Questioning the Promise of Carbon Tax Border Adjustments

Most economists agree that the best way to reduce carbon dioxide emissions that cause global warming is by implementing a carbon tax, and making it more expensive to buy products and services with a high carbon content. Yet by putting a price on carbon, countries may drive up costs for domestic businesses, putting them at a competitive disadvantage to foreign competitors from countries where no carbon price exists.

Two experts in climate law and economics look at the most commonly proposed solution to protect American businesses from the competitive impacts of a carbon tax. The solution, known as a border adjustment, would ensure that American and imported goods are subject to the same carbon price.

The tool seems simple enough, and in fact every carbon tax proposal in Congress this year features a border adjustment. Yet research suggests that the economic protections promised by border adjustments may not be as great as commonly assumed.

David Weisbach is a professor of law at the University of Chicago. Sam Kortum is an economics professor at Yale University. Their work has focused on the role of taxation in addressing climate change, and potential competitive implications of a carbon tax.

Full Transcript

Andy Stone: Welcome to the Energy Policy Now podcast from the Kleinman Center for Energy Policy at the University of Pennsylvania. I’m Andy Stone. Most economists agree that the best way to reduce carbon dioxide emissions that cause global warming is by implementing a carbon tax and making it more expensive to buy products and services with a high carbon content. Yet by putting a price on carbon, countries may drive up costs for domestic businesses, putting them at a competitive disadvantage to foreign competitors from countries where no carbon price exists. On today’s podcast, we’ll be looking at the most commonly proposed solution to protect American businesses from the competitive impacts of a carbon tax. The solution, known as a “border adjustment,” would ensure that American and imported goods are subject to the same carbon price.

The tool seems simple enough, and in fact, every carbon tax proposal in Congress this year features a border adjustment. Yet research suggests that the economic protections promised by border adjustments may not be as great as commonly assumed. Here to talk about border adjustments are David Weisbach, a Professor of Law at the University of Chicago, and Sam Kortum, an Economist at Yale University. The two have focused on the role of taxation in addressing climate change. David and Sam, welcome to the podcast.

David Weisbach: Thanks for having me.

Samuel Kortum: Yes, thanks.

Stone: David, there are more than half a dozen carbon tax bills now making the rounds in Congress. All the bills include border adjustments. What are border adjustments, and what problems are they intended to address?

Weisbach: Well, to understand border adjustments, it’s helpful to back up just a bit to explain how the line carbon tax is imposed, and then show how border adjustments change the carbon tax. So recall that carbon emissions come mostly from burning fossil fuel. And carbon taxes are designed to address fossil fuels — petroleum, coal, and natural gas. The simplest way to tax fossil fuels is just to tax them when they come out of the ground, or somewhere close to that, such as when they’re processed. So you can think of this as a tax, if you will, on domestic extraction. And many carbon tax proposals do exactly that. Other proposals and many existing cap-and-trade systems instead impose a tax when the fossil fuels are burned. So they’re using production.

In either case, fossil fuels or goods that are produced abroad and imported to the U.S. would not bear a tax, and fossil fuels extracted and goods produced here and exported would bear a tax. We can think of border adjustments as kind of reversing that, right? There are taxes on the carbon content of imports, whether it’s fossil fuels or goods. And so when a good is imported, we add the tax, and that means all imports bear the same kind of tax as goods produced here or fossil fuels that are extracted here. Border adjustments also include a rebate of prior taxes paid on exports. So that means if fossil fuels are extracted here or goods are produced here, they have the tax removed when they’re exported, which means when they’re sold abroad, they don’t bear a tax unless the recipient country happens to impose one. If you think this through, what this means is border adjustments have the effect of shifting the tax from extraction or production onto domestic consumption.

Stone: I just wanted to point out here that the border adjustments really apply when one country has a carbon tax, and the countries that it trades with don’t.

