Why the IRA’s Carbon Capture Tax Credit Could Increase Greenhouse Emissions

New research raises doubt around the climate benefits of the 45Q tax credit for carbon capture and storage for fossil fuel powerplants.

The Inflation Reduction Act earmarks billions of dollars of incentives for carbon capture and storage from coal and gas-fired powerplants. Ideally, the incentive will provide a path for fossil generators to reduce their greenhouse gas emissions as the electric grid transitions to cleaner resources and to net zero.

Yet recent research calls into question the climate impact of the IRA’s carbon capture tax credit, known as 45Q. The report, co-authored by a former deputy assistant secretary for the Department of Energy’s Office of Carbon Management, finds that 45Q could lead to an increase in greenhouse gas emissions by incentivizing coal and gas generators to extend their working lives and maximize their output. The result could be billions of dollars of taxpayer money spent with no climate benefit.

Emily Grubert, report co-author and now an associate professor of sustainable energy policy at the Keough School of Global Affairs at the University of Notre Dame, examines the costs and climate impacts of carbon capture and storage under the IRA. Grubert explains how the 45Q tax credit could lead to unintended climate impacts. She also discusses the need for robust review of proposed carbon capture projects, and strong regulatory guardrails, if 45Q and CCS are to deliver climate benefits.

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 Inflation Reduction Act earmarks billions of dollars of incentives for carbon capture and storage, including for carbon capture from coal and natural gas-fired power plants. The incentive could provide these generators with a path to reduce their climate emissions, as the electric grid transitions to cleaner resources and to net zero. Yet recent research calls into question the climate benefit of the IRA’s carbon capture tax credit, known as 45Q. In fact, the research which we will discuss in detail suggests that 45Q could lead to an increase in greenhouse gas emissions by encouraging coal and gas generators to maximize their output and lifespans. The result could be billions of dollars of taxpayer money spent, with no climate benefit.

Today’s guest is co-author of the report on the costs and climate impacts of carbon capture and storage under the IRA. Emily Grubert is an Associate Professor of Sustainable Energy Policy at the Keough School of Global Affairs at the University of Notre Dame. She’ll explain how the 45Q tax credit could lead to unintended climate impacts. She’ll also discuss the need for robust review of proposed carbon capture projects and strong regulatory guardrails if 45Q, and by extension carbon capture and storage, are to deliver climate benefits.

Emily, welcome to the podcast.

Emily Grubert: Thank you so much for having me.

Stone: Earlier this year, you and co-author Frances Sawyer of Pleiades Strategy published a paper that looks at the cost in climate impact of carbon capture and storage under incentives provided by the IRA. Before we discuss the report’s findings, could you start us out by explaining how 45Q works and the range of carbon reduction activities that it covers?

Grubert: Yes, absolutely. There are a couple of different things that 45Q specifically addresses. It’s a carbon oxide sequestration  credit, which in this case means that it provides incentives for storing or utilizing either carbon monoxide or carbon dioxide. In most cases, what we have seen, this tax credit has actually been around for quite a while, and the IRA just increased its value, rather than introducing it for the first time. What we have seen generally is that people have used this for CO2, not carbon monoxide. But effectively what it does is that it pays projects’ proponents a specific amount of money, depending on how exactly they’re managing their CO or CO2 to store or utilize one of these carbon oxides. And so it’s effectively a production tax credit for stored carbon oxides, specifically CO2.

Stone: Okay, so 45Q, as you’ve said, has been around a while. It was introduced in 2008. The amount that it pays out for carbon capture has been increased under the IRA. Why the increase, and what does that increase say about the current view on the role of CCS?

Grubert: Yes, it’s kind of interesting, especially in the context of the Inflation Reduction Act being named as it is. A lot of the increase was really catching up with inflation relative to the 2008 credit. It was $50 a ton for the most common thing that people tend to think about in the climate space, which is essentially permanent dedicated geologic storage of CO2. So it was 50. Now it’s $85, assuming that you’re meeting some labor and other requirements associated with how you manage it. But it’s actually not that big of an increase relative to what you might expect, just because it wasn’t inflation indexed before.

So it is bigger, and that is potentially meaningful, but it’s not a huge amount bigger than it used to be. I think the expectation is basically that it’s large enough to encourage more projects to go forward, also because it has caught up with inflation. I think one of the more meaningful things that changed actually under the Inflation Reduction Act around 45Q is that the threshold for total emissions that you’re required to be managing is actually a lot smaller. So the eligibility has widened, in addition to the tax credit increasing a little bit. I don’t mean to say that the increase isn’t meaningful at all. It is, but it’s not quite as significant of an increase as it looks like because of some of that inflation stuff.

