A New Social Cost of Carbon, with Brian C. Prest and Kevin Rennert

Notable Quotes

• Defining the social cost of carbon: “The social cost of carbon is an economic measure of the marginal damage from an additional ton of greenhouse gas emissions … It basically says that if you have a baseline of greenhouse gas emissions, and you emit an additional ton of greenhouse gas emissions, what is the effective dollar cost to society—the social cost from the addition of that ton of greenhouse gas emissions? Or you can flip that around and say, What is the dollar benefit of not emitting that ton of greenhouse gas emissions? That’s what the social cost of carbon will tell you.” —Kevin Rennert (2:40)
• Revealing the new estimate of the social cost of carbon: “Here is the big reveal, or estimate, for the [social cost of carbon]: After putting together all the pieces—the socioeconomics, the emissions, the climate modeling, the temperature [change], the projections, the climate impact studies, and the discounting—we ultimately come to an estimate for the [social cost of carbon] of about $185 per ton of carbon dioxide. How does that compare to the other estimates out there? The natural benchmark estimate is the existing interim estimate used by the federal government, which amounts to $51 per ton of carbon dioxide.” —Brian C. Prest (7:21)
• Importance of open-source, transparent research: “What the federal government ends up deciding to do is still very much to be determined. We were coming up with our science and our updates, and we pursued the process with transparency in the hopes that it will facilitate the government updates. But again, the decision on what they will use is up to the federal government in the end … The change in the value is important, but another important aspect of the work that we’ve done is its transparency and the fact that it has all gone through peer review.” —Kevin Rennert (13:47)
• Considering uncertainties that affect the estimate: “One big uncertainty is how big the economy is going to be 100, 200, or even 300 years from now … If you look at long-run projections of the rate of economic growth, it could be as low as less than one percent by 2300, or it could be more than three percent. Those differences may sound small, but the power of compounding means small changes in growth rates can have big changes in outcomes in the size of the economy, and hence the degree of vulnerability to a changing climate.”
  —Brian C. Prest (18:30)

Top of the Stack

• “Comprehensive Evidence Implies a Higher Social Cost of CO2” by Kevin Rennert, Frank Errickson, Brian C. Prest, Lisa Rennels, Richard G. Newell, William Pizer, Cora Kingdon, Jordan Wingenroth, Roger Cooke, Bryan Parthum, David Smith, Kevin Cromar, Delavane Diaz, Frances C. Moore, Ulrich K. Müller, Richard J. Plevin, Adrian E. Raftery, Hana Ševčíková, Hannah Sheets, James H. Stock, Tammy Tan, Mark Watson, Tony E. Wong, and David Anthoff
• “An Updated Social Cost of Carbon: Calculating the Cost of Climate Change” RFF Live event
• “The Social Cost of Carbon: Reaching a New Estimate” by Brian C. Prest, Jordan Wingenroth, and Kevin Rennert
• Social Cost of Carbon Initiative at Resources for the Future
• Mimi Integrated Assessment Modeling Framework
• The Rule of Five: Making Climate History at the Supreme Court by Richard J. Lazarus

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 two RFF fellows, Brian C. Prest and Kevin Rennert. Brian and Kevin, along with a large team of collaborators, recently released a paper in the journal Nature that provides a new estimate of the social cost of carbon (SCC). In today's episode, I'll ask them to describe what's new about their estimate, how it differs from previous numbers, and what it implies for policymaking. We'll also talk about the important uncertainties that surround their estimate, as well as some critics’ arguments that we should put less focus on using the social cost of carbon. Stay with us.

All right, Kevin Rennert and Brian Prest from Resources For The Future. My colleagues, my friends—welcome back to Resources Radio.

Kevin Rennert: It's wonderful to be here.

Brian C. Prest: Thank you for having us.

Daniel Raimi: Because you've both been on the show before, we're not going to ask you how you got interested in energy and environmental topics. But—because the topic that we're going to talk about today, the social cost of carbon and your updates to the social cost of carbon, has been done with a substantial team of folks, and I know that you want to acknowledge their contributions at the outset—we are going to ask you: Can you tell us who you worked with on this project?

Kevin Rennert: It has been a tremendous team of folks. It's been a collaboration primarily between RFF and the University of California, Berkeley (UC Berkeley). The core team out at UC Berkeley has been David Anthoff and his lab, with Lisa Rennels and also Cora Kingdon. Here at RFF, it's been Brian, myself, Jordan Wingenroth, Roger Cooke, Richard Newell, and Billy Pizer. Frank Errickson at Princeton also has been a part of that core team.