Weisbach: Yes, you can do that when both have taxes, in which case there would be bilateral border adjustments. For example, suppose we had border adjustments with Canada. We would remove the tax when we export a good to Canada, and then Canada would impose a tax on its import of that same good. So you can do it either bilaterally, or you can do it with a country that doesn’t have a carbon tax at all. That motivation tends to be in the case where countries don’t have carbon taxes.

But either way, what they do is they shift the tax from reduction or extraction onto domestic consumption, right? Because all goods that are bought by U.S. consumers would bear the tax, regardless of whether they’re produced here or produced abroad. And all goods that are produced here only bear the tax if they have to be consumed here. If they are exported, the tax is removed. So you’ve shifted the tax off of domestic extraction or production and onto domestic consumption. And what that’s designed to do is remove the incentive to shift extraction or production abroad — something known as “leakage.”

Stone: And when you say, “This is shifted into consumers,” with a border adjustment, consumers would not have the option per my understanding to, say, purchase goods produced in a foreign country that may not be subject to a carbon price, and then the consumer avoids that carbon price in that way. The border adjustment again shifts this onto consumers because consumers no longer have a way to escape that. Is that correct?

Weisbach: Right, domestic consumers. If they went to the foreign country and consumed there, obviously they wouldn’t bear the tax. But domestic consumers would always bear the tax, because it wouldn’t matter who produced the good — domestic producers or foreign producers — either way, the goods they buy would bear the tax.

Stone: Now, what goods would be subject to a border adjustment?

Weisbach: So in theory, all goods, but in practice, actually, most proposals only impose a border adjustment on a small fraction of goods, which are goods that have a high energy content and that are exposed to trade, such as steel, aluminum, chemical, paper, cement, and maybe a few others — and of course fossil fuels themselves. In practice, they tend to be relatively narrow, but in theory it’s everything.

Stone: Now, in the United States, we don’t have a carbon tax at this point, but where have border adjustments been tried elsewhere?

Weisbach: Really not very many places. Maybe there are a few very minor exceptions out there. There are serious concerns about whether you can implement them, and whether they’re illegal. And as a result, most places that have carbon taxes or carbon prices, such as cap-and-trade systems, don’t do that. For example, the E.U. cap-and-trade system does not generally have a border adjustment, and instead most countries try to track the vulnerable industries in simpler ways.

Stone: Now we’re going to talk about that legality issue that you mentioned in a few moments, but before we go onto that, I want to bring up the fact that several years ago, you and Sam published research that concluded, a bit counterintuitively, that border adjustments may not provide the economic protection they’re assumed to offer. You just talked about that a little bit. Can you take that a little bit further for me? What exactly does that mean?

Weisbach: Right, so the basic question we wanted to ask is whether border adjustments make the American people overall better off — and not just the industries that are complaining about the carbon tax, but really everybody. And leaving aside the difficulties of implementing border adjustments which we’ll come back to, and leaving aside the question of whether they’re legal, which we’ll also come back to, we found that they probably don’t help U.S. consumers. And the question is why not?

So go back to thinking about border adjustments as shifting the tax from U.S. production or extraction to U.S. consumption, all right? If the tax were on U.S. production, let’s say — so we didn’t have border adjustments — then people all over the world who buy goods made in the U.S. will bear the tax no matter where they live. And that means that some foreign people will buy U.S.-produced goods and bear some of that tax, right? And U.S. consumers would not bear any carbon tax, to the extent that they buy foreign goods, but this is without border adjustments.

Now compare that to a world with border adjustments. What has changed? Now all U.S. consumers bear the tax, no matter what they buy. Only U.S. consumers bear the tax. What the border adjustment has done effectively is shift the tax off of foreign consumers and put it on U.S. consumers. Foreign consumers are really happy about this. They no longer have to bear any tax, since they were buying U.S.-produced goods. And U.S. consumers — well, not so much, right? Now they face a tax on everything they buy, instead of a tax on only part of what they buy. That, of course, might be offset by possible benefits of having certain types of production occur domestically. Border adjustments might prevent shifting of highly mobile, energy-intensive industries abroad. And if you believe this is bad — it might not be — but if you believe it’s bad, then that in effect might offset the costs of border adjustments to U.S. consumers. But overall, we think that it tends to be the case that border adjustments are probably bad for U.S. consumers.

