Do Carbon Taxes Hurt the Economy?, with Gilbert Metcalf

Notable Quotes

• Even a modest carbon price can curb emissions: “What do we need to do to cut emissions by 35 percent? What’s the tax rate that we need to do that? … [The US Department of the Treasury in 2016] found that a carbon tax in the $50 to $70 range would collect a gross revenue of 1.1 to 1.2 percent of GDP. We find that the tax-revenue-to-GDP ratio would be only roughly a fifth of that—about 0.2 percent—which is much lower. That suggests we’re going to need a much lower carbon tax than the $50 to $70 rate that the Treasury study suggested.” (6:59)
• Carbon pricing incentivizes the adoption of cleaner technologies: “If you think about investing in clean technologies, these are obviously technologies that have a higher up-front cost—otherwise, firms would have been investing in them, anyway. Carbon pricing incentivizes firms to make these investments in the [more costly] but lower-polluting technologies. As that carbon price increases over time, it increases the share of firms for whom it now becomes profitable to make this investment in clean technology.” (10:41)
• Policy uncertainty can disrupt decarbonization efforts: “Policy risk is huge, both for the business community and for the environment. Businesses are much less likely to make these expensive, long-lived capital investments if they risk facing policy changes that lower or eliminate a price on carbon pollution. That means we may not get the kinds of reductions that we need … I think it’s really critical to try to build as much policy certainty into whatever policies we end up enacting in Washington.” (18:28)

Top of the Stack

• “The Macroeconomic Effects of a Carbon Tax to Meet the US Paris Agreement Target: The Role of Firm Creation and Technology Adoption” by Alan Finkelstein Shapiro and Gilbert Metcalf
• The Power Broker: Robert Moses and the Fall of New York by Robert A. Caro
• Chicago Architecture Foundation Center River Cruise

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 Gilbert Metcalf, John DiBiaggio Professor of Citizenship and Public Service at Tufts University. With his coauthor, Alan Finkelstein Shapiro, Gib recently published a working paper on how a carbon tax designed to meet the US's climate target under the Paris Agreement would affect the US economy. They estimate the effects not just on overall GDP, but also on employment, labor force participation, wages, and companies’ decisions about investing in clean energy technologies. We discuss their modeling efforts, the surprising results and the implications for policymakers in today's podcast. Stay with us.

Okay, Gib Metcalf from Tufts University, also a university fellow at RFF. Welcome back to Resources Radio.

Gilbert Metcalf: Thank you Daniel. It's a pleasure to be here again.

Daniel Raimi: So Gib, we've had you on the show before, but it's been a little while. So can you remind our listeners how you got started working on environmental issues?

Gilbert Metcalf: I got started after I graduated from college in the mid-’70s. In fact, it was mainly through anti-nuclear organizing in western Massachusetts. I was part of the Clamshell Alliance occupation of the Seabrook Nuclear Power Plant site in 1977 and spent two weeks in a New Hampshire armory, along with a thousand other close friends and neighbors. That really got me thinking about issues of demand and really the importance of economics in demand forecasting and policy development, which all led me to enroll in a masters program in environmental and resource economics at the University of Massachusetts. That's really where I got my start.

Daniel Raimi: That's great. The Clamshell Alliance, where does the name come from?

Gilbert Metcalf: Well, the Seabrook Nuclear Power Plant is on the New Hampshire coast, and there are a lot of clam diggers in the area, and it was a name that resonated with the local group that started it.

Daniel Raimi: Very cool. We're not going to talk more about clams, sorry to disappoint you listeners. We are going to instead talk about a really fascinating new working paper that you recently published with Alan Finkelstein Shapiro called “The Macroeconomic Effects of a Carbon Tax to Meet the U.S. Paris Agreement Target.” So we're going to dig into that today. We'll have a link to the working paper in the show notes so people can read along if they'd like as we talk. But can you give us a start by helping us understand how this paper fits into the broader research landscape and how it contributes to the pretty substantial body of literature that's out there on the economic effects of carbon pricing?

