Power Struggle: The Electric Grid’s Natural Gas Challenge

Over 40% of U.S. electricity is generated by gas-fired power plants yet, double the role the fuel played two decades ago. Yet the past few years have exposed risks arising from our growing reliance on gas-fired generation.

Major power outages in Texas and the Eastern U.S. have highlighted the fact that gas generators are vulnerable to disruption of the natural gas supply networks that fuel them. And, while the electricity and natural gas systems have become increasingly interdependent, there remains surprisingly limited coordination of the planning, operation, and regulation of the two industries. This fact complicates efforts to address reliability concerns.

Seth Blumsack, director of the Center for Energy Law and Policy at Penn State University, discusses the challenge of coordinating the nation’s natural gas and electricity systems as gas has become the predominant fuel for generators, and a key balancing resource for intermittent renewable energy. Blumsack explains the growing interdependence of the nation’s natural gas and electricity networks, and explores efforts to address reliability concerns through better coordination of the systems.

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. Natural gas has become the most important fuel for electricity generation in the United States. Today, over 40% of the nation’s electricity is produced by gas-fired power plants, double the role that fuel played just 20 years ago. Yet the importance of gas to our electricity system extends beyond its simple preponderance as a fuel for generators. Gas-fired power plants are flexible and can quickly raise and lower electric output. This makes them well suited to balancing the variability of wind and solar power and a complement to carbon-free generation.

Yet the past two years have exposed risks arising from our growing reliance on gas-fired generation. Large-scale power outages in Texas and the Eastern US have highlighted the fact that gas generators are vulnerable to disruption of the natural gas supply networks that fuel them. And while the electricity and natural gas systems have become increasingly interdependent, there remains limited coordination of the planning, operation, and regulation of the two industries. This fact complicates efforts to address reliability concerns.

On the podcast, we will explore the challenge of coordinating the nation’s natural gas and electricity systems with Seth Blumsack, Director of the Center for Energy, Law and Policy at Penn State University. Seth’s research looks at energy system risks and at planning issues that impact the resilience of the electricity system. He’ll walk us through the challenges relating to gas and electric interdependence and look at efforts to better coordinate the systems. Seth, welcome to the podcast.

Seth Blumsack: Thank you. Thank you for having me. It’s great to be here.

Stone: In the last 15 years, natural gas has become a much more important fuel for electricity generation. Could you provide an introduction to why and how the role of gas has changed?

Blumsack: Yes, I think that what you’ve hit on is just a really fundamental way that our power system has changed. If you look back 15 or 20 years ago, natural gas was really kind of a marginal fuel in many parts of our power grid. What has happened and what has changed that, in a sense, can be summed up in a single word, which is “fracking.” So this technological development that has just opened up huge natural gas reserves in the US has really paved the way for gas to really displace coal as the major fuel for power generation in many parts of the North American power grid.

Stone: So this dependence on gas can create vulnerabilities, as we’ve seen in recent years with Winter Storm Uri in Texas in 2021 and Elliott in the Eastern US in 2022, where we had some outages. Can you tell us about the storms and the vulnerabilities that they revealed?

Blumsack: Both Winter Storm Uri and Winter Storm Elliott were events that were pretty extreme. You had very, very cold temperatures in Texas and in swaths of the Northeastern US. They were pretty extreme, but not unheard of. Texas, for example, has these kinds of extreme winter cold events about once every ten years, if you look back historically. These were extreme but not unheard-of events. What made both of these storms really difficult for grid operators was that basically when these storms happened, you had a couple of things go on. One is you had a simultaneously high demand for both gas and electricity, which is historically something that grid operators have not always planned for. I think they are thinking more about those kinds of simultaneous peaks now, but historically, that’s a pretty unusual thing. Grid operators have lived in this world where demand is really, really high during the summer for air conditioning. It’s much lower during the winter because people use a greater variety of fuels for heating.

Stone: And these events happen during the winter, right?

Blumsack: These events happen during the winter, and so it creates these simultaneous stresses on two systems at once. Back when the preponderance of our electricity generation came from coal, not natural gas, then that kind of simultaneous peak didn’t matter as much to the power grid. Now that the grid is more dependent on natural gas power generation, these cold snaps are stressing the power grid and increasing electricity demand at exactly the same time that they’re stressing the natural gas grid and increasing natural gas demand. So these gas power plants basically find themselves stuck competing for this fuel that’s in very, very high demand.

