Upending Conventional Wisdom on Carbon Pricing in the Power Sector, with Ryan Kellogg

Notable Quotes

• Targeting fossil fuel generators to exit the market: “The potentially happy coincidence is that, if the most expensive plants to keep running (in terms of private dollars that have to be paid in order to generate the megawatt-hours) also happen to be the ones that have high emissions rates, then a [clean electricity standard] and carbon pricing, which really target those high emitters, are going to wind up wiping out the fossil generators in the same order—in the order of highest emissions rate down to the lowest emissions rate.” (17:56)
• Carbon pricing can account for fluctuations in natural gas prices: “You can get to a point where the natural gas plants do become more expensive than the coal plants. Then you do have this problem that a clean electricity standard drives the gas plants out of the market while the coal plants linger around for a long time. That would clearly be a bad outcome, and that’s where having something like a carbon price actually becomes really important.” (21:49)
• Stabilizing fossil fuel demand after deep decarbonization: “One effect of [decarbonization itself] is going to be a reduction of demand for fossil fuels across the board, which is going to tend to drive prices down. And one of the really big challenges of decarbonization is continuing to push clean sources of energy in a world where fossil energy is no longer scarce and is actually plentiful and available. Policy innovation has just put us in a position where we don’t want to buy it anymore.” (23:09)

Top of the Stack

• “Carbon Pricing, Clean Electricity Standards, and Clean Electricity Subsidies on the Path to Zero Emissions” by Severin Borenstein and Ryan Kellogg
• Superpower: One Man’s Quest to Transform American Energy by Russell Gold

The Full Transcript

Daniel Raimi: Hello, and welcome to Resources Radio, a weekly podcast from Resources for the Future. I'm your host, Daniel Raimi.

Today, we talk with Ryan Kellogg, professor at the Harris School of Public Policy and affiliated faculty at the Energy Policy Institute at the University of Chicago. Ryan and his coauthor, Severin Borenstein, recently released a fascinating new working paper that upends some conventional wisdom in energy economics. The paper, “Carbon Pricing, Clean Electricity Standards, and Clean Electricity Subsidies on the Path to Net Zero Emissions,” finds that carbon pricing, long the preferred tool of most economists, may actually not be as efficient as other policy options. In today's conversation, Ryan helps us understand why economic theory has led many to favor carbon pricing and why, in the messy real world, the conventional wisdom may turn out to be wrong. Stay with us.

Ryan Kellogg, my friend from the University of Chicago, welcome to Resources Radio.

Ryan Kellogg: Thank you, Daniel. It's great being here.

Daniel Raimi: So Ryan, I'm shocked that we haven't had you on the podcast before. You work on fascinating topics, and you've written so many great papers over the years. I'm thrilled that we finally get the chance to talk to you. And we're going to talk specifically about a paper that you've co-authored with Severin Borenstein from the University of California, Berkeley.

But before we do that, we ask all of our guests how they got interested in working on energy or environmental issues, whether at a young age or later in life. What drew you into this field?

Ryan Kellogg: There are two pieces to the story. One is how I got interested in economics, and then the other is how I got interested in energy and environment. And believe it or not, the economics piece actually came first, dating all the way back to second grade, which is not a typical place people get introduced to economics. I remember this very vividly, though.

My second-grade teacher, Mrs. Kramer, was interested, one, in economics; and two, in theater. So she wrote a play, a fully-fledged play, that was economics centered. I don't remember the plot anymore; it had something to do with rats. We all had rat costumes, and we were dealing with scarcity of cheese or something. I had one line. I distinctly remember it. I have no idea why this is imprinted in my brain, but it probably has something to do with why I do what I do. And the line was, "Economics is the study of making choices brought about by scarcity." I kid you not. That was my line in second grade.

Daniel Raimi: That's pretty great.

Ryan Kellogg: Mrs. Kramer was fantastic.

After that, I took this long detour into science and engineering. I went to college at Rice University in Texas, down in Houston, with the full intent of becoming a chemical engineer. But while I was there, I became increasingly interested in economics and, more broadly, in environment and energy. I graduated from Rice with a chemical engineering degree. This is in Texas: you get a chemical engineering degree in Texas, and you're probably going to wind up working for an oil company, which is exactly what I did. I worked for BP for four years, which was fascinating. I learned a lot about the world and, obviously, the oil and gas industry in particular.