Gilbert Metcalf: Sure. Alan and I, as we looked at this literature, noticed that much of it was computable general equilibrium modeling, or CGE models. We were really interested in understanding the employment impacts of carbon pricing, but these models—these big, large-scale computer models—generally speaking are full employment models. So they can model shifts in employment between sectors, but they can't capture overall employment effects. To do that, we added standard labor market frictions that can lead to unemployment: things like search costs and the costs of posting vacancies. This is something that my coauthor, Alan Finkelstein Shapiro, had studied in great, great detail. So he's a real expert in that area so it was a great match to write this paper with him.

Now, we also wanted to capture the dynamics of firm entry and exit as well as policy-induced technology adoption, because we felt like these were going to be critical features in the model. In fact, it turns out that they make the difference between carbon pricing leading to lower gross domestic product (GDP) growth, versus pricing having a zero or even a modestly positive impact on GDP growth. This sort of provides the theoretical underpinnings for the results that I found in another paper—an empirical paper I wrote with my colleague Jim Stock from Harvard—where we looked at carbon taxes in European countries and found absolutely zero negative impact on employment or GDP growth. So, my paper with Alan really helps to understand what's going on with the data.

Daniel Raimi: That's great. We're going to break down each component of that analysis in the next couple of minutes, so people can really dig into both the top line findings on GDP, and also findings on these specific issues related to labor markets and businesses. But before we do that, I'd like to ask you about just one specific issue in the title. The title refers to the US Paris Agreement target, and I imagine depending on when the paper was written, you might imagine one target or another target. So can you just help us understand what level of emissions reductions you're estimating here or that you're modeling out? And then what is the level of carbon tax that the model finds is needed to achieve that target?

Gilbert Metcalf: Let's talk about the target first. The Biden administration, in their recent commitment under the Paris Agreement—the 2015 agreement to reduce greenhouse gas emissions globally—committed to reducing US emissions by 50 to 52 percent relative to 2005, by the end of this decade, by 2030. Given progress that we've made in reducing our emissions since 2005, along with Energy Information Administration projections of emissions under a business-as-usual scenario for the rest of this decade, this translates into a need to cut emissions by 35 percent by 2030. So that's the target that we're looking at.

What do we need to do to cut emissions by 35 percent? What's the tax rate that we need to do that? Our model is really not constructed in a way that allows us to spit out a tax rate in dollars per ton. A way to answer your question is to look at the carbon tax revenue collected relative to GDP, which our model does report. If you look at a 2016 US Department of the Treasury analysis that was done at the end of the Obama administration, they found that a carbon tax in the $50 to $70 range would collect a gross revenue of 1.1 to 1.2 percent of GDP. So a little over 1 percent. We find that the tax-revenue-to-GDP ratio would be only roughly a fifth of that, about 0.2 percent, which is much lower. So that suggests we're going to need a much lower carbon tax than the $50 to $70 rate that the Treasury study suggested.

We bench tested that estimate by tweaking our model. We find if we turn off the ability for firms to enter and remove the technology adoption channels we include in our paper, then, in fact, we find that the revenue-to-GDP ratio jumps to about 1.4 percent, which is much, much closer to the Treasury estimate. So the bottom line is that, based on our modeling, much lower tax rates will be needed to hit the Paris Agreement target than models suggest when they don't allow for firm entry and technology adoption.

Daniel Raimi: So those dynamics around firm entry and technology adoption are really, really important, which is one of the things that really stood out to me in this paper. I'm going to ask you about that next, but first I want to ask about the top line numbers, those headline numbers that people often look at when economic effects of carbon pricing analyses come out. In terms of macroeconomic outcomes like GDP, as well as employment impacts, what are the top line findings from your analysis?

Gilbert Metcalf: So our top line findings on GDP is that it actually increases very modestly in the baseline model. However, if we turn off the firm entry and technology adoption channels, then it actually falls by about half a percentage point. Employment is interesting. Our findings are very similar to those from work that Marc Hafstead and Rob Williams from RFF found in a paper that they published in 2018. We find that unemployment increases a little bit, but what's interesting from our paper is that labor force participation is still increasing. Part of this just has to do with the reallocation of job creation towards the sector using green technologies. With an increase of people entering the labor force, then not surprisingly that can put upwards pressure on unemployment.

