capacity and transmission process expansion, not behind, because the consequences, I don't think we really want to fully understand economic challenges, national security challenges, economic prosperity, health and safety. I think we've learned that in recent blackouts that have occurred during winter periods in particular. Okay.

I don't feel old, but I'm young enough to remember when we had a blackout when I was younger. It was kind of a fun activity. You went outside and you played soccer for a little bit, went inside, played board games, and you woke up and everything was back to normal. But really, the reality is the winter conditions where we expect some of the outages to to occur, where the risk is popping up even further, is really challenging for health and safety security. So that's going to be a hard thing to understand because we really haven't lived in that world. So we're hoping they can get the message out and so we can really have long term, durable solutions.

Kinsey Grant Baker: And before we dig into what some of those potential solutions might look like, just because I'm curious, what regions are at the highest risk and why? You mentioned throughout this conversation that it sort of depends on where you are, the reliability that you're looking at. Tell me a little bit more about some of the hotspots and the not-so-hotspots.

John Moura: I think we talked about Texas as a hotspot. Those challenges have, it's quite remarkable, right? Where, you know, and for decades, Texas has always been built around air conditioning and summer loads and, you know, prepare for the summer. And the whole system is really designed around that. And what we've come to find is that risk is really transitioning, still some in the summer, but really transitioning to where it's more concentrated during winter. Winter is really where the resource adequacy risks are in Texas. And again, largely because the resource choice, right? They've built a lot of solar and storage, very helpful for summer peaks.

Some other challenging areas are in MISO, in PJM, two of the largest what we call independent system operators, regional transmission organizations. They're across the country for many years. Again, markets have been working as markets do. We've talked a lot about the long history of the system. Yeah.

I think this is a good example where markets came in about a couple of decades ago, helped revolutionize how we think about bulk supply of generation and transmission and utilities and their involvement and have done what markets do. We were probably at the time 60% reserve margins in some places, and markets have made areas more efficient. And in doing that, have really kind of gotten the margins down to right where they need to be. And the challenge with that is, you know, you can kind of hover as you put solutions in place, as you build generation, and as generation might get delayed or deferred, your reliability targets might hover below or above your targets. The risk factor is how long it takes to build a certain generation. And so what contributes to that? Supply chain disruptions, early retirement of resource of generation, as well as the permitting and kind of supply choice in different states. So, you know, PJM and MISO have some challenges that are, have to deal with getting the right resource mix and choice of supply and keeping generation online that kind of wants to retire, but again, are needed for, for reliability. Those are the kind of couple areas of the country.

You know, the other notable area of interest is kind of New England area that's got plenty of capacity, plenty of generation, but suffers from energy, potential energy deficits during your kind of real pro long cold winter. You know, they're at the end of a pipeline. They have a lot of generation that doesn't have firm energy, rights to natural gas generation and so a lot of generation is you know flying standby you know can get can get gas if uh natural gas is available but those long and persistent you know winter cold cold spells um can can really deplete uh not only the gas but also the oil that they have in reserve so that's why that area um gets covered

Kinsey Grant Baker: Yeah, you know, I think you bring up a lot of the big questions that utility leaders are looking at when they're trying to figure out how to build a better grid, right? We're talking about building more effective transmission, making sure that we have the right resource mix. Yeah helping to ensure that permitting reform happens at any rate, right? I want to hear your perspective on what makes the biggest difference here. You mentioned specifically the pace of generation and transmission build-out was a concern. What would fix it? Tell me a little bit about what some of the solutions look like in your view, from policymakers, from utility leaders. What could fix this problem?

John Moura: You know, the beauty about our system, you know, you can say it's very challenging and bureaucratic and so many decision makers that it's hard to get things done. We also can kind of step back and say it's brilliant because everybody has a say. And really, this is how we make decisions together on some of the most important things crown jewels of our, of our, in our, in our, in our country and across North America. The challenges are making decisions and having the right type of data so that those decisions are, can be reputably and confidently, um, it, uh, can, can be informed by those. And so it's one thing that NERC really tries to do really, really, really, really hard. And, um, and kind of get that message about what are the reliability risks and how do we make decisions around whether or not more resources are needed.

And so fundamental to that, at the heart of that, I think one of the things that makes the most difference is, you know, states have what we call resource adequacy criteria. And those are that does this really state's responsibility to to to enforce those. Many of those are dated. Many of those have been relinquished to the ISO RTOs and markets to administer. Some areas are advancing. They're capturing the right types of resource adequacy criteria that's needed in kind of this modern world where we're going to a much different resource mix. Fundamentally, reserve margin. We talked about that a couple of times. It really doesn't make a lot of sense when you have a lot of solar or a lot of wind on your system. It's really a dated metric. And if you're building only to a reserve margin, you're, you have maybe a lot of solar and wind. It's really an inefficient, really ineffective way to assess reliability and to be confident about where you are and and therefore very difficult to make decisions. So a good example of this is.

If you've got a lot of wind and solar and you have maybe a 15% reserve margin target, but yet you know there are challenges on your system, energy challenges during certain times of the year. It's going to be hard to write any checks or to justify additional infrastructure in those systems because they're above the reserve margin target. They meet criteria. It's like, well, yeah, but we have to understand that that criteria, it really isn't, doesn't have any energy engineering soundness to it anymore. Now that the resource mix is so drastically changed. So there's this huge, I think, transition of learning and instituting better resource adequacy policies and criteria yard. We are seeing advancements there. Um, but I think that that is the biggest thing that can really move, um, and help make decisions for reliability.

Kinsey Grant Baker: Yeah. I mean, context is king, right? And understanding how...