Then what really got me more interested, one, in academics and two, in energy and environmental policy, was the most fascinating thing I did while I was with BP: economic analysis on this project that never got built, the Alaska Gas Pipeline project. It's a literal pipe dream to take natural gas from the Prudhoe Bay Oil Field—way up north in Alaska, the start point of the Trans-Alaska pipeline, 800 miles north of Anchorage—and build a natural gas pipeline from there all the way down to Chicago. This was 2001, 2002; a long time ago now. A $20-billion project.

The economics then were shaky, and this was in a world where people thought the United States would be importing gas and there would be a gas shortage because there was no such thing as shale then. It's a really good thing the project didn't get built because it would've finished right when the shale boom was kicking into high gear and natural gas prices in North America tanked. We missed all that. That was good.

But part of my experience was with policy because the pipeline was so intertwined with state policy and federal policy. Both the United States and Canada were doing a lot of work trying to model sensitivities towards various tax credits, regulatory policies, and so on. There was an energy bill up for debate in Congress at the time. That part was really fascinating.

Then, as I wanted to pursue a career in energy and energy economics, I thought, Do I want to be doing that working in private industry, or do I want to take a more academic route? In 2003, I applied to graduate school, went off to do a PhD, and then became the research egghead that I am today.

Daniel Raimi: That's great. Next time I see you, I'm going to bring some cheese, and we'll decide how to divide it up between ourselves. I can say from firsthand experience: you may be an egghead, but you are also pretty fun to hang out with.

Let's talk now about this new paper, which you've coauthored with Severin Borenstein, who we've had on the show. The paper is focused on three types of policies aimed at reducing emissions from the electricity sector. Those three types of policies are carbon pricing, clean electricity standards (CES), and clean electricity subsidies. We're going to talk about those three types of policies. Our listeners probably have an intuition as to what they all are, but to get us all grounded, can you give us a quick primer on how each of those are supposed to work?

Ryan Kellogg: You bet. Let's start with carbon pricing, which is the easiest to get your arms around. The basic idea of carbon pricing is to put a price on carbon emissions from fossil fuel generators. It's probably easiest to think about in the context of an explicit carbon tax, where if you're a coal-fire generator or a natural gas–fire generator, you pay a price on every ton of carbon dioxide that comes out of your stack: $40 a ton, $50 a ton, $100 a ton—wherever that price is set.

You can implement that with an explicit carbon tax, so the money goes straight to the Feds, or, alternatively—and this is effectively part of the system that California has and the northeastern states have—you set up a cap-and-trade market where for every pound of carbon dioxide you emit, you have to find and buy an emissions permit on the open market. That acts very similarly to a carbon tax, in terms of the price incentive that it places on trying to reduce your emissions.

Now, moving on to clean electricity standards, or a CES, the mechanism is different in a couple important ways. One, unlike carbon pricing, a CES doesn't directly tax fossil generators’ emissions. Instead, what you're going to do under a CES is classify all the generators into two types: either you're clean, or you're dirty, where clean typically is meant to mean basically zero emissions or really close to zero emissions.

What would qualify as clean? Obviously wind and solar; also nuclear; probably geothermal. In principle, you could imagine a fossil generator with really good carbon capture and sequestration bolted onto it so that all the carbon emissions wind up getting re-injected into the ground. That’s a very expensive technology and not developed at scale, but in principle it would be covered under a CES.

Once you've made that distinction between, "All right, these plants are clean; these other plants are dirty," the clean electricity standard then says, "All right, let's say in a given year, 2030, we're going to mandate that all generation hit a certain clean electricity share target." For instance, 60 percent of all generation needs to be clean by a certain date. That clean generation percentage can ramp up over time. Right now, around 40 percent of US electricity generation is clean, including hydro and nuclear. You can imagine ramping that up to 80 or 90 percent over a period of 15 years. That electricity standard incentivizes clean generation and penalizes anything that's dirty.

Daniel Raimi: Got it. And then our third one is clean electricity subsidies.