Stone: You mentioned a moment ago, and you also stated in your research, and I’m quoting from that paper here. You say that 45Q is “effectively a carbon dioxide production tax credit.” And in this, herein lies, it seems, what may be a fundamental — I don’t know if “flaw” is the correct word to use here — but with the incentive, that could lead it to increase greenhouse gas emissions, rather than reduce them. I wonder if you could explain how that might work?

Grubert: Yes, I think this is where some of the history of 45Q is actually important. As you said, it was initially passed in 2008, and the stated intent or the description of the stated intent at the time was to try to create a capability within the United States for geologic storage of CO2 or utilization of CO2. So it was really designed to try to encourage this stuff to happen in the first place, and as a result, it really incentivizes having a lot of CO2 to store or utilize — or carbon monoxide. But again, most people use CO2, so it was really developed as a way to kind of kick-start a storage industry.

When we’ve started now more recently to use it in the context of climate emissions reductions, the fact that it’s not really fit for the purpose of emissions reductions — again, it’s incentivizing maximized storage, as opposed to minimizing emissions — is starting to show up. The specific way, as you allude to, that that happens is basically that the more CO2 you have, the more CO2 you’re able to store, and the more tax credit you’re able to claim. That in and of itself doesn’t necessarily increase emissions. CCS is rarely complete, so there are some additional emissions that are still going to make it to the atmosphere, even with the CO2 capture system. But the bigger issue from our perspective, especially in the power sector, is essentially that because that tax credit is only available for about twelve years, it does not necessarily incentivize long-term mitigation; but it does incentivize really significant capital investments in plants that may extend their lifetime in addition to encouraging them to run more while the capture unit is actually still being paid for by the subsidy.

Stone: One could argue here that even if fossil plants are creating more emissions than some of them are obviously capturing, that is the main point, right? They’re providing needed electricity and doing so with fewer emissions. That’s really the face value justification for 45Q, right?

Grubert: At this point in the power sector, yes. I think there is still an element of it that is really trying to develop the industry overall. There are debates about what that exactly looks like. CCS is not just a power sector thing, and I would actually probably argue that in the US context, at least, CCS is not that valuable in the power sector but could have some pretty important implications in other sectors where there are fewer options for mitigation. But yes, the argument is essentially that by encouraging capture, you’re reducing emissions.

I think the thing that becomes challenging in the power sector is that because we have other ways to generate electricity, you end up blocking replacements, as well. I don’t want to be cavalier about how difficult it is to build out lots and lots of new renewable energy and zero-carbon resources, but in a situation where you’re encouraging a lot of large fossil-fired power plants, many of which are very old in this country, to basically put billion-dollar investments onto their plants and then run as much as they can — you are creating challenges for actually having other things stepping in. Just having that massive additional fossil energy, which again is not carbon-neutral. It’s actually not consistent with the executive order that asks for carbon pollution-free power by 2035, just because you’re not capturing all of those emissions. You end up with some kind of interesting maladaptations that probably are not driving toward full decarbonization of the power sector.

Stone: I want to get into the opportunity costs of 45Q, versus some other carbon mitigation strategies in just a moment, but I want to go with something you just said, as well. You talked about the capital investment that would be made onto coal plants, onto natural gas plants to capture the carbon. That’s interesting because when you’re making that capital expenditure, you assume that you’re going to want to use that for as long as possible to get your money back, right? And this is a critical point. You point out in the research that 45Q leads to greenhouse gas reductions where plant life is not extended. But where it is extended, that’s when you run into the problems.

Grubert: Yes, exactly. And some research that I had done a couple of years ago essentially shows that most of the US’s fossil fleet, about 75% at the time — and that has changed a little bit with some retirements and some additions — but about 75% of the US fossil-fired fleet would reach end of life by 2035, regardless. And so the argument for life extension with these ordered billion-dollar investments is pretty strong, I think. And we have seen a number of project proponents talking about the fact that adding the capturing unit really gives people the opportunity to go and do other overhauls, things like this that just put a lot of capital into these plants, which is likely to extend their lives by a couple of decades.