On the actual paper whose results we're talking about today, there are even more coauthors because we’ve been collaborating with a huge number of institutions to come up with this update.

Daniel Raimi: We're not going to go into the background of the project here in much detail, but I know that it's been a multiyear, multi-institution, herculean effort to come up with this new estimate of the social cost of carbon that we're going to talk about today.

Before we get into it and tell everybody what the number is, let's just define what the social cost of carbon is. Most listeners probably know, but for those of us who could use a refresher, can you define the term for us? And then tell us how the SCC is used in policymaking in the United States and elsewhere.

Kevin Rennert: The social cost of carbon is an economic measure of the marginal damage from an additional ton of greenhouse gas emissions. To unpack that a bit, it basically says that if you have a baseline of greenhouse gas emissions, and you emit an additional ton of greenhouse gas emissions, what is the effective dollar cost to society—the social cost from the addition of that ton of greenhouse gas emissions? Or you can flip that around and say, What is the dollar benefit of not emitting that ton of greenhouse gas emissions? That’s what the social cost of carbon will tell you.

It's an important number and an important concept because oftentimes policymakers want to evaluate trade-offs between the decisions that they're making. They might say, “I'm thinking about imposing a regulation," for example, and that regulation is going to have some costs associated with it in terms of potentially changing the equipment that someone who’s polluting might need to use to control that pollution, or the standard I'm going to set for energy efficiency for a particular line of products.

When they're setting those standards and thinking about the cost, they also want to know what the benefits might be. The social cost of carbon allows them to quantify the emissions that would be reduced by taking that action and see what the dollar benefit to society is as a result, and then compare those with the costs. They have some more information for that process.

The SCC gets used by the federal government in the United States, in regulatory actions, and in efficiency standards, but it also gets used at the state level. There are states that think about it in terms of having their utilities that are producing electricity think about the social cost of carbon as they're planning additional electric power coming onto the grid. They're evaluating whether it should be natural gas or renewables, things like that. There are some states that actually pay their utilities based on the value of the social cost of carbon for zero-emissions generation and other things of that nature.

Daniel Raimi: That's great background. We are going to get to the number in just a second, but to produce an estimate of the social cost of carbon, there are several really important analytical steps that you have to take to get there. What are those key steps that you and the team have taken?

Kevin Rennert: That's exactly right. To unpack the number, you have to realize that it's a very rich intellectual subject because it combines all these different steps. If you want to know what the cost to society is, you need to know where society is heading. You need very long-run projections of socioeconomics: what the emissions are going to be, what the actual income is going to be, what the population is going to be.

You also need to have information about what the climate system is going to do. You're going to need to have a climate model involved. Then you need to look at the changes in the climate and be able to translate those into economic impacts. You need some information from research studies that will translate them into economic damages. Finally, these damages are going to persist for a long time because greenhouse gas emissions stay in the atmosphere for a very long time. You want to sum up those damages and discount them back into a net present-value term.

All of those different pieces are required to estimate the social cost of carbon. There was a set of recommendations from the National Academies of Sciences that looked at the models that were out there for estimating the SCC back in 2017, and it found that all of them needed substantial updates across these four pieces.

Our team embarked on research and pulling in research from the literature to improve all these different pieces. The first piece was obviously these long run projections of socioeconomics. We had an entire research project working with some of the top demographers and economic growth–projection economists to look at what's happening with future economic growth and population and future projections of emissions. We brought in a state-of-the-science climate model. We looked to the state of the science from the literature from economic impact. Brian, Richard Newell, and Billy Pizer also came up with a novel method for determining the economic discounting piece, as well.

Daniel Raimi: We could almost certainly spend entire podcast episodes talking about any one of those components, but we're going to zip through them and move on to the big drumroll. What results do you come to after you take all these steps? What is your estimate for the SCC, and how does it compare to some of the other estimates that folks listening may have heard of?

Brian C. Prest: Here is the big reveal, or estimate, for the SCC: After putting together all the pieces—the socioeconomics, the emissions, the climate modeling, the temperature, the projections, the climate impact studies, and the discounting—we ultimately come to an estimate for the SCC of about $185 per ton of carbon dioxide.

How does that compare to the other estimates out there? The natural benchmark estimate is the existing interim estimate used by the federal government, which amounts to $51 per ton of carbon dioxide. In the paper, we also do a stepwise comparison, where we look at the estimate from Bill Nordhaus’s DICE model, which came out to a number that's very similar to that $51 federal estimate. (Their number was about $44.) Then we incrementally add different pieces of our update to show how the different pieces that Kevin alluded to drive that change from $44 to $185, which is our estimate.