Ryan Kellogg: A clean electricity subsidy, which also annoyingly has the initial CES, in the paper we call them zero-emission subsidies to try to make the distinction, but we'll call them clean electricity subsidies. They're almost the flip side of carbon pricing. You don't really do anything with the dirty generators, in terms of policy that directly affects them. Instead, you subsidize electricity generated (megawatt-hours generated) by any source that's designated as clean. The idea there is, by subsidizing the clean generation, you're going to give them a leg up in the electricity markets so that clean generators can better push out the dirty generators over time.

Daniel Raimi: In practice, in the United States, we've done the third policy, right? We have clean electricity subsidies that are in place, and at the federal level, we have no electricity system–wide carbon pricing or clean electricity standard.

Ryan Kellogg: That's right. At the federal level, of those three policies—carbon pricing, clean electricity standard, clean electricity subsidy—all we've done is clean electricity subsidy. In particular, the tax credits for wind and solar generation have been around for the past 20 years or so.

Different states have implemented carbon pricing: California and some of the northeastern states, for instance. A lot of states have implemented something like a clean electricity standard. These are renewable portfolio standards that dictate a certain share of electricity procured by utilities needs to come from renewable sources. Typically, those target wind and solar and not things like large hydro or nuclear or geothermal or fossil with carbon capture.

Daniel Raimi: Great. We've got some really good background in place. Let's go now to the analysis that you carry out with Severin.

One of the starting points for the analysis is the common wisdom among most economists who study energy and electricity policy, which is that carbon pricing in general will tend to reduce emissions at the lowest cost to society. Can you give us the logic for that underpinning common wisdom? Then we'll upend it in a couple minutes. But first, what's the common wisdom?

Ryan Kellogg: Here's the what I think of as the three-piece version of the common wisdom. And I'll add: these pieces of logic had been my own personal thinking about why carbon pricing was, in some sense, the efficient, market-based policy for decarbonizing the grid.

Part one goes back to this distinction between carbon pricing that taxes emissions versus clean electricity standards that separate generators into two categories: clean versus dirty. The worry with the CES is that, within dirty, there's variation in how dirty different generators are. In particular, coal plants have really high emissions rates of carbon dioxide per megawatt-hour of electricity that they generate. More modern gas plants—especially combined-cycle natural gas plants, highly efficient natural gas plants that get the most that they can out of the gas—have substantially lower rates of emissions than coal plants. Carbon pricing distinguishes between those two types of generation. If you're a coal plant, you're going to be paying a much higher emissions tax per unit of electricity that you generate than if you're an efficient natural gas plant.

A CES doesn't distinguish between that at all. From the perspective of the CES, a coal plant is dirty. A gas plant is dirty. You effectively pay the same market penalty per megawatt-hour that you generate. That leads to the concern that as you implement a CES, you're going to be pushing fossil generation off the grid and replacing it with clean energy, but you're not necessarily going to be displacing the dirtiest fossil generation. That's worry number one with the CES, and carbon pricing has a leg up on efficiency from that perspective.

Part two of the story is about the idea of getting electricity prices right. One of the mechanisms for reducing emissions from the electricity sector is to use less electricity, at least as long as the grid still has some fossil fuels in it. And the nice thing about a carbon price—and we can upend this later—but, in principle, a carbon price bakes into the price of power the cost of emissions, so end-use consumers can see that cost and respond to it by reducing their use of electricity, whether it's turning their air conditioning down, investing in energy efficiency—all the different ways you might imagine industrial, commercial, or residential consumers trying to conserve power. Carbon pricing is really good at sending those sorts of price signals.

Clean electricity standards are less good. And if you think about a clean electricity subsidy, that actually lowers the price of electricity, rather than raise it. In some sense, it moves in the opposite direction of a carbon price.

The final and third piece is about government revenue. A nice side benefit of carbon pricing is that you're generating revenue for the government, and that's revenue that the government can use however it pleases. It can reduce other taxes like income taxes and things like that. It can spend the revenue on various other programs. Because it's all going into the big fiscal budget, it can be redistributed as progressively or regressively as people want it to be distributed. That's all nice.