So specifically Boundary Dam, which is a Canadian plant with CCS, explicitly talks about the fact that adding the CCS is a way to extend the plant’s lifetime. In that case, I think 30 years is their stated target. It might be 20, but basically this is not an unlikely outcome, and it’s something that people are pretty upfront about. I think what the challenge really is in the 45Q context is you can do those plant life extensions without necessarily always having to operate the capture unit. So the likelihood of turning off the capture unit, and then allowing the plant to keep burning afterwards is really where the big risks come from. We have seen some project proponents talking about potentially taking that route, as well.

Stone: So those carbon capture units would be shut down because they’re expensive to operate, and after the twelve years of the tax credit, after the tax credit expires, it simply wouldn’t make economic sense for a power plant to continue to operate its carbon capture equipment, right?

Grubert: Exactly. This is one of the big differences between the kinds of tax credits that we see for carbon capture versus for something like solar or wind. With a solar or wind farm, operating those plants is pretty cheap, and so basically you use the tax credit to get the thing built. But once it is built, you need to maybe do some light maintenance and mow the lawn occasionally — that sort of thing. But it’s not like you’re paying for fuel anymore once you’ve earned the tax credit. If it’s built, it can just run. With carbon capture, it’s an incredibly energy-intensive process, and because of that and because of some other consumable materials that you end up with in those situations, you do actually end up with this, as you say, pretty high operating cost. So if you’re not required to do that, and nobody is paying you to keep doing it, there’s no real benefit to continuing the capture. But again, if you’ve overhauled your whole plant fairly recently, there’s not a lot preventing you from just turning off the capture unit and continuing to run your now updated plant.

Stone: So if we’re looking at the energy transition and the need for natural gas plants to continue to operate for some time, as dispatchable resources that can balance the output from intermittent wind and solar plants, it would seem that 45Q could play an important role in cleaning up the carbon emissions from gas generators that, again, can play an important role in balancing the output of renewables. Yet your research points out, quite ironically, that the economics are such that it doesn’t make sense for gas plants to make use of 45Q. Could you explain?

Grubert: Yes, I think there are a couple of dynamics going on. First of all, with natural gas plants, the subsidy is worth a lot less because there is a lot less CO2 in the flue gas. So not only is it actually harder to do the capture because you’re separating a more dilute gas, but because the credit is based on tons of CO2 and not by degree of mitigation, it actually is a lot more costly, relative to what the credit is.

So by our calculations, just based on how much CO2 you tend to see in a coal plant versus a gas plant, the tax credit is worth about 9 cents a kilowatt hour for a coal plant and about 3 cents for a natural gas plant. This is all compared to about 2.6 cents per kilowatt hour that’s usually offered to wind farms at this point, and to zero-carbon electricity going forward. So yes, the tax credit is much more significant for coal, and also just the costs are quite a bit lower relative to how much CO2 that you’re able to get because it’s an easier thing.

I think with regard to just what we should be trying to incentivize, though, there are a few things that are a little bit challenging about natural gas plants with regard to CCS. One is that CCS does nothing for methane emissions, and methane emissions probably actually go up per unit of electricity when you add a CCS unit, just because they are pretty energy-intensive, and you’re not doing anything about the methane. So upstream methane emissions can be pretty significant, even if you have a capturing unit for the CO2 at the plant itself.

The other thing in regard to the argument that people do make about having flexible, up-and-down dispatchable resources available during the transition in the form of natural gas is that adding a CCS unit actually really pretty severely limits the flexibility that you have with one of those plants. Because of the way that CCS works, you’re effectively turning a plant that’s pretty flexible because it has a gas turbine, essentially. So think about a stove, where you can turn the flame on and off. That’s basically how gas turbines work. Steam plants, by contrast, like a coal plant or a steam natural gas plant operate more like a tea kettle, so you turn on a stove, and you have to wait a while for the water to heat up. So they’re much less flexible. Adding a CCS unit to a gas plant makes its characteristics look a lot more like a big steam unit. So they’re not as easily able to turn on and off quickly, those kinds of things.

The triple problem here is that yes, the tax credit is not as significant for gas, and costs are higher. You end up with quite a bit of methane burden because of the energy intensity of CCS. And adding CCS actually probably reduces some of the characteristics that are the best argument for keeping gas in the system for a little longer. So I think in general I tend to be on the side of thinking about what other flexible resources we can look to in this transition, because gas is going to struggle, especially in a situation where we’re trying to decarbonize quickly.