Stone: It’s important to understand something here I wanted to ask you about, as well. Gas is different from other fossil fuels in that it’s pretty much delivered just in time. There is very little storage relative to how much is coming through the pipeline and being used at that moment. Why is this the case, and what are the implications for reliability of electric generators because of that?

Blumsack: Just to be totally clear, the US actually has a tremendous amount of gas storage. Gas has been treated differently than other fuels for power generation, and this is primarily a contrast with coal. Very few gas power generators have the capability to store fuel on site. If you’ve ever been to a coal power plant, you might see the giant piles of coal that are right next to the power plant. We simply have not had that kind of on-site fuel storage for natural gas.

Stone: So there is storage, but it’s not next to the plants themselves.

Blumsack: Yes, there is plenty of storage basically for the bulk gas delivery system, but there’s very, very little on-site storage at gas plants. Gas plants have largely relied on being able to get what are effectively just-in-time deliveries of fuel. During these cold weather events, when the pipeline system gets stressed, and there’s a lot of competition for that gas, getting those just-in-time deliveries has been more challenging for some gas plants.

Stone: Take a step back. What was the actual cause of the outages we saw in Uri and in Elliott? Was it the gas pipelines having some sort of problem? Was it the generators themselves not being able to operate because of the cold weather? What exactly was the root of the bulk of the disruptions?

Blumsack: I think what you’re getting at is the other really important element of how these kinds of events can really highlight the risks and vulnerabilities in just how interconnected and interdependent the gas and power systems have become. When you have these really cold weather events, there are multiple things that happened in Texas essentially simultaneously. One thing that happened was that you had basically what are called “freeze-offs” at natural gas supply points. What happens is that basically the gas that comes out of the ground is not pure gas. It’s not pure methane. There’s water and other stuff in there, and then it goes to a processing plant. Then what comes out of the processing plant is the gas that goes into the pipeline.

Stone: That’s the pure gas.

Blumsack: That’s the pure gas, yes. In Texas, you had a number of natural gas supply points that essentially were not weatherized. They were still producing gas, but this gas had some water and other stuff in it, and the well heads literally froze. So that delivered a shock to the gas supply system. You had similar problems with cold weather operations at some gas power plants and other kinds of power plants, too, for sure.

So at these power plants, they weren’t designed to be able to operate in this cold weather, and there were equipment malfunctions where things broke, and the plants basically had to shut down — which was also a contributor to the shortage of electricity.

Stone: Those are weatherization issues, right? Those plants could have potentially worked, but my understanding is that the investments weren’t made to make sure that they would work in those extreme conditions.

Blumsack: That’s right. So the operators of those plants didn’t make the investments to be able to operate in those kinds of cold weather conditions.

Stone: So summing up the condition, we’ve already gotten to this. We’ve got these increasingly interdependent gas and electric systems, and the gas pipeline system is also dependent on the reliability of electricity systems to run those pipelines. Is that right?

Blumsack: In some ways it is. We have gotten a lot of very stark reminders of how dependent the power grid can be on the natural gas supply system and the natural gas transmission system, but that dependency goes the other way, too. There are swaths of the nation’s pipeline system that are very dependent on electricity. And so what do they use electricity for? Well, they use electricity to run compressors, to maintain sufficient pressure to push all the gas molecules down the pipeline.

There are certain parts of the nation’s gas pipeline infrastructure that are very, very heavily dependent on electricity to run the compressors. The gas processing plants also rely on electricity to run. So this two-way dependence in situations like Winter Storm Uri in Texas basically creates this feedback loop where you have problems on the natural gas supply system, which leads to gas power plants coming off line, which leads to electricity shortages, which leads to blackouts, which leads to further problems on the natural gas supply system being able to move gas, not just to power plants, but also to homes that heat with natural gas.

And so you have this two-way dependence that in these extreme weather situations can, and as we saw in Texas during Winter Storm Uri, can create really bad situations. People can die, and people did die because there were shortages of fuel and power in their homes.

Stone: So what we’ve got is this very clear scenario, and now we’re getting to the main point of our conversation today. You’ve got this incredible interdependence between the two systems. The electricity system — now 40% of the electricity generated comes from gas-fired generators. If that gas supply is not reliable, then the generators cannot operate. And also if the generators cannot operate, then the gas pipeline system can’t operate, and you’ve got this vicious cycle, whatever you want to call it.