That's sort of the top line on unemployment. As do Hafstead and Williams, we also see big sectoral shifts, with employment falling significantly in the carbon-intensive sectors and rising in green sectors. So it's that sectoral shift that we always think about when we think about green jobs and the transition to a zero carbon economy.

Daniel Raimi: Now let's get to these issues of firm entry and technology adoption, which are clearly playing a really big role here. Can you just talk a little bit about how the carbon price affects firm entry or business formation, and in particular, how it affects a business’s decision about adopting or investing in clean energy technologies and what that all means for these top line results?

Gilbert Metcalf: If you think about investing in clean technologies, these are obviously technologies that have a higher up-front cost; otherwise firms would have been investing in them anyway. Carbon pricing incentivizes firms to make these investments in the higher cost but lower-polluting technologies. As that carbon price increases over time, it increases the share of firms for whom it now becomes profitable to make this investment in clean technology. We don't actually find much of an impact of the carbon price itself on overall business formation, but it clearly affects what kind of technology you're going to adopt when you do decide to enter an industry. You have to choose either a polluting or a non-polluting technology, and as the price goes up, then there’s clearly a greater incentive to choose that cleaner technology. Intuitively, what's going on is that these channels of business formation and technology adoption simply provide greater flexibility to the economy over time to adjust to higher carbon prices.

Daniel Raimi: To get a little more intuition for what's going on here, I think it might be helpful if you could help us understand what you mean when you say clean technologies or green technologies. Can you give us an example, maybe, of a hypothetical business and how it behaves in your model versus how it might behave in a model that does not have this kind of technological adoption built in?

Gilbert Metcalf: So in our model, we have firms that are deciding whether or not it's profitable to enter a market. Part of that calculation includes decisions they're going to make on the kind of machinery to put into their factory, the kind of boilers to put into their factory. Just think of it that way. So a firm decides to enter and makes a calculation that given the carbon price, and given all the other prices in the economy, that it makes sense—for example, if it's a trucking enterprise—to put in a fleet of all electric vehicles versus diesel vehicles. There’s going to be a higher up-front cost, but assuming that the electricity they're getting is going to be cleaner than what you get from burning diesel, then you have a cleaner technology or clean technology, assuming zero carbon electricity. So that'd be the kind of decision that a firm is thinking about making. Alternatively, you could think of an aluminum smelter that has access to hydroelectric electricity, and therefore puts in an electric smelter as opposed to a coal-fired smelter.

Daniel Raimi: Then just one more follow-up question on that. Can you help us understand how making those decisions ultimately leads to a higher top line GDP relative to a world in which the business is invested in, let's say the diesel trucks or the fossil fuel–powered smelter?

Gilbert Metcalf: It just gives you an additional margin of adjustment to help us hit lower emission targets, since the question is “how do we reduce emissions?” Well, we can reduce emissions by either switching to clean technology, which is one of the channels we're allowing for. But, if we're not allowing for that channel, then the only way to reduce emissions is to reduce demand for the goods and services that are being produced with that polluting technology. That means that more of the onus is placed on consumers to be choosing products that are low carbon, as opposed on supplies to produce the same product in a lower carbon way. So with additional channels to operate on, you don't need as high a carbon tax rate to hit a target—35 percent reduction—and that means we're putting less drag on the economy.

Daniel Raimi: That makes a lot of sense. I know you look at these effects over time. Obviously you're only looking up to 2030, but can you talk us through a little bit about what you find in the early stages of the modeling versus the later stages?

Gilbert Metcalf: We assume we hit a steady state in 5 years or 20 quarters. We're finding that as you slowly ramp up the carbon price—we ramp it up over time in the model—that we move in the direction of slowly beginning to see workers shifting away from working in dirty technologies to clean technologies, among other things, like more businesses when they are forming, deciding to form using clean technology, and so forth. So the steady state results really tell the same story as the transitional results.

I think the interesting thing that comes out of both looking at the transition as well as the steady state is that unemployment really hits the long-term higher level—meaning that we have higher unemployment and higher labor force participation—pretty quickly in the transition. But two things to that. One is it's not a very big increase in unemployment. Secondly, we find that real wages are higher across the board for all workers. Of course that's not surprising, because what's driving the increased labor force participation is the attraction of higher wages. That's an important point for policymakers, since obviously higher wages help with the cost of transitioning from one sector to another. That sort of takes some of the sting out of making that change.