Stone: There’s a really interesting detail you have in the report here that says generally the “weighted average capacity factor,” meaning the percentage of time that a natural gas plant will actually be up and operating, right now is around 33%. But you say that these facilities would actually need to run at 70% of their capacity for the full twelve years to break even, and that’s versus about four years break-even for a coal plant.

Grubert: Yes, exactly, which is again to this point, where you actually lose quite a bit of flexibility, not just because of the steam plant thing, but because the plant needs to be running more often to try to maximize the amount of tax credit it’s getting from CO2 storage, which also means it’s not turning off as often. So if you’ve got a plant that needs to stay on 70% of the time, and all of a sudden you have a big wind day or a big solar day, and you want those gas plants to turn off to accommodate it, that’s not how this is incentivized at the moment.

Stone: Let’s look at the costs and the opportunity costs. What is the cost of reducing carbon via carbon capture and storage, as it applies to a fossil fuel plant versus an alternative such as investing in renewables?

Grubert: Yes, this is partially why we wanted to do this study, because the answer you come to there is a little different if you think about it from a system optimization perspective, relative to a profit-maximizing perspective. We really wanted to look at this from the decision location of a plant owner, thinking about what to do. At the very lowest, the societal costs of running a CCS unit is basically the cost of the tax credit, plus whatever the CapEx looks like, so at the very least $85 a ton, if that’s what’s going to be being paid out. What we see is that because of CapEx and because a lot of the time you are actually adding some emissions because capturing CO2 also includes the new CO2 that you’re generating because you have this extra energy input to run the capture unit. I think those kinds of things, then also getting into some methane, it tends to be quite a bit higher than that.

So we see in situations where we assume that abatement is actually happening, so you’re not on a life cycle basis actually increasing emissions, that you are cutting them. We tend in this analysis to see those costs looking to be about $85 at the lowest, up to a couple of hundred dollars a ton, in some cases maybe even several hundred dollars a ton. In general what’s interesting is that CCS is always going to add costs to running a fossil-fired power plant. This is not something that’s going to increase revenue for you, basically. And to the extent that you do have revenue for that CO2, that’s public funding at this point. But even without CCS, a lot of the work that has looked at relative costs has concluded that a brand new wind or solar plant is actually cheaper, from the ground up than continuing to operate existing coal plants.

That’s not quite as true for gas at this point, but renewables are already significantly cheaper, even than unabated coal plants, and that disparity probably pretty significantly widens once you start to add CCS. So this is an expensive option. That doesn’t mean that there is never any reason to do it. I think in a lot of other countries, the argument is a bit stronger, particularly where you have more, newer plants, potentially more efficient plants that may be more CCS-ready than ours. But in the US, a lot of the conversation is basically, “Should we put a billion-dollar unit onto a 50-year-old coal plant, given that coal plants tend to last about 50 years?” The kinds of costs we’re talking about for doing something like that are substantially higher than the alternatives, but from a utilities perspective, that might make sense because if you want to keep you asset around, and you can make that happen profitably, there may be a financial argument to do that, even if the societal cost is quite high.

Stone: That’s an interesting point that you just brought up, particularly within the context of COP 28, which is just about to get started here, right? One of the discussions right now is how much money should be transferred to developing countries that have fossil resources so that they can apply, for example, carbon capture and storage technologies to those resources so they can take advantage of them themselves. You’re saying that the calculus is a bit different.

Grubert: Yes, I think so. And I think in the particular context of should CCS be an excuse to develop a lot more resources that aren’t developed yet? I’m a lot less enthusiastic about that one. I think that there are some reasonable arguments for applying carbon capture and storage in places that have a lot of relatively young CCS plants, have pretty high energy burdens and energy poverty, and don’t have a lot of domestic resources in addition to what they already have.

So if you’ve already got a bunch of coal plants, and your wind resources are terrible, then I think that that argument is a lot stronger. But within the United States, I think there’s not a particularly strong argument for power sector CCS.

Stone: There’s also the issue here that there are very few working examples of power sector CCS. There’s the Petra Nova plant in Texas, which was actually operating for a few years, closed in 2020. It has recently been re-opened, the carbon capture component of it, but really this is still a very new technology. It’s risky in its own right.