On top of that, as you mentioned a little bit earlier, we’ve got this shift to winter demand peaks for the electricity system, or at least we’re moving in that area. So you’ve got more electricity needed in the winter. At the same time, you need the gas to heat homes and do all that, so there is a double whammy hit on the gas system. The gas system and the gas pipeline system really predate the growing dependence on the electricity before that gas. What was the gas pipeline system and the gas supply system designed to serve in this country, and how has that changed?

Blumsack: For many years, power plants were pretty marginal customers for natural gas companies. We had been using natural gas for power generation for a long time, but again, until the fracking boom, supplying natural gas power plants was not really the main business of basically gas producers or gas transmission companies, the companies that own and operate the pipelines.

So we had this incredibly extensive natural gas pipeline system in the continental US. It’s very well interconnected. It has a lot of redundancies, but it was basically designed for natural gas producers, basically to be able to move gas to gas utilities that would ultimately serve homes and businesses.

Stone: That’s for heating and cooking?

Blumsack: For heating and cooking, all of that kind of stuff, and also to industry. Industry and manufacturing uses a lot of natural gas for heat. They use it as feedstocks for petrochemical processes, so there is a wide variety of different uses for natural gas in industry, manufacturing, and chemicals. So this extensive gas supply system that we have was really designed to serve those kinds of customers and wasn’t really designed to serve gas-fired power plants, the way that it is now.

Stone: So this is all new load, and one of the questions is how does the electricity system incentivize the gas system to serve its additional demand? And that’s a whole can of worms, as I understand.

Blumsack: It is. And I think to start talking about this, I think it’s worth going back to the fact that gas power plants have historically gotten just-in-time gas deliveries, and why that has happened. Why do gas power plants treat their gas supply differently than a gas utility, which might have a long-term, very heavily guaranteed fuel delivery contract? So we’re going to get a little bit nerdy and jargony here, if that’s okay.

Stone: Yes.

Blumsack: All right. If you boil it down, there are basically two types of gas supply and delivery contracts. One of them, we call “firm” contracts, which is essentially the highest priority gas delivery, where if you have a firm delivery contract, then you’re guaranteed to get the contracted gas, if you want to take it, unless there’s some kind of force majeure, kind of really extreme event.

Stone: Something happens with the pipeline.

Blumsack: Something happens with the pipeline, or a gas processing plant, if something really bad happens. So those are the kinds of contracts for gas delivery that gas utilities and large industrial manufacturing chemical customers have historically had. The other kind of gas delivery contract is called “non-firm.” A non-firm contract basically gives you the right to take gas from the pipeline system, if there is gas available. If the capacity of the pipeline is not totally used up with deliveries to other customers, then if you have a non-firm contract, basically you can take the gas that you need.

Stone: And this is what generators are using?

Blumsack: Historically, power plants have used non-firm contracts. There were a couple of reasons for this. One is, as you might expect, a non-firm contract is going to be a lot cheaper to get than a firm delivery contract. So if you’re the owner of a gas plant, like a utility, and you have a gas-fired power plant, and you have this regulator watching over your shoulder to make sure you’re not spending too much of your rate payer’s money, that non-firm contract is going to look a lot more palatable.

So even as we have opened up electricity markets to competition, if you’re the owner of a gas-fired power plant, and you’re competing with other power plants to sell electricity, you want your costs to be low. That’s also going to motivate you to get the non-firm contract.

Stone: You want to make sure you clear into the market.

Blumsack: Yes, so you want to make sure that you clear into the market. You want to make sure that you’re able to sell, otherwise you don’t make any money. Historically, the fact that gas power plants got these non-firm, non-guaranteed, interruptible gas supply contracts was not a big deal because there were just very, very few circumstances where you would have constraints on the gas supply system that would require non-firm contract holders to have their gas curtailed.

It wasn’t really a big deal if you did have curtailments to gas-fired power plants. There were plenty of other coal or oil or other kinds of power plants that could fill in that gap. But as gas has become such a huge part of the power supply system, when those non-firm contracts get interrupted, it’s much more consequential because you don’t necessarily have all of the other supply sources to take its place. This is really the reason that, one, we have basically built the gas supply system out to serve gas utilities and industry, because they were the ones with the really fixed, long-term contracts.