Daniel Raimi: The next question I wanted to ask you is a question that it's applicable for pretty much any study that's trying to model the future, but it's a question about certainty and how the model treats certainty for businesses. In this analysis, and please correct me if I'm wrong, I think the assumption is that the carbon tax policy is durable and that businesses can plan around it. They can plan for the future by buying those electric trucks or investing in that clean smelter. But of course, in our real world and current political environment, a predictable carbon price is not something that's really on the table. And even if a carbon price were on the table, there might be considerable uncertainty about how long it sticks around and how it might get changed in the future. So how do you think your findings might change if we started to account for those uncertainties associated with the real world of policymaking?

Gilbert Metcalf: Yeah, that's a really, really important point and you're absolutely correct that our paper does not allow for any uncertainty. The policy is put in place, and it's going to be there, and businesses can depend on it being there. And it's also certainly the case that in the real world, policy uncertainty is a major issue, as we've just seen over the past four years. Policy risk is huge, both for the business community and for the environment. Businesses are much less likely to make these expensive, long-lived capital investments if they risk facing policy changes that lower or eliminate a price on carbon pollution. That means we may not get the kinds of reductions that we need in order to hit the targets that we want to hit for emission reductions. I think it's really critical to try to build as much policy certainty into whatever policies we end up enacting in Washington.

One of the great debates is whether we have greater policy certainty by putting a carbon tax in place that raises revenue and provides a constituency for that revenue, or whether it is better to use the regulatory approach because regulation is kind of sticky: once it's in place, it's hard to undo. I'm not an expert in that area so I don't know the answer, but it's absolutely the case that this uncertainty is a big deal. There's also another level to it. Policy uncertainty is one thing, but there's also technology uncertainty. We could be thinking about investing in particular types of, say, wind turbine technologies that we need to pay off over a 20 year period to recoup our investment in, but if a new technology comes along that's cheaper, more efficient, or better, it could end up making my technology obsolete. That's just a whole other level of uncertainty that kind of makes this a very, very complicated problem.

Daniel Raimi: Do you think it's fair to say that if we were to incorporate policy uncertainty into the model that we would end up with less rosy outcomes, like worst impacts on GDP? Or is it just kind of not possible to answer that question with the current framework?

Gilbert Metcalf: It really depends on the kind of uncertainty you're thinking about. I wrote a paper almost 20 years ago in the Economic Journal with Kevin Hassett, and Kevin and I were looking at investment tax credits. This wasn't energy-specific, but it was about investment tax credits in the federal income tax. This is a policy that has bounced on and off over the years from its initial inception in the early 60’s. We used a stochastic process as sort of a random walk in continuous time, something called the geometric Brownian motion. And what we found is that when the policy flips off and away from the investment tax credit, obviously investment goes down. But when it flips back on you see a big rush to invest because you want to take advantage of that tax credit when it's there.

It turned out that in the aggregate, you could find overall investment increasing or decreasing. It really just depended on the type of uncertainty. The same thing could be true here. Depending on the kind of policy risk, you could see a flood of investment occurring during the good times, and that could outweigh the dampening in investment in the bad times, but it could also go the other way. So I'm skeptical that a model would give us a definitive answer, because I think it would really depend on how you model that kind of uncertainty.

Daniel Raimi: So Gib, we've touched on I think some of the central points of the paper, but we're really only scratching the surface. So I just wanted to kind of have this placeholder question, where I could ask you to talk about any other parts of the paper that you think are important or interesting or counterintuitive. Is there anything else you want to highlight?

Gilbert Metcalf: Well, one thing I've already highlighted a little bit, but is so striking that it really bears just emphasizing, is that it was very surprising to how much lower the tax rate collection was in our model than in the model where we turn off the other channels. I find that outcome very encouraging for the kind of policies we're looking at, where we're setting an emissions reduction target and hoping to use carbon pricing—whether implicit pricing through regulation or explicit pricing through some sort of pricing mechanism carbon fee, or a cap and trade—I find that incredibly encouraging and I think that's something we need to dig into more. So that's one thing that I want to really emphasize.