Stone: So Sam, we haven’t forgotten about you here. I was wondering if you could tell us about the model that you used in the research, the model that you used to understand the outcomes from a border adjustment. What were the key assumptions that were put into that model?

Kortum: Well, like many economic models, it’s a very simplified story that we tried to capture in a formal way through the models, so that we could focus on the key issues. So one of the things it simplified is we think of a world with just two countries — the country that’s imposing the tax, and then the other country that’s just carrying on with business as usual.

Another feature of the model is that there are these two layers of trade. There is trade. There is energy being extracted in each of these countries, the taxing country and the other country. That’s traded. Then there are goods being manufactured in each country, and those are being traded. And then ultimately, consumers consume those goods. And so, even though it’s simplified, it does capture the basic issues that we’ve been talking about here, about the different layers of carbon moving through the world economy.

Then within — and I should say one other key assumption that maybe sometimes surprises people who are non-economists is it is a bit more of a long-run model in the sense that unemployment per se is not a feature of the model. There’s always something for people to be employed doing. They can produce services. They can produce energy. They can produce manufactured goods so that arguments about carbon taxes based on employment are not something we can speak to in this analysis. So that’s both a short-coming and a benefit that lets us focus on maybe the more long-run features.

Stone: Let me jump in on that point, if I may, for just a moment. From reading the paper, what I understood was that there may be short-term employment impacts as certain industries can no longer compete because of the high level of carbon content in their products and services. But over time, I think from what I understood, the assumption is that many of those jobs would move to lower carbon areas. And in the longer term, it will be a net zero for employment. Is that generally what you’ve assumed here?

Kortum: Yes, that’s exactly right. So we built in a kind of mobility of labor to the areas where the jobs are. And so within that setting, we can think about taxing carbon at these different levels — at the wellhead, as David was describing. Or we can think of it being taxed at the level of production. Or we can think of it being pushed all the way down to consumption. And we can compare the outcomes from taxing at these different levels.

An important thing to keep in mind when we make that comparison, to make that comparison fair, you need to always hold fixed the global emissions of carbon that are achieved with policies. So that needs to be held fixed, because that’s all that matters in the end. When you have a carbon policy, you shouldn’t really care about anything else other than the total global emissions, in terms of the effectiveness of the policy. So we hold that fixed and then say, “What policy would be able to get us that outcome at the least cost to the economy?” And as David said, a big part of that is least cost to the U.S. consumers.  

Stone: So Sam, what did the model show the impacts to be on industry in the country with the carbon tax?

Kortum: Yes, so it started with an extraction tax. Suppose you did nothing, you just tax at the wellhead, and there are no border adjustments whatsoever. Well, then what you’re going to do is reduce the size of your energy sector. You’re going to export less energy or start importing more energy than before. So that’s one layer.

Now, you could have border adjustments on the energy. So for example, you could have a tax on the imported energy and rebate of the tax when the energy is exported. And so in that case, you’ve moved the tax up to the level of producers. And in that case, what you’re going to find is that you’ve reduced the size of your goods-producing sector, and you start to export less goods and import more goods. At that point, your energy sector is kind of left in the same situation as without a tax, but you’ve caused these shifts in the production sector of the economy. And so finally you could have the other layer of border adjustments on the goods — in other words, a tax on the energy content of the imported goods, and a removal of that tax when those goods are exported. And then you move it all the way up to the consumer paying the tax.