A CES doesn't generate any revenue for the government. And then, of course, the clean electricity subsidy actually involves the government paying out revenue. And that revenue ultimately has to come from somewhere, either additional taxation, cutbacks and spending elsewhere, or deficit increases that have to be paid for somewhere down the road. Those budget differences are important, though it is useful to point out that if we're thinking about deep decarbonization of the grid, one thing people sometimes forget about carbon pricing is that if you actually drive emissions to zero, you're not generating any revenue anymore. You might have a really high carbon price and a really high carbon tax rate. But if you drive all the fossil fuels out of the grid, you're multiplying that tax rate times zero carbon dioxide, so the tax revenue ultimately goes away with carbon pricing if you crank down at it.

Daniel Raimi: Over the long term, for sure. That is extremely helpful.

Let's talk now about comparing some of your analysis and findings with regard to carbon pricing versus clean electricity standards. We're going to talk about subsidies after that, but electricity standards first. Your analysis shows that, under certain circumstances, using a CES could be just as efficient, from society's perspective, as a carbon price. Can you help us understand how you get there?

Ryan Kellogg: This was the big, surprising result to Severin and myself as we worked on this paper, and this really cuts at the first reason I gave a few minutes ago for why economists have been skeptical of a CES. A CES doesn't do a great job of targeting the really dirty coal plants because it lumps together coal and gas together in a category called dirty. If you think about ramping up a CES over time from 40 percent now to something more like 80 or even 90 percent in the future—and as renewables penetrate the grid and start pushing out fossil fuels—what fossil fuel is going to exit the market first?

Electricity markets are intensely competitive. The fossil generators that are going to leave the markets first are the ones that have the highest ongoing operating costs, the ones that the operators find the most expensive to keep around. They're the ones that are going to have the hardest time competing with all the new green generation that's being pushed by the CES policy.

So, if you want to think about which fossil generators are going to leave the market first, you need to think about the rank ordering of the most expensive fossil plants to keep running. What happens as you look at cheaper and cheaper plants as you go down the merit order? The potentially happy coincidence is that, if the most expensive plants to keep running (in terms of private dollars that have to be paid in order to generate the megawatt-hours) also happen to be the ones that have high emissions rates, then a CES and carbon pricing, which really target those high emitters, are going to wind up wiping out the fossil generators in the same order—in the order of highest emissions rate down to the lowest emissions rate.

But this really comes down to trying to understand how correlated emissions rates and private operating costs are across all the fossil generators in the United States. Put another way, do units that have high operating costs also have high emissions rates, and do units that have low operating costs also have low emissions rates?

Severin and I go to the data to see what that correlation actually looks like. The Energy Information Administration (EIA) provides all this data readily. We gather all this data for the year 2019, which we think of as the last pre-COVID normal year. What we see—it's really shockingly stark in the data—is that the big, dirty coal plants that have high emissions rates, under 2019 conditions, are much more expensive to keep running than the newer, more efficient combined-cycle gas turbines.

The upshot of this is that—if you think about modeling, and we model this in a very simple way—if you model what happens as you crank up a CES from 40 to 90 percent clean energy over time and compare that to a policy of cranking up a carbon tax that would also eventually get the grid to 90 percent clean, this gets you to a carbon tax on the order of $150 per ton of carbon dioxide. It's a big tax.

The differences in emissions between those two policies during the transition is tiny: on the order of about two percent of total emissions during the transition. That's coming from the fact that both the CES and a carbon price are effectively getting rid of the fossil plants in the same order. The coal plants go away first, and then you start taking out the more efficient natural gas plants.

So this worry that I and many economists and others have had about a CES not targeting the dirty plants you want to get out of the grid may not matter enough because those really dirty plants happen to also be really expensive to run, and they're going to have a really hard time competing in the presence of a CES.

Daniel Raimi: One thing that I certainly thought about that conclusion and that I imagine some of our listeners are thinking about is what happens if natural gas prices change, which, of course, they have historically; natural gas prices have been pretty volatile. Over much of the last 10 to 15 years, natural gas prices have been really low because of the shale revolution, but, just in the last six months to a year, natural gas prices have gone up dramatically. They've more than doubled and maybe even more than tripled during some times over the last year or so. What happens in future scenarios where natural gas prices are substantially higher than you might assume in a paper that focuses on 2019?