Grubert: Yes, it’s an interesting thing because we’ve tried it a number of times. I think depending on who you talk to, it has been the case that the projects have failed because of financial conditions or other things. But yes, it is technologically complicated. I tend to fall on the side of, “We know it’s possible, but we also know that it’s pretty expensive, and it’s not trivial to do, particularly in a retrofit context.” I think this is also one of the places worth thinking about how we invest in these technologies to try to understand where we could get benefits gets really tricky because there are places where CCS is the only option that we have for emissions mitigation.

We’re working with new chemistries and things like that, but cement plants tend to stand out to me as one where there is not a ton of other options because a lot of the CO2 that you have is just a by-product of the process itself. You’re driving carbonates into cement basically, plus CO2, and it’s not really a fuel substitution issue. So thinking about where CCS has a role to play because there’s not a ton of other options, I think really merits a lot of investigation. But I tend to see people making the argument that investing in power sector CCS demonstrations and things like that in the US makes sense because they allow us to learn more about how we might translate it to other industries or other countries.

I think the point that stands out to me about that is that with power sector stuff, we’re almost exclusively talking about retrofits. So essentially, like to your point about project risk, these are basically all first-of-a-kind projects on existing, old plants that we’re not totally sure about their status. Maybe some of them were thinking about retiring, and it’s not necessarily the case that we’re going to learn a lot that’s translatable from doing those kinds of projects. I think particularly in the context of using what we learn in those contexts to inform international work, we use pretty different technology for our coal plants that are still operating than a lot of places.

So learning how to do a CCS retrofit on a 50-year-old sub-bituminous, pulverized coal plant, operating at subcritical temperatures in the US is not necessarily going to do a lot to teach you how to do it on an ultra, super-critical bituminous coal plant in China that was built in the last ten years. So I think there’s some sense that doing any CCS is going to help us with learning curves, and I think that there’s a real need to be very targeted in what exactly we’re trying to learn and how that gets translated.

Stone: It’s very interesting, the point you’re bringing up. You used to work at the Department of Energy in an important role related to what we’re talking about today. And taking that a little step further, in September of this year, you published an opinion article in Utility Dive that harshly criticized the DOE for what you call “its error-ridden analysis of a proposed North Dakota coal plant CCS project.” And this gets to the issue of really needing strong guardrails and analysis around proposed projects. You state that DOE’s analysis overstated the project’s potential climate benefits and understated its costs. What happened in this case?

Grubert: Yes, it’s still a little unclear. I’m hoping that we do find out more, but essentially, at least what I was looking at was the life cycle emissions evaluation that they conducted as part of an environmental assessment. Essentially there were a lot of pretty significant math errors and also some assumption errors and these kinds of things. But they somehow did not really use best practice in calculating how the emissions might change.

There are a couple of examples that are more on the side of people just weren’t checking this all that closely, so pretty egregious errors in just thinking about what global warming potentials look like, and concluding that SSX, for example, was going to be a major, major contributor to the overall emissions associated with the plant. But in general, it just wasn’t a careful analysis and is something that I think should have been caught at a couple of levels. What’s challenging is that there are a bunch of additional nuances with how we do life cycle assessment, particularly related to things like how much longer would the plant have continued to run? And are we actually changing the baseline scenarios by installing units on these plants?

These are difficult things and involve a lot of judgment calls and a lot of evaluation. The analysis that came out on this particular plant didn’t even really get to that level because it was just not particularly well done. What was a little complicated about that from my perspective is that we’re starting to see quite a bit more policy analysis and policy requirements to use life cycle evaluations of what emissions might look like, particularly with hydrogen. A lot of the way that some of those tax credits are going to be given out really depends on very high-quality life cycle analysis that even people who are really experts in the field disagree about how to address. So when you have these statutory requirements to do life cycle analysis that could lead to — depending on which analysis you look at — maybe trillions of dollars of tax credits going out. But the evidence that we have so far suggests that there’s not a whole lot of existing attention to how to do these really well. That worries me quite a bit.

I think the other thing with the Tundra LCA that was challenging was that they effectively — again, I think there were some really significant flaws in the way the analysis was done, so the numbers aren’t exactly right, but they eventually concluded essentially that the plant was going to lead to large increases in greenhouse gas emissions and recommended moving ahead anyway, even though the entire point of a CCS unit is to reduce emissions. And so that also, the fact that the LCA was poorly done, but also that based on the conclusions, they essentially recommended moving forward, even though the analysis suggested that the plant was not going to achieve its goals. Both of those things were fairly concerning to me.