And even though that is now not having any effects on the global production of goods or global production of energy in terms of the taxing country versus the other country, as David pointed out, that may look like a good thing, but that doesn’t necessarily make U.S. consumers better off. And in fact, one of our main points is that’s unlikely to be the optimal policy. In other words, having the full border adjustments, say, push the tax all the way up to the consumer is typically not the best policy. And in fact, it can be the worst among those three that we just discussed.

Stone: That’s a pretty bleak outlook for it. So were there any notable benefits to the home country, the country with the carbon tax?

Kortum: Well, just a second. Let me step back, because there could be huge benefits in having a carbon tax. It’s really an issue of whether you should leave it on the extraction sector, the energy sector, or push it downstream either to the production sector or to the consumers. So when you say it’s a kind of glum scenario, I guess I don’t agree. It’s just saying, “Let’s be smart about the way the taxing is done,” and we shouldn’t have a presumption that the best thing is to always push it down to the level of the U.S. consumers.

Stone: David, another point that the two of you make is that one of the major issues with border adjustments, as well, is that they imply large administrative costs. As we’ve spoken already, these are complex issues, right? And there are a lot of different ways that a border adjustment can be implemented. There are many different ways and levels at which a carbon tax can be implemented. So tell us about the administrative costs. How great are they, and are they greater than the actual economic — any benefits that may come?

Weisbach: Let me just back up, though, to adjust the point that Sam was just making, whether there are any benefits to border adjustments. The benefits that are claimed — and they may be true — is that it avoids the shifting of, say, our fossil fuel extraction sector abroad or shifting of heavy industry abroad. That’s the claim. And the question is how big is that benefit, and whether we can quantify it?

To some extent, that disruption is going to occur, anyway. Carbon-intensive industries are going to have to retool in the long-run. So to the extent that we think that border adjustments are really preventing those industries from shifting out of the U.S., it’s not really clear how much in the long-run you’re going to stop those industries from having to retool, anyway, right? In some ways, it’s a rearguard action, and so again, it’s not clear that consumers are better off — they’re probably worse off with border adjustments. And it’s not clear that border adjustments, by trying to continue to protect these industries from retooling are really doing something that’s good for the U.S. in the long-run.

Anyway, we don’t model that, but the main thing that people argue about is, “Gosh, these heavy industries are going to be hurt. They’re going to have to shift to lower carbon production methods.” And they’re going to have to do that anyway in the long-run. But let me turn to the administrative costs. And this, I think, is really the key. Even if you’re on the fence regarding the economics, you might think that the cost to U.S. consumers is worth it in order to protect certain heavy industry that is still occurring here. Even if you thought that, the real problem with border adjustments is how do you actually implement them? And implementing them on any kind of comprehensive basis, I think, would be almost impossible. The costs would be overwhelming.

So imagine that you’re a customs officer. You’re sitting in a port in Los Angeles, and a cargo ship shows up, and it’s loaded with automobiles. Your job — you’ve got to figure out what the emissions are from the production of each type of car that’s on the ship. And remember, those cars might have been each produced in six different countries, using different methodologies, different energy technologies, different fuel sources, and so forth.

Stone: And when you’re talking about emissions, you’re not talking about emissions out of the tailpipe. You’re talking about emissions that were involved in the actual manufacture of those cars.

Weisbach: Yes, precisely. So you’ve got steel in a car. When you made that steel in, say, South Korea, there were emissions associated with that. And what a comprehensive border adjustment would do would be to impose a tax on all of the emissions that came from producing that car — just as if the car were produced here. If we had a carbon tax here, then any use of fossil fuels in producing that car, including making the steel, would bear a tax. And so when we import that car, we have to impose the same tax as if it were produced here.

There’s simply no way that you can do that with the number of goods and the number of types of goods, the complex production processes around the world — there’s no way that we can impose any reasonably accurate border adjustments on most of the goods that are coming into the U.S. And that’s the reason why most border adjustments are limited to simple raw materials, such as steel or chemicals. But even then, the problem may be insurmountable. There are thousands of different types of imported chemicals. Each one of those chemicals might be imported from different places, made use in different production technologies and different energy sources. At best, the border adjustment proposals we have out there are extremely crude and extremely inaccurate. There’s simply no way to make them accurate.