Stone: Now we’re expecting them to service the electricity system, but they can’t enter into these firm contracts that give the financial incentive for the gas industry to do anything. Is that it?

Blumsack: There are some gas power plants that have entered into firm contracts. Starting probably about ten years ago, there was another kind of extreme cold weather event in the US Northeast, and we didn’t have big blackouts at that time, but we almost did.

Stone: Was that the polar vortex?

Blumsack: Yes, that was the 2014 or 2015 polar vortex, and that event happened at the same time that the role of gas in the power grid was really ramping up. That started to wake — especially in the Northeast — that started to open some people’s eyes about these vulnerabilities, so you started to have some gas-fired power plants signing firm contracts, but not all of them did. Ten years ago, there still were not a lot of major incentives for the gas power generators to sign these contracts or find some other way to ensure gas deliveries.

Stone: There’s a whole question I want to get to a little bit later in this conversation. It’s how long will we need gas? Is this the right path for us? That’s a different conversation, but we will touch on it in a moment. But I want to take a step back, and I feel like I meant to ask this question 5 or 7 or 8 minutes ago, and I didn’t, but basically what we’re seeing here is this increasing interdependence, but the two systems are not planned together, they’re not coordinated in an operating sense fully, and they’re not regulated together — yet they’re increasingly interdependent. Could you introduce us to this lack of coordination that we’re seeing between these two industries?

Blumsack: Yes, so basically because the gas pipeline system, our overall gas supply system, grew up in a very different way than our power grid did, there has always been some interconnection, some use of gas for power generation and all of that, but there was never really a need to think about any kind of joint planning for gas and electric transmission. There was never really any need to think about trying to harmonize how the two systems are operated. Both gas and electricity have very different regulatory systems, and so basically on the gas side, FERC is given some jurisdiction for siting and pricing on interstate gas lines. FERC doesn’t have that same jurisdiction over the power grid, so they don’t have the same kind of siting authority.

So FERC, especially recently, with their electric transmission order that came out just a few days ago, has really tried to encourage regional power grid planning, but you really don’t have the same kind of system-level planning on the natural gas side.

Stone: I think it’s also interesting there’s no mandatory reliability standards for the gas pipeline network, as well. Is that correct?

Blumsack: That is.

Stone: Like there is for the NERC via FERC insurance to the reliability of the electricity system, but there isn’t something equivalent for the gas system.

Blumsack: That’s also correct, and it’s another big aspect of the regulatory difference between gas and electricity. In electricity, for many years, we have had NERC, the North America and Electric Reliability Corporation, whose function is to set reliability standards for utilities and power grid operators. Those standards can be enforced by FERC, right? So even though it cuts across these two entities, there is a system for power grid planning with reliability as a huge objective for that, and a way to monitor compliance, to make sure that the grid is reliable.

On the gas side, there’s no NERC for gas. There are regulations for gas pipelines that are mostly focused on safety because we don’t want gas pipelines to blow up. But the reliability of the gas system, the reliable operation of gas pipelines, that’s not something that is regulated in the same way that it is regulated for the power grid. On the gas side, the reliable operation is largely market-based, so those firm gas customers, if there’s a problem on the pipeline, if the pipeline doesn’t operate reliably and if those gas deliveries can’t get made, then there’s some financial recourse that is written into the supply contract, but there’s no overt regulation.

Stone:  So your work really focuses on the planning aspect of this, right? Planning to resolve the issues of gas-electric coordination and the lack thereof. A number of recent initiatives have come from the industry. Recently some of the big markets such as PJM and ISO New England released a paper with suggestions as to how to better coordinate the systems. NARUC, NERC, I think have all come out with reports and position papers in recent years to solve this. What are the key factors and the solutions that you see for this from the planning perspective?

Blumsack: The work that I have done on these gas-electric planning issues really, I think, at its core suggests a couple of things on the planning side that would really help ensure a more resilient power grid, one that is better able to withstand disruptions and interruptions in the gas supply system.

One thing which we think that could be done actually at fairly low cost in many areas would be to require or somehow otherwise incentivize gas generators that either don’t have any on-site gas storage or don’t have what we call “dual fuel capability,” which basically means they could run on natural gas, but they could also run on fuel oil for a short period of time, and you can store the oil on site. That would basically let the power plant ride through the disruption on the natural gas system.