Stone: Now is there some driver of this? Is there some reason that the DOE would move forward with something that doesn’t look like it’s going to deliver what it’s supposed to deliver? It sounds like it’s almost hell-bent on seeing CCS happen.

Grubert: I can’t comment on their motivations particularly. I will say that with a lot of the grants and tax credits that exist around CCS at the moment, it’s not really about emissions reductions. So congressionally speaking, under the Bipartisan Infrastructure Law, also known as the Infrastructure Investment and Jobs Act, DOE is required to build two demonstration CCS units at coal plants. So that is pretty independent of what the actual emissions impact might be. This is also true for needing to build a couple on gas plants, and needing to build a couple on industrial facilities that aren’t purposed for power generation.

So at some level, it’s basically that Congress required them to build a couple of these. And then similarly, with 45Q, which is a Department of Treasury program, not a DOE one, although they do talk to each other, obviously. There is no requirement that those tax credits are associated with climate benefits. It’s really just about counting how much CO2 you’re storing. And so in both of the really big programs that are resulting in activity here, there is not a link to a climate requirement, and I think that’s part of the issue.

Stone: So there’s no base-lining for these plants, right? There is no base-lining that says you get the tax credit if you go below a certain pre-set carbon output. Basically you get the tax credit for sequestering carbon, but again, there’s no climate goal specifically associated with that.

Grubert: Not a required one. And I think that in a couple of cases, there are efforts within grant programs and other things to try to put some of those guardrails on, but what’s actually statutory, no. There’s not really any requirement to reduce climate impact.

Stone: It’s interesting. Another point that you bring out in the paper is that CCS, even given the fact that it may be high-cost, with questionable climate benefit, it may still be attractive to traditional vertically-integrated electric utilities in parts of the country where competitive electricity markets have not really been introduced. And I’m talking, I guess, most prominently in parts of the Southeast and the Western US. In these areas, high-cost solutions can still be very profitable. I wonder if you can explain that?

Grubert: Yes, absolutely. And I think this is one of the places where we wanted to do this analysis really, really conscious of who is actually making the decision, and who is getting the incentive, as opposed to looking at a broader cost optimization. There has been some really good work out there thinking about cost optimization for the whole system under the Inflation Reduction Act — where does that take you? What we wanted to say was for the plant owners themselves who are trying to figure out how to maximize their profits or continue to operate whatever, how are they likely to behave under these incentive structures?

One of the things that I think is kind of interesting is that utilities in the United States disproportionately own fossil generation, so they own most of the fossil generation, but also they don’t own most of the renewable generation. So the kinds of skill sets and asset management strategies and things like that that they tend to have are much more biased toward fossil than you might expect from just looking at basic costs of new-build kinds of facilities, or even the full distribution of capacity that we have in the country. And so there is already kind of a starting bias to say, “Well, can we continue to operate the things that we have and be profitable that way?” That’s probably more attractive than tearing it all down, having to pay for that, and then also building something that you’re not that familiar with, and you don’t have the right workforce for, et cetera.

The other thing that’s interesting about this from a profit management structure and from just a governance structure in the way that we handle electricity capacity in the US is that utilities are typically able to earn a rate of return on capital investment. So the more they spend on capital, the more they’re actually able to pull back from ratepayers and use with a profit margin that’s generally agreed between them and their public utility commission or something like that. And so in a situation like CCS, where you have an opportunity to put a massive amount of capital expenditure into plants that you already have, that in many cases were already fully amortized and are maybe reaching end of life, there’s a really sort of hidden incentive to invest more in these types of facilities than you might expect from a pure cost perspective, because again, that capital investment is really where they’re going to be making their profits, and they’re a little bit immune to it at some level because again, more capital investment is actually better for them in many cases.

Just the fact that a plant is expensive is not really a barrier in the way that you tend to see with some other kinds of resources. And then in addition to being able to pull a rate of return on their capital investment, because of the tax credit, you’re also getting paid for the operational part of it, at least for twelve years or so. There’s not, again, as much of a barrier as you might expect to doing some very, very expensive things. And because, again, utilities disproportionately own fossil generation, this opportunity for big capital investment with operational costs covered for a period is really attractive from a profit-maximizing perspective, even though it’s much more expensive than alternatives from a system-wide perspective.

Stone: One of the most dramatic numbers that you put in the paper is — I guess this is a worst-case scenario, I think we’d call it — 3.5 trillion dollars invested and almost a doubling of carbon emissions as a result.