And when they’re inaccurate and crude, then you introduce all kinds of problems, because they’re not creating the right kinds of incentives that you want. That is, the imaginary border adjustments, where we tax foreign goods the same way we tax U.S. goods is simply not something that anyone thinks we can actually do.

Stone: So David, given all these complexities, and also the fact that again, we have more than a half dozen carbon fee proposals in Congress right now, all of which include border adjustments — the bottom line — Is support for border adjustments justified?

Weisbach: Well, I don’t think so. I think the supporters of border adjustments are the affected industries, right? They are the loudest voices out there, and they have an easy time making their views known in Washington. But nobody is out there speaking for consumers or for the American public generally. Politicians naturally react to the loudest voices that they hear, to the industries that have access to them. They can’t dismiss industry voices altogether, right? They need to put together a coalition to pass a tax, and industries may have the power to block taxes altogether. But we should view the calls for border adjustments by industry as just that. And what politicians need to do is figure out a way of putting the other coalition, of satisfying this industry so that carbon taxes aren’t blocked, however they need to. But we shouldn’t view border adjustments as justified on an economic basis.

Stone: Sam, are there better alternatives to border adjustments? I think you mentioned some alternatives earlier on.

Kortum: Yes, so I go back to this is sort of the question that David and I are asking right now in the research we’re doing. And I might be repeating a bit, but the first alternative would just be to impose a carbon tax at the level of extraction of fossil fuels and just stop there. Now that’s certainly a policy that doesn’t involve border adjustments, and it also doesn’t affect the price that U.S. producers of goods are facing. But that may not be the optimal thing, either.

What we’re tending to find is sort of a mix of partial border adjustments as what is the best thing. For example, David mentioned the difficulties of imposing border adjustments, but that doesn’t apply really to that first layer of the actual energy exports and energy imports. So we can do that pretty easily, but we may not want to do that one for one with whatever our carbon tax is. If we don’t do it one for one, then we leave a bit of the tax burden on the energy sector, and we’re finding that to be typically a good thing to do.

Then if you go downstream to the border adjustment for the goods sector, when you’re talking about exports, you’re allowing the goods producers to get rebated on the energy taxes they paid in the production of goods. You might want to do that a little more in a smarter way, because in a sense their facing that carbon tax is a good thing, because that gets them to produce the exported goods in a more energy-efficient way. That’s kind of the beauty of a carbon tax. Everybody is trying to figure out how to avoid it by the techniques of production that they adopt.

And so you want to leave that incentive there, but you may be able to support your exporters who are getting priced out of the market through other subsidies, based on the number of units they export, rather than the carbon content of what they’re exporting. And that way, you retain the incentives to produce goods in an energy-efficient way, yet protect some of those producers from getting priced out of the market due to your carbon tax policy. So those are some of the things that we’re finding as we try to explore better alternatives to just a simple border adjustment that’s complete.

Stone: David, are we seeing any of these alternatives in the current crop of carbon fee proposals in Congress or elsewhere?

Weisbach: Well, I think what you saw in the E.U. with their cap-and-trade system was something even simpler than what Sam was suggesting, which is they just subsidize the at-risk industries by giving them free permits, or something like that. So they have passed this cap-and-trade system. Industries are worried about what they’re going to do. And so they just basically give them money. We saw that in the Waxman-Markey Bill which made it through the House several years ago, which is they basically subsidized at-risk industries — so it’s something very, very simple.

Stone: Doesn’t that take away some of the incentive, though, to reduce carbon emissions?

Weisbach: It depends on how you do it, but it might. But if you simply gave them money, then it wouldn’t, because they’d still have to bear the tax on any goods they produce. It’s not been an incentive to produce clean goods, even though they now have just extra money. If you lower the tax rate on them or lower their permit requirements in the cap-and-trade system, then it would have that diluting effect.