The first step is where we implement all of the socioeconomic uncertainty, the emissions, and the modern climate model that Kevin mentioned. But we keep the DICE model’s damage function, which is a widely used and understood function for estimating how temperature affects GDP losses. All of those pieces increase the SCC by about a third.

Next, we update those damage functions and the DICE simple damage function to our sector-specific damage functions, where we account for impacts on four major sectors: heat-related mortality, agriculture, energy expenditures from heating and cooling buildings, and coastal impacts from sea level rise. When we do, that increases the SCC by about another third.

The last piece is where we update the all-important discount rate. The SCC is well known to be sensitive to the discount rate because of the long-lived impacts of greenhouse gas emissions. There, we update the discount rate from the historically used three percent to two percent, which is more reflective of recent trends in market interest rates that have been declining for decades now. When we update that, the SCC goes from about $80 at a three percent discount rate to $185 at a two percent near-term discount rate. That last piece approximately doubles the SCC.

Daniel Raimi: Fascinating. That's probably surprising to our listeners, but it's just always so interesting to understand how important that discount rate is to the ultimate number. Because it is so important, it's probably worth taking a moment to explain to our listeners. What is the discount rate? Why is it used in the social cost of carbon? And how does what appears to be a small tweak in the discount rate from three to two percent lead us to such a substantial change in the ultimate number?

Brian C. Prest: The discount rate involves taking impacts that might be felt in the future, 10, 20, or a hundred or more years down the line, and converting it into a value in present value—essentially today's equivalent value. How much are we willing to pay today to mitigate a dollar or more of impacts 50 years from now?

This is important because there are many reasons that people and society value having benefits earlier rather than later. The big change when you go from three percent to two percent owes to the power of compounding. A small change in interest rate extended over a hundred years can lead to a very large change in the present value of those impacts.

Daniel Raimi: Your projections go through the year 2300, is that right?

Brian C. Prest: That's correct.

Daniel Raimi: When you go out that far, there's a huge effect. Can you dig one level deeper? Why is it that you and the other folks on the team have decided that two percent is the right number in this context?

Brian C. Prest: Two percent has been a focal point for many economists in recent years for an updated discount rate. The reason is simple and straightforward: during the Bush administration, back in 2003, the Office of Management and Budget issued some guidance for how federal analysts should do cost-benefit analysis.

One piece of that was what discount rates to use. They landed on a three percent discount rate for discounting what you call “consumption impacts,” such as climate impacts. That was based on a 30-year look-back, where they reviewed the real rates of return on government debt on average for the previous 30 years. That number came to about three percent, so they recommended three percent.

Of course, things have changed quite a bit. Interest rates have come down over the past 20 years. If you do that same 30-year analysis that was done in 2003—but not looking back from 2003, looking back from today—that same calculation gets you about a two percent discount rate. It's really the same logic underlying long-standing guidance for discounting in cost-benefit analysis.

Daniel Raimi: Thank you. Kevin, I'd like to turn to you and ask this next question: What is this higher SCC that is more than three times higher than the current number being used by the Environmental Protection Agency (EPA)? What does that mean for climate policy in the United States?

Kevin Rennert: I should make a clear distinction here that we are putting out this number as a result of our own independent research. What the federal government ends up deciding to do is still very much to be determined. We were coming up with our science and our updates, and we pursued the process with transparency in the hopes that it will facilitate the government updates. But again, the decision on what they will use is up to the federal government in the end.

So, what if these numbers that get pulled in by the federal government in the SCC were increased substantially? The answer to your question is that it motivates more stringent policies when it comes to reducing greenhouse gas emissions. That's because, as the different regulatory agencies—and I mentioned energy efficiency standards and other ways it can get used—think about the trade-offs of different actions that could be taken, more stringent actions will suddenly pass a benefit-cost test that they might not have passed with a lower number.

That's not to say that a higher number necessarily means that the federal government will take actions that it wouldn't have taken otherwise. In many cases, the benefit-cost test is one component of information that they look at. And just knowing that something passes doesn't mean that if the SCC were to be $185, that actually ends up getting passed on to consumers. It's a piece of information that they use to decide about the level of action that is taken.

The change in the value is important, but another important aspect of the work that we’ve done is its transparency and the fact that it has all gone through peer review. The paper that we were putting out was peer reviewed, and all of the components that we built, the model that we're talking about, and the results from today were also already independently and scientifically reviewed.