From a system-wide planning perspective, this is one thing that we think could be done at pretty low cost, that would allow the power grid to ride through a large number of potential disruptions on the gas supply system.

Stone: And this would be a requirement for all gas generators, to have this dual fuel capability?

Blumsack: So whether you call it a “requirement” or some kind of a market incentive — because there are going to be different ways of doing things in different regions, because we don’t have a uniform set of planning requirements for the entire power grid — so things are done on a regional basis. Whether that takes the form of a market incentive or whether that would take the form of a regulatory requirement, either way, one possible planning solution is just to be able to store more of this fuel on site. And again, based on what data we have on gas interruptions, you wouldn’t need a ton of gas storage on site at a power plant in order to be able to handle most disruptions. It’s on the order of several hours. That would add a lot of resiliency to the power grid at a pretty low cost. That’s one possible solution. There are grid operators that, in various ways, are actually exploring this kind of solution right now.

Another planning strategy that our work suggests, which is a little more difficult to accomplish right now, because it would require some coordination across both the gas system and the electric system from a planning perspective, would be basically to be able to essentially identify what I would call “vulnerability hot spots.” So the areas of these two systems, that if something were to happen in this area, would be particularly consequential. We think we can figure out where those hot spots might be, and to make both gas and electric transmission planning decisions that can effectively minimize the risk of disruption at these hot spots, right? And those planning decisions may include redundancies in gas connections for gas-fired power plants. I know that there are some gas-fired power plants that have also looked at this solution, kind of thinking about how the gas grid relies on the power grid, thinking about basically having almost dual fuel capabilities for compressors and the gas processing plants. It’s less clear how much of that is going on.

So this kind of planning solution is pretty simple, at least in concept. It would require some coordination across these systems. I’ll say that in the aftermath of Winter Storm Uri, there has been some attempt in Texas to do this kind of planning. Texas is now requiring the identification of critical pieces of gas-electric interconnection. Once you’ve identified those, what you do about it is not totally clear yet, but Texas has taken some steps in this particular direction.

Stone: That’s an interesting point because Texas is its own market, but a lot of the requirements for gas system reliability actually go to the states, right? There’s, again, no national coordination on that.

Blumsack: Yes, that’s right.

Stone: And it’s interesting because FERC Chairman Willie Phillips has called for Congress to name an agency to oversee the natural gas system and its reliability. He did this a couple of years ago following Elliott. Is that something that Congress might do?

Blumsack: I think the story of how we wound up with NERC is fairly instructive here. So it took Congress about 40 years to figure out that they really had to give NERC teeth. The creation story of NERC was that after a large blackout in the 1960s, the utility industry basically feared the wrath of Congress, that Congress was going to come down on them with a hammer and regulate them really stringently for reliability. As a way to avoid that kind of heavy-handed regulation, the utility industry basically created what is now NERC as a voluntary standards body, and everybody in the utility industry said, “We will voluntarily abide by the standards that NERC sets.”

So for really almost 40 years, after that blackout in the 1960s, we actually did not have formally, Congressionally-mandated reliability regulation. It was really another big blackout which happened in August of 2003 that affected a huge part of the Northeast, and all of the various root causes of that blackout — which was a very complex event — that event basically pushed Congress towards really designating NERC as a formal standards body and giving FERC essentially authority to punish.

Stone: [OVERTALK] The Energy Policy Act of 2005.

Blumsack: Yes. So in a way, I think you probably would have to have a similarly disastrous event in order for Congress, especially the way that Congress tends to operate these days, to really create a formal reliability regulatory structure for the gas system. But at the same time, I do think that it is something that needs to happen because the power grid is so dependent on the natural gas supply system, and the way that we monitor and set standards for reliability for this supply system which is so key to the power grid is just years behind what we do for electricity.

Stone: So we’ve gone through the planning aspects, but there are operational challenges here, as well, in real time as it’s the operation of the electricity system and the gas systems that are causing us problems. Can you give us a high level overview of what some of those key issues operationally are?