Stone: Got it. Now David, you touched briefly earlier on the question of whether border adjustments are actually legal under World Trade Organization rules. And the World Trade Organization generally seeks to eliminate barriers to free trade. So bottom line — would border adjustments be legal under the WTO?

Weisbach: Oh, I don’t think we know the bottom line. No one has ever really done this before. The WTO has never really ruled on this, and so there are enormous questions as to whether they’re legal or not. And no one really knows. The legal analysis, then, relies on analogies to other cases which don’t look very much like the current proposals for border adjustments. And so there’s enormous legal uncertainty. And one reason why is because we know that these types of inbound tariffs and outbound subsidies can be used to block trade, right? Or to block [?] domestic industries.

And so distinguishing genuine environmentally-motivated border adjustments from proposals that simply protect your domestic industries is very, very difficult. And that means that the WTO will face difficult questions when this is eventually considered. And so the bottom line on the legal issue is there is a lot of risk that they’re illegal under the WTO, but no one really knows.

Stone: Sam, let me ask you a final question here. Under the Paris Climate Agreement, all countries are supposed to cut their carbon emissions. Given that, is a border adjustment even relevant, or as relevant as it once may be have been?

Kortum: Yes, so let me answer that in theory, and then in practice. So the vision people had, I think, is that the global emissions — it’s a global externality. It doesn’t matter if the U.S. produces the emissions or China produces the emissions. It’s the sum of the two that matters. And so that’s the problem that needs to be solved. And if every country could get together, the optimal thing would be to have a common tax rate, and we’d be done. We could forget about border adjustments. And in fact, that common rate could be taxed at the wellhead, and then there could be some transfers of the tax revenue, to make it fair. We wouldn’t need border adjustments. So that’s kind of the beauty of it.

You know, David described all the difficulties of implementing them. You’d do away with that. Okay, now in practice, there are two issues that come up. One is that the Paris Agreement doesn’t really have teeth, so we’re still in the world where some countries are going to impose much stronger policies than other countries. And many countries will probably have no policy whatsoever. So that’s one issue.

And then the second one is what David touched on earlier, which is even if everybody has a policy, but they’re kind of different, you may need border adjustments to move between one country that’s taxing at the wellhead, and one country that’s taxing when the carbon fuels are combusted, and so on. But I feel like that’s more of a technical issue.

I think the big picture is that if we could somehow, all countries get together, impose a common policy, that’s just by far the best way to solve the issue of climate change. And what we’ve been talking about today is the second-best situations, where only some countries have a policy. And then what are you going to do about it to make that policy not be undercut by the other country? And that’s where border adjustments come in.

Weisbach: Can I add one thing to that? The Paris Agreement contemplates different emissions reductions commitments by different countries, which means different implicit carbon prices. And the question is, imagine three countries — the United States, Canada, and China — and they all have different carbon prices. One might be 0. One might be $25 a ton. And one would be $50 a ton. And then the question is — How do you think about border adjustments in that kind of world? Do we have border adjustments for the $50 country? Only with the $0 country? Or with the $25 country? And how do you coordinate it among multiple different countries when there are different carbon prices? The implementation problem is then starting to get exponentially more difficult.

Stone: David and Sam, thanks for talking.

Weisbach: Thanks very much.

Kortum: Thank you for having us.

Stone: Today’s guests have been David Weisbach of the University of Chicago Law School, and Sam Kortum of the Yale University Department of Economics. Keep up to date with Energy Policy Now by subscribing to the podcast on iTunes or wherever you get your podcasts. And for more energy and environmental policy news, blogs, and research, visit the Kleinman Center for Energy Policy’s website. Our web address is kleinmanenergy.upenn.edu. Thanks for listening to Energy Policy Now, and have a great day.