As the federal government looks to pull in additional information, that scrutiny is going to make it more likely that, in addition to the scrutiny that the federal government would put their updated numbers through, it will be legally defensible as well. Because you want to be sure that your regulations themselves are based on as solid and scientifically-credible evidence as possible to make sure that they're not vulnerable to legal challenge, as well.

Daniel Raimi: Legal challenges will certainly come once EPA updates its number. It's already in being litigated right now, right?

Kevin Rennert: That's right.

Daniel Raimi: We've talked about the central $185 value that the team has come up with. We haven't talked yet much about the uncertainties around that number, and the uncertainties are substantial. There's a 5–95 percent confidence interval around that $185 value that is as low as $44 and as high as $400. Can you talk a bit about why that range is so large and what some of the biggest contributors to that wide range might be?

Brian C. Prest: As Kevin mentioned, a key contribution of our work here is that we quantitatively characterized uncertainties in all of the key pieces: the socioeconomics population, GDP projections, emissions projections, uncertainties in how the climate system responds to a pulse of emissions, and uncertainties on how rising temperatures and sea levels affect society; then, also, uncertainty in the proper discount rate out into the future. We’ve accounted for all of those key things, which has proven to be pretty important.

In terms of a quantitative disaggregation or untangling of all of the different contributions of each of these uncertain inputs, we haven't done a formal analysis of that yet. We would like to and plan to do so in future work. One thing that I can say is that a key uncertainty is the socioeconomics: understanding the range of potential future populations that might be affected by climate, the size of the economy that might be vulnerable to climate impacts, and, finally, a likely range of potential future emissions trajectories into the future. These are key uncertainties, and we plan to do a more quantitative disaggregation of each one's contribution in the future.

Daniel Raimi: For the sake of our listeners who haven't worked with large data sets, projecting into the future, and uncertainty, can you pick a random example from the work and give us a sense of how wide the range of uncertainty is from whatever element of the work might be a good example?

Brian C. Prest: One big uncertainty is how big the economy is going to be 100, 200, or even 300 years from now. We’ve drawn on recent econometric work to understand the volatility and range of uncertainty for future economic growth rates, as well as done our own work to subject that to expert elicitation. If you look at long-run projections of the rate of economic growth, it could be as low as less than one percent by 2300, or it could be more than three percent. Those differences may sound small, but the power of compounding means small changes in growth rates can have big changes in outcomes in the size of the economy, and hence the degree of vulnerability to a changing climate.

Daniel Raimi: Folks can, I think, wrap their heads around that economic growth uncertainty because we live with it every day, right? We hear about it on the news almost every night.

A couple more questions now before we go to our Top of the Stack segment, which is to ask you to a bit more about uncertainty and, particularly, what Donald Rumsfeld would call the “known unknowns”: potential climate impacts that we are confident are out there and could happen but that you were not able to quantify in your calculations. I’m curious about what some of those big-ticket items might be and how significant you think their effects could be if they were to be incorporated into the SCC in the future.

Brian C. Prest: We have accounted for four major sectors in a clear, dollar-valued, monetized impact. Those are heat-related mortality, agriculture, energy expenditures for heating and cooling, and coastal impacts from sea level rise. But there are lots of impacts from climate change that aren't on that list. We hope to include those in the future.

In fact, the modeling platform that we have developed for this purpose is designed to easily facilitate the addition of new sectors and new impacts over time. I’ll focus on two examples that we hope to include in the future. One missing piece is the impact of climate change on global biodiversity loss. There is a literature on the impact of rising temperatures on species loss, but putting a dollar value on that is not a simple or straightforward task. We have ongoing work where we plan to put that dollar value on the impact of biodiversity loss. I'm not going to speculate on what the dollar value is yet, but it could be a nontrivial addition to the impacts that we have modeled.

Another piece that we haven't monetized yet is ocean acidification. Rising carbon dioxide levels in the atmosphere get absorbed by the ocean. This raises the acidity of the ocean (or lowers the pH). That has many different effects, including degradation of corals, as well as impacts on shellfish, which has economic consequences for fisheries. Our model actually has a component where we quantify the impact of rising carbon dioxide levels on ocean acidity. We can report the change in pH levels of the ocean, but we don't yet have a way to translate that into dollar terms. That would be a great place for additional work to continue to contribute to our estimates.

Then there are tons of other options out there: impacts on migration, conflict, labor productivity, and so on. All these things are not included in our estimate, but could be with additional work in the future.

Daniel Raimi: For our research audience that's listening out there—still lots of work to do and ways to contribute to this number! And we wouldn't think of this SCC as the final SCC, right? But it's a substantial update to what we have.