Blumsack: One of them that I mentioned a little earlier has to do with a kind of contract, the kind of supply contract that’s sort of non-firm, just-in-time delivery contract that gas power plants have. And so there are and have been economic drivers for that, but because the gas plants are basically — if they have these non-firm contracts, they’re always looking for these just-in-time deliveries — then what they’re basically forced into is rather than being able to schedule guaranteed deliveries a long time in advance, they’re always almost going out into the spot market for gas deliveries, right? And this is an area where the way that the gas spot market works and the way that the electricity market works have not historically been very well aligned. It goes back to something that is just really, really fundamental, which is if you’re going to sign a contract or receive a dispatch order from a grid operator or whatever to supply gas or electricity on some day, then what exactly does a “day” mean? And this just sounds totally —

Stone: Times, right?

Blumsack: Yes, right. And this is one of those just weird historical differences that didn’t really make a huge difference for many, many years and now is kind of part of that gas-electric coordination thing that we have to grapple with. If you say that you’re going to schedule electricity deliveries for a day, that means what you and I might think it means. It means that whatever schedule you’re on, it’s going to start at midnight, and it’s going to end at 11:59, right? So the electric day and the calendar day kind of coordinate, right?

The gas system operates on a different definition of a day. The gas day has historically started at 10:00 a.m. Eastern time and has ended at 9:59 a.m. Eastern time on the following calendar day. This is how gas deliveries are scheduled. They’re not scheduled midnight to midnight. They’re scheduled 10 a.m. one calendar day to 10 a.m. the next calendar day.

Stone: So I’m an electric generator. I’m committed for the next day. I’m supposed to buy gas for that day, but the day is not the same day. There are two days I’m straddling here.

Blumsack: Yes, that’s exactly right. So you basically have to go into the gas market for two different gas days, which just makes things much more complicated for the gas generator. It’s also complicated because if you’re the gas generator, and you get dispatched to produce on some schedule for some day, then what you’ve got to do is you’ve got to buy part of the gas day, and then you’ve got to buy part of the following gas day. That certainly makes things more complicated from a market perspective. It also could potentially lead to some financial exposure for the gas plant, because they may wind up with essentially buying gas deliveries that they don’t need.

Stone: And also the weekend, right? The gas market is a 3-day weekend?

Blumsack: Yes, so over the weekend, the electric day just keeps going. There’s an electric day Saturday and an electric day Sunday. The gas day basically straddles not just one day and another day, but one day and the entire weekend. For the gas power plant, that makes things especially complicated because you may have to go out and try to schedule fuel on, say, a Friday, not really knowing what your dispatch instructions are going to be on Sunday. So that’s complicated, and it’s a risk.

Stone: It’s a financial risk.

Blumsack: Yes, it’s a financial risk for the gas power plants because they’re faced with this choice where either they may have to wind up kind of over-buying or potentially under-buying gas for some electric day. In the gas market, you have basically the daily scheduled market for the gas day. And then within a gas day, there are a series of what are called “intraday markets,” which basically allow gas buyers and gas sellers to adjust how much they want to get delivered in the middle of a gas day. And those intraday markets are meant to be an adjustment mechanism.

If a gas utility finds that it unexpectedly needs a little bit more gas, it goes out into the intraday market. So these markets, these intraday markets, historically are not terribly liquid, which just means there aren’t the same large pool of buyers and sellers as there would be for the daily market. And so they’re not very liquid, and they can be very volatile. During Winter Storm Uri or during Elliott or even during the polar vortex ten years ago, you’ve heard these stories of astronomical gas prices. What you are really seeing there was the volatility in the intraday market.

Stone: Due to the illiquidity of that market.

Blumsack: Due to the illiquidity. You had gas power plants that were scrambling, trying to get gas on this intraday market, and maybe not just power plants, maybe other buyers, too, but not as many sellers. You can envision ten hungry dogs going after five delicious bones. [LAUGHTER]

Stone: There’s kind of an elephant in the room here that I want to address, and I mentioned it earlier in this conversation. Where is gas demand going? What is the role going to be for gas in the future? And before I ask you your specific thoughts on that, I just want to point out the fact that gas is important, not only because it is a bigger part of our generation mix, but as we go through the energy transition, gas and highly flexible combustion turbines are very helpful in balancing the ups and downs of wind and solar output.

So basically what we have is a situation here where we’ve got a lot more gas that’s needed. We don’t know how much we’re going to need 10 or 15 years down the line, because nobody knows where gas is going to go, but we know that gas demand might also be increasingly volatile if a bigger portion of that gas demand is as a balancing resource that you cannot really predict on any given day how much of it you’re going to need, how much are you going to run your gas? How much wind and solar is going to be output, and how much are you going to have to balance to that?