Ryan Kellogg: If you look at a scenario where gas prices are much higher than they were in 2019 (which was about $3 per million British thermal units, about the same price that had prevailed for much of the 2010s, mostly due to shale gas), without a commensurate increase in the coal price, you can get a reversal. That is, you can get to a point where the natural gas plants do become more expensive than the coal plants. Then you do have this problem that a clean electricity standard drives the gas plants out of the market while the coal plants linger around for a long time. That would clearly be a bad outcome, and that's where having something like a carbon price actually becomes really important. That scenario is out there. It's something to worry about.

That said, Severin and I don't worry about it too much for two reasons. One: look at natural gas prices now, which have, at various points in time this year, been triple what they had historically been, going back to 2019 and earlier. Coal prices are also really high, so coal is still going to have a hard time competing on the margin as you add more renewables to the grid.

Then, two: As you think about a world where we are enacting these stringent policies and aggressively decarbonizing the grid, that's a world where natural gas isn't scarce because we've been driving natural gas and other fossil fuels effectively out of the market. That's a world where natural gas is cheap because the grid isn't demanding it anymore because of these aggressive decarbonization policies. That is, one effect of decarbonization itself is going to be a reduction of demand for fossil fuels across the board, which is going to tend to drive prices down. And one of the really big challenges of decarbonization is continuing to push clean sources of energy in a world where fossil energy is no longer scarce and is actually plentiful and available. Policy innovation has just put us in a position where we don't want to buy it anymore.

Daniel Raimi: That's a great point. We could talk about price issues and expectations about future prices for a long time, and maybe we will when we're done recording the podcast, but let's talk now about the subsidy option in comparing subsidies with the carbon price.

One of the concerns about clean electricity subsidies is that they'll tend to reduce electricity prices, which will encourage people to use more electricity, which will result in more pollution overall as long as there are dirty sources on the grid. What do you find when you look at this question of whether carbon pricing or subsidies would send better or worse pricing signals to consumers?

Ryan Kellogg: The conventional story that a carbon price is really good at sending the right price to consumers in trying to get consumers to conserve, especially when dirty generation is still on the grid, hinges on retail price signals reflecting actual costs of generation, including the costs of any emissions. That story about retail pricing, though, turns out to be wrong for the United States.

This point really piggybacks off of some past work Severin Borenstein, my coauthor on this paper, has done with Jim Bushnell over the past few years. (Jim Bushnell is a professor at the University of California, Davis.) They look at retail markups of electricity prices that consumers pay across the US. Let me say a bit about what that means.

Let me think about my own electricity bill here in Chicago. I'm a customer of ComEd. They're the big distribution utility that serves Chicagoland. My electricity rate is about 11-and-a-half cents per kilowatt-hour. Electricity's cheap here in the Midwest. If I look at my bill breakdown—and I sometimes do because I'm an energy geek, I don't think most people do this—what I'll see is that of that 11-and-a-half cents, about four to five cents of that is ComEd's cost of procuring power on wholesale markets or generating some of that power itself. That's the actual private, wholesale market cost of power, what it pays to generators. The rest of that difference, something like seven cents per kilowatt-hour, pays for ComEd's fixed distribution costs: what it costs to maintain and operate all the distribution power lines, pay people who work in the central office, all of the fixed costs of running a big utility. But that's all paid on a dollar-per-kilowatt-hour basis.

Those fixed costs, they don't really vary with how much electricity I consume, but, in principle, they do discourage me from doing things that would increase my electricity consumption because I'm paying 11-and-a-half cents rather than four cents. For instance, if I'm thinking about getting an electric vehicle (EV), which is one of the things I'm thinking about, the electricity rate factors into that trade-off.

Here in Chicago, in the Midwest, we still have a fairly dirty grid. Paying a somewhat marked-up price for power is maybe okay from that perspective, but if I look at California or New York or parts of New England that have cleaner grids that don't have a lot of coal on them—unlike here in Chicago—retail electricity prices are much higher than the actual cost of generation, including any costs you assign to emissions coming from the fossil generators that are on the grid.

Retail rates in California and New York are in excess of 30 cents per kilowatt-hour. For most customers, that's really high. (And by New York, I mean New York City.) As the grid gets cleaner across the United States, what places like California and New York are experiencing now is going to happen everywhere. We're going to have high retail markups that discourage electricity use rather than promote it when the grid is clean.