Grubert: Yes, and that number is actually only for the 45Q tax credits. That’s in addition to probably about another, I think 1.2 trillion dollars capital investments that would fall on the backs of ratepayers, as well. And it’s the worst-case scenario that we’ve looked at. If anyone wants to play with the models, all this stuff is adjustable so you can try to make your own scenarios, too. But this basically assumes the plants increase their capacity factor while they’re able to access the tax credit, and then also extend their lifespan by about 20 years, which is again consistent with Boundary Dam and a couple of other plants that we’ve seen.

So that scenario, yes, is a huge amount of money that actually results in pretty significant increases to life cycle emissions relative to a count of actual, where the plants would have closed by the time they reached end of life.

Stone: A few minutes ago, you mentioned that 45Q doesn’t necessarily accomplish the overall goal of propelling us towards net zero. And that’s really critical, given the current administration’s 2035 net zero grid and 2050 net zero economy goals. And there’s kind of a technical issue in here, but it explains why 45Q could also potentially slow the build-out of renewables. You talk about negative pricing and how 45Q could enable negative pricing that would allow plants with carbon capture to continue operating at the expense of renewables. Could you explain that?

Grubert: Yes, it really is something that we haven’t observed yet, obviously, because there aren’t really any carbon capture plants that are operating in the US at any significant scale. But hypothetically, we’ve seen this a little bit with other production tax credit-style implementations, so particularly wind has caused a situation where basically you’re getting paid a lot more than it costs you to run at the margin because of this production tax credit. And so you’re actually willing to accept negative power prices because the tax credit is worth more than the negative power price you see.

Stone: And to be clear, that’s the production tax credit for wind power we’re talking about, right?

Grubert: Yes, exactly. And so this being kind of a similar situation, where you have a production tax credit, but based on the amount of CO2 that you’re able to store — which also means it’s related to the amount of CO2 you’re able to generate — has a pretty similar dynamic. And particularly because the numbers are so big, so again, this doesn’t fully account for all of the costs that are going into this, but the tax credit for coal is worth about 9 cents a kilowatt hour, and for gas is worth about 3 cents a kilowatt hour. When you think about marginal situations and also think about the fact that again, these are big steam units that are difficult to turn on and off, you are likely to be able to weather some situations where you might even be able to take negative pricing to keep your plant on, and also to keep generating that CO2 to store underground.

So the fact that there is such an incentive for these plants to operate at high capacity factor, I think even in a situation where you limit the problem of extending the lifespans or turning off the capturing unit at some point — those kinds of things — the fact that you’re incentivizing these plants to basically become base loaders, and base loaders that are willing again to take negative price with both because of the tax credit and because it’s a huge challenge to turn them on and off. Yes, you start to run into some pretty significant issues, and particularly when that’s knocking renewables out of the money for the hours where they’re really making most of their money at certain times of year. This could be, I think, a bigger deal than I had initially anticipated. It depends on the idea that you end up with a lot of CCS build-out. That may or may not happen, but the incentive is certainly there.

Stone: Going back to that 3.5 trillion dollars of money that we talked about that would be spent for no climate benefits. It sounds pretty dramatic. Some have criticized those numbers as out of scale, criticized your research. What’s your response to that?

Grubert: Yes, and I think there are a couple of different aspects of this. One of the things, just to be really clear about it, and especially if people do want to go play with this model, is we assume that all incentivized CCS retrofits actually work. So if there’s a profit margin that is available to be had, then a plant will choose to put a CCS retrofit on, and that will be fully successful. It won’t be delayed. No cost overruns, those kinds of things. I think that one of the places where that tends to lead people to be like, “Wow, you’re really overestimating what this could be,” is that in general we don’t actually assume that every plant that’s capable of installing CCS is actually going to do that, partially because, as you said, we’ve had some pretty challenging times with the technology. It’s unlikely that all plants are actually capable of putting a CCS retrofit on, just because of space constraints and things like that. So the number is really big because it assumes basically that the incentives presented by the tax credit are able to be picked up by everybody. Personally, I actually don’t think that we’re going to see essentially universal uptake of CCS by coal-fired power plants, but that is the incentive. And we’re really trying to highlight what the incentive structure looks like.

The other way that that number gets so big is mostly because we assume a lot of uptake, but the way that the cost per ton gets so big is because we assume that there’s a lot of plant life extension and that it is legal for plants to turn off the CCS units once they are no longer receiving money to keep them operating. And again, because of this large OpEx associated with capture, there’s not a ton of incentive to run, unless you’re either required to or somebody is paying you to.