That further complicates, I would imagine, this need to coordinate these two industries and to have the gas system invest in infrastructure to support the electric system if this demand is just so unpredictable.

Blumsack: Yes, and I think there are a couple of things that you are hitting on that are really important to this future story. One is that in very, very few cases have we seen major pipeline investments to support gas-fired power plants. Building a gas pipeline involves just huge risk exposure on both sides because if you’re going to build the pipeline, there’s so much capital that’s invested in it, that you had better be sure that there’s a buyer.

On the buyer’s side, gas is a wonderful fuel in a lot of ways, but there’s really only one good way to move it around, which is in a pipeline. So on the buyer’s side, you’re dependent on that pipeline if you need the gas. What has really underwritten almost all of the major pipeline investments in this country are basically long-term contracts between buyers and sellers. For the pipeline company, it ensures that they are going to get paid for the pipeline, and for the buyer, it ensures that there is going to be a pipeline there to deliver the gas.

There really have not been a lot of major pipeline investments to support gas-fired power plants because gas-fired power plants have not really been in a position to sign these very, very long-term supply contracts because even without this balancing role that gas plants play, the output of gas plants over a day is still going to be pretty variable. Many gas plants don’t operate like nuclear plants, where you set them at basically one output level, and they stay there for a long time. This is the beauty of the flexibility of the gas turbine is that you can do this. You can ramp it up or ramp it down to match demand or renewables or whatever, but that doesn’t make for the kind of long-term contract that is going to get pipe in the ground.

So I think looking forward, if we are going to use gas for more of this kind of balancing function, then we’re going to have to figure out a way to pay for the infrastructure. The other element, which I think has the potential to influence these questions of how we pay for this gas infrastructure is electrification. And if we’re successful at pulling a lot of gas stoves and gas furnaces and other things out of buildings and replacing it with electricity, if the electricity demand growth that would accompany major investments in data centers as PJM, the mid-Atlantic grid operator, is starting to see and expect.

Stone: Yes, you’ve got a lot of new data centers in Virginia and the D.C. area, right?

Blumsack: Yes, in Virginia and the D.C. area, which is part of the regional PJM power grid. When you talk about electrification, there’s one. There’s this sort of shifting of demand basically from the burner tip to the power grid. And then there’s this new demand from these new, large sources of electricity. And just to use PJM as an example, they are expecting to need substantial amounts of new capacity in order to meet this rising electric demand, a lot of which — at least over the next ten years — is likely, I would say, to come from gas.

But again, looking 20 years into the future, as energy storage costs continue to fall, and as renewables costs continue to fall, as more states may adopt renewable portfolio standards or other policy measures to encourage renewables, what does that look like 20 years from now?

Stone: But that’s the fundamental kind of final question I wanted to ask you about here, right? Should these investments be made in gas infrastructure to support the electricity industry? Obviously it sounds like there’s a need, but there are technological changes to storage. There’s battery storage that’s coming along. Given climate change, given the priorities that we have, where does gas fall in the next 10 or 20 years?

Blumsack: Yes, I think there’s a lot of risk there from multiple angles. I do think that if you look over the next 10 years, I think it’s pretty likely that gas is going to continue to play a pretty big role.

Stone: But who is going to invest in that infrastructure? [OVERTALK]

Blumsack: Who is going to invest in all the infrastructure, right? A 10-year underwrite of a pipeline, ordinarily — my sense is that’s not going to cut it.

Stone: It gets stranded pretty quickly.

Blumsack: Yes, something’s going to get stranded there. So if you think about 20 years out, and crystal balls are always really dangerous, but with the falling cost of energy storage and other potential technologies that could also provide this balancing service, then what that means is a lot of those gas plants are going to face a lot of competition, which is a good thing. But because gas is so dependent on this very easily-stranded infrastructure, it does create a lot of risk.

I know that how you handle this risk is an issue that a lot of state utility commissions are grappling with right now, as their utilities look towards the future and make decisions about what to build.

Stone: Seth, thank you very much for talking.

Blumsack: Thank you so much. I enjoyed the conversation.

Stone: Today’s guest has been Seth Blumsack, Director of the Center for Energy, Law and Policy at Penn State University.