Getting back to carbon pricing, now—what does carbon pricing do? It's going to increase the price of electricity even more, which exacerbates all these utility retail price markups that already exist. So, if consumers now pay even more for what will be clean electricity—which is particularly problematic if one of our decarbonization goals is to use clean electricity to decarbonize other sectors like transportation via EVs, because why would I buy an EV if electricity costs 30 cents per kilowatt-hour?—getting retail prices right from that perspective seems to argue more for a policy like a clean energy subsidy that leads to lower clean electricity prices, rather than higher clean electricity prices. It's a big turnaround from the conventional thinking on carbon pricing versus subsidizing clean energy.

Daniel Raimi: Based on some of the feedback I've seen online to the paper, this work is definitely getting people thinking and helping people question their assumptions about what they may have thought beforehand about carbon pricing and these other types of policies. It's a really great contribution. It's a fascinating paper. Of course, we'll have a link to it in the show notes for people to check out.

Now we're going to go to our last question that we ask all of our guests, which is asking you, Ryan, to recommend something that is at the top of your literal or your metaphorical reading stack, something you've read or you've watched or you've heard that you think is great and that you think our listeners would enjoy.

Ryan Kellogg: I recently read the book Superpower by Russell Gold. I highly recommend it. It's a book about an effort by this company, Clean Line Energy Partners, and, in particular, by the owner-CEO of this company, Michael Skelly, to develop a large, utility-scale, independent wind farm in the Oklahoma panhandle, where nobody really lives or very few people live but there's an awful lot of wind, and to try and develop a transmission line to get that wind to major population centers out East, Memphis in particular, as the plug-in point to the eastern grid.

This was around 2010 or so, and the book does a fabulous job of telling the story of all the obstacles that Michael Skelly and his company ran into while trying to get this big, interstate, clean electricity transmission line built: securing permission from the various states and municipalities that the line would have to go through, and trying to find a utility buyer on the other end (in this case, the Tennessee Valley Authority) to actually want to buy the electricity rather than using their own generators. The book does a nice job of laying out how substantial each of these obstacles were.

I won't spoil the ending, but the obstacles are serious and they speak to things that Severin and I don't really talk about in our own paper. Subsidies for clean electricity are good, useful things, and we should be doing them, but their efficacy is going to be limited by the fact that, to get the most that we can out of renewable technologies like wind and solar, we need to find ways to get electricity from the parts of the country that have good wind and solar resources—which tends to be right down the center of the country, which is not very densely populated—out to the major population centers.

Daniel Raimi: It is a really fascinating book. I read it when it first came out, and it's great. There's actually an announcement this week–we’re recording this the last week of July—from the Midwest Interconnection System Operator that there was a new agreement to build out lots of new high-voltage transmission lines. I'm not sure if that's actually going to go forward as planned, but I did at least see that announcement, which is maybe a hopeful note in this story.

Ryan Kellogg: I have some hope for that. There is some controversy with regards to that announcement. The projects that are being considered—my understanding is that they are utility-sponsored projects, and some of the independent transmission-line builders are arguing that they were cut out of the process. I haven't dug into the facts quite enough to have a full grasp of that, though.

Daniel Raimi: Interesting. Definitely a topic for a future podcast.

Well, Ryan Kellogg from the University of Chicago, once again thank you so much for coming onto the show today, for helping us understand this fascinating paper, and for sharing the work with us. We really appreciate it.

Ryan Kellogg: Thanks for having me, Daniel. This has been a lot of fun.

Daniel Raimi: You've been listening to Resources Radio, a podcast from Resources for the Future (RFF). If you have a minute, we'd really appreciate you leaving us a rating or comment on your podcast platform of choice. Also, feel free to send us your suggestions for future episodes.

This podcast is made possible with the generous financial support of our listeners. You can help us continue producing these kinds of discussions on the topics that you care about by making a donation to Resources for the Future online at rff.org/donate.

RFF is an independent, nonprofit research institution in Washington, DC. Our mission is to improve environmental, energy, and natural resource decisions through impartial economic research and policy engagement. The views expressed on this podcast are solely those of the podcast guests and may differ from those of RFF experts, its officers, or its directors. RFF does not take positions on specific legislative proposals.

Resources Radio is produced by Elizabeth Wason with music by me, Daniel Raimi. Join us next week for another episode.