You've both mentioned the open-source transparency and the platform that you all have developed to share the information and the results. Can you point our listeners to where they can go to find that information, access it, look at the data, analyze the models—all that stuff?

Kevin Rennert: Sure. If you go to rff.org/scc, you'll find all of the information that RFF has on the social cost of carbon aggregated in one place. There are some pretty clear links there to the Mimi platform, as well as out to David Anthoff's lab.

Daniel Raimi: Does Mimi stand for something, or is it just a cute name?

Kevin Rennert: It's just a cute name. There's a long-standing joke about it, but I can't go into it here.

Daniel Raimi: One more question before we go to our Top of the Stack, which is a question that I'm sure some of our listeners are thinking about. There are some who have criticized the entire approach that we're talking about today—who are critical of using an SCC approach. Folks will argue instead that the best way to think about dealing with climate change is not to try to balance benefits and costs, but instead to pick a clear temperature target—1.5 or two degrees Celsius by 2100—and then throw everything we need to throw at the problem to try to limit global warming to that temperature target. Can you talk a little bit about the relative merits of the two different approaches and how you all as researchers think about them?

Kevin Rennert: That's a great question that comes up a lot in discussions about the social cost of carbon. I'll go back to the definition.

Again, the social cost of carbon is the future damages from an incremental increase in carbon dioxide emissions in a year—or really the avoided damages from a decrease in those emissions. It's telling you the incremental (or additional) benefit of abatement. In the benefit-cost framework, which is used by the federal government, the idea is that you're quantifying the incremental benefits of abatement that are associated with the policy action, and you're comparing them to the incremental costs of that action for abatement. The social cost of carbon is a consistent metric for comparing the benefits to the costs.

This is different from the other approach that you're asking about, in which you're assessing the effective carbon price from a policy that's specifying an environmental target, which could be emissions or temperature. The carbon price that's associated with a policy specifying an environmental target provides a measure of the marginal cost of abatement, not the damages. That's really useful in evaluating policy cost-effectiveness, but it's not an alternative way to value the damages from carbon dioxide emissions.

Both of these approaches give you different information, and both of them are useful. The SCC, though, is the one that's consistent with the framework of analysis that the US government actually uses in its regulatory analysis, so it's appropriate to think about it in that framework.

Daniel Raimi: Thanks to both of you so much for coming on the show today and helping us understand this update to the social cost of carbon. As we often do in our episodes, we’ve really only scratched the surface here. We would encourage folks to read the paper and check out the RFF social cost of carbon event.

Before we close out, let's ask you both the same question we ask all of our guests, which is to recommend something to our listeners that's at the top of your literal or metaphorical reading stack. Kevin, what's at the top of your stack?

Kevin Rennert: I'm going to duck this question because I've been on vacation for the last two weeks, and I've been reading a romance novel that was written by a friend of mine from high school. I'm not going to talk about that here on the podcast.

Daniel Raimi: Yes. I've heard it was a little steamy.

Kevin Rennert: It was.

Daniel Raimi: Maybe that’s best for another conversation. Okay, Kevin pleads the fifth. Brian, what's at the top of your stack?

Brian C. Prest: I just read a great book called The Rule Of Five by Richard Lazarus. This book is the story of Massachusetts v. EPA, which was a landmark 2007 Supreme Court ruling that ultimately led to the Endangerment Finding at EPA, where EPA determined that carbon dioxide was an air pollutant. This triggered a bunch of authority for the EPA to regulate greenhouse gas emissions from cars and power plants and so on.

The book is in some sense a history book and nonfiction, but it's written like a legal thriller: it's gripping and exciting and interesting. It tells the story from a petition to EPA in the early 2000s up through the big Supreme Court case, and it has a lot of interesting legal history and background about the Supreme Court. All of this is the history that settles up to where we are today, with EPA pursuing regulations on greenhouse gas emissions from various sectors.

Daniel Raimi: So it's like a thriller? Like, Anthony Kennedy getting in a car chase or something?

Brian C. Prest: You joke, but Kennedy does play a major role in this, and it sometimes wasn't clear which way he was going to go. Understanding the history just makes for a very interesting story.

Daniel Raimi: That sounds fascinating. Thank you, Brian.

And thank you to both of you, Kevin Rennert and Brian C. Prest from RFF, for coming on the show and helping us understand your team's update to the social cost of carbon.

Kevin Rennert: Thanks so much, Daniel.

Brian C. Prest: It was great to be here.

Daniel Raimi: You've been listening to Resources Radio, a podcast from Resources For The Future, or 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.