We have actually seen project proponents talk about this publicly, so I think related to at least one project, people saying, “It’s not in the best interest of our ratepayers to continue to run the capturing unit after those twelve years are up.” So we do expect that this behavior is possible, but I think the best counter-argument to this bad-case scenario happening is that that’s a relatively easy loophole to close regulatorily. You could end up in a situation where you require that plants cannot turn off that capital CCS unit once it has actually been installed. There’s a little bit of effort to do that under the EPA 111 rules. They’re not law yet, though, and the IRA 45Q credit is, I think, the other challenge with the proposed EPA rules that would close the loophole — at least of being able to turn off the capture unit — is that EPA actually only evaluates compliance with carbon capture requirement based on capture and does not actually require that that CO2 would be stored in the proposed rule. So we’ll see where that one lands, but hypothetically, that issue of being able to turn off the capture unit at the end of when you’re actually receiving a tax credit is closable.

I think the other place where we’ve gotten a little bit of pushback is just that our capital assumptions are pretty stunning in some cases, so in addition to that 3.5, 3.6 trillion dollars of tax credit that would be issued under the highest deployment scenario that we have looked at, that also engages about 1.2 trillion dollars of capital investment, and that’s based essentially on some DOE estimates of the difference in cost between a new unit that has CCS and a new unit that doesn’t have CCS.

That’s not the same thing as looking at retrofits. We actually think we’re probably underestimating costs that way, but one of the things that has been sort of interesting since the paper came out is that — I think it was Project Tundra, actually, the one with the challenging LCA — has released new estimates of how much their plant conversion is, like with the costs. What they’re saying publicly is pretty much spot on with what we’re estimating for what the costs would be under our model.

So I’m still pretty comfortable with those cost numbers, and they are shocking. But again, those really, really big numbers assume that anyone who is incentivized to build CCS does and succeeds at it, which is probably unlikely.

Stone: So kind of the final question here then is: 45Q is now law. It’s fact. It’s out there. How can we ensure that that money is best spent, in terms of keeping costs within reason and maximizing climate benefits. What are your thoughts?

Grubert: Yes, absolutely. And I think this is one of these good news stories, where what’s out there is scary and has a lot of really bad incentives, but a lot of them are closeable, I think, without necessarily too much effort. So I think the biggest one to me is having actual law or regulatory frameworks to prevent plants from turning off a capture unit once it’s installed. This issue of being able to run it while you’re getting a tax credit, and then go back to running your plant unabated at the end of that is really, I think, where the biggest problem comes from. And that’s again relatively easily solved, potentially by EPA, potentially by states.

Another one, just because plants exist within states is that as state agencies [UNINTEL] and things like that start to go into the process of approving rate-based requests and stuff like that, as plants start to say, “We are actually going to install capture,” is at that level really trying to tie their decisions to say yes or no to whether the plant is creating climate benefits. In some cases, they might not be, and I think that we need to also be fairly comfortable with the idea that probably it’s better to close a lot of these plants than to have them keep running with CCS in many, many cases. But I think the state regulatory apparatus is another backstop that exists and maybe isn’t quite ready for this, necessarily.

The other big issues are really related to making sure that we’re accounting for what’s happening with the carbon pretty well, and I think that also is solvable but will require some more people power and a little bit more direct deliberation about what exactly those things look like. But again, the chance that the very bad scenario happens, I think is quite low, just because these projects are complex. They’re expensive. They’re hard to do, that sort of thing. But the incentive under the law, as it currently exists, is not great.

Stone: Emily, thank you very much for talking.

Grubert: Thank you so much for having me.

Stone: Today’s guest has been Emily Grubert, an Associate Professor of Sustainable Energy Policy at the Keough School of Global Affairs at the University of Notre Dame.


Emily Grubert

Associate Professor of Sustainable Energy Policy, University of Notre Dame
Emily Grubert is an associate professor of sustainable energy policy at the Keough School of Global Affairs at the University of Notre Dame, and former deputy assistant secretary in the Office of Carbon Management at the U.S. Department of Energy.

Andy Stone

Energy Policy Now Host and Producer
Andy Stone is producer and host of Energy Policy Now, the Kleinman Center’s podcast series. He previously worked in business planning with PJM Interconnection and was a senior energy reporter at Forbes Magazine.