The second point I'll make is less about the specifics of this paper, but I really enjoyed writing this paper with Alan Finkelstein Shapiro, who's a macroeconomist who studies macro-labor markets. It's my second paper at this intersection, having done an empirical paper with Jim Stock, and I think this intersection of environmental economics and macroeconomics is a really, really important one, given the focus of policymakers on what the macro-impacts of environmental policy might be. I just want to say how cool it is to be doing this work and to be seeing others working at this intersection, including people like Lint Barrage at UC Santa Barbara and Garth Heutel at Georgia State University, and of course Marc Hafstead and others at RFF. I think it's an intersection of two fields that haven’t yet been as well populated as it could or should be.

Daniel Raimi: Yeah, that's such a great point. I don't do this type of modeling, but in the work that I do, these questions about employment impacts in particular are so prominent from the policymaking community. Oftentimes we just don't have great tools to answer those questions, so it's really fantastic that you and Marc and others are really stepping in to fill that gap.

Gilbert Metcalf: Well, I think we have the tools, but what we haven't done is use them as well as we might, and so it's great to see people doing that.

Daniel Raimi: Well, Gib Metcalf from Tufts, thanks again for joining us. Let's go now to our last question, which is our Top of the Stack question. So, asking you to recommend something that you've read or watched or heard related to the environment, even if it's sort of just tangentially related to the environment, that you think people would enjoy. I'll start with an experience that I had over the weekend. I took my first post-pandemic vacation to Chicago and it was wonderful, and one of the things that I did was go on an Architecture Foundation riverboat tour. If you've never done this in Chicago, it has this great river in the middle of the city and it has these different branches, and you can take these boat tours that take you all along the river, and you learn about the history of Chicago and about the buildings, but you also learn about the reversal of the Chicago River.

You learn about the pollution of the Chicago River in the sort of early industrial era, and hear some real kind of bone-chilling stories about the amount of animal carcasses and other waste that was dumped into the river early in the city's history. So it's a really fantastic experience and you learn about architecture and history and environmental issues too. So if you're ever in Chicago and you've got some free time, definitely go on a boat tour. But how about you Gib, what's on top of your stack?

Gilbert Metcalf: Well, that sounds fascinating. Well, on my stack, I'm currently reading Robert Caro's biography of Robert Moses called The Power Broker, which is actually an old book published in the ’70s. Robert Moses singlehandedly created the New York State park system, one of the largest and best state park systems in the country. It really transformed recreational opportunities for New York state residents, particularly for low-income New York City residents who could now get to beaches on Long Island, like Jones Beach and other beaches they couldn't access before. To see how he did that is quite remarkable. Moses is also responsible for contributing to a car culture in the New York metropolitan area that bedevils drivers to this day. For example, he refused to allow public transit options in order to access the popular beaches on Long Island that he had built. I think this is a terrific read. It's spellbinding. Caro is a terrific writer and it's a masterful meditation on the uses and misuses of power for civic and environmental improvement.

Daniel Raimi: It's such a great book. I remember reading it quite a long time ago, and if I remember correctly, there are quite a few issues about the siting of infrastructure, which has ended up having really big impacts on environmental justice issues with freeways being placed in certain neighborhoods and displacement of certain communities. I can't remember, is that covered in detail in the book too?

Gilbert Metcalf: He's got a chapter called “One Mile,” which talks about the destruction of a mile-long segment of a highway in order to build this highway that destroys an entire community of lower and middle income New Yorkers in the Bronx area. So yes. lt also touches on him destroying farms out in Long Island. So yes, there are huge environmental justice issues that Caro focuses on sharply in the book.

Daniel Raimi: Fascinating. Well, we'll have links to both the boat tour and the book, and of course the paper in the show notes for all of you to check them out. And just say thank you one more time to Gib for coming on the show and teaching us about this really interesting new work. Thanks so much Gib.

Gilbert Metcalf: Well thanks Daniel, it's been a pleasure.

Daniel Raimi: You've been listening to Resources Radio. Learn how to support Resources for the Future at rff.org/support. 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. Resources Radio is a podcast from Resources for the Future. 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.