To cap or to tax? That is the question, at least for U.S. policy makers intent on reducing the nation's carbon footprint by making emissions costlier. There are impassioned advocates for both systems, but upon closer inspection, cap-and-trade schemes and carbon taxes can start to look like two sides of the same coin. Our experts discuss the benefits and drawbacks of each, and pinpoint the real choices policy makers need to make if they want the market to get to work on global warming.
In today's climate change debate, carbon taxes lurk as a sleeper option compared to the cap-and-trade legislation pending in Congress. Carbon taxes aren't without their downsides, but they should receive more consideration despite the almost Pavlovian aversion to taxes.
Carbon taxes are fairly laser-like in the way that they target carbon dioxide, which comprises 85 percent of U.S. greenhouse gas emissions. Carbon taxes are based on the carbon content of fossil fuel, and most fuel that's consumed emits carbon dioxide in direct proportion to its carbon content. Because of this correlation, taxing carbon can serve as an easy surrogate for taxing carbon dioxide emissions, avoiding the need for end-of-tailpipe or end-of-smokestack monitoring technology. Refunds or exemptions can provide relief in situations where the carbon is consumed without producing emissions (such as when natural gas is used as a feedstock in making petrochemicals) or when carbon dioxide is sequestered.
Cap-and-trade regimes, which distribute a limited number of emissions allowances, also target carbon dioxide, but carbon taxes are simpler. As with a gas tax, a carbon tax can be imposed early in the production or distribution cycle of fossil fuels. Congress just sets the tax, and the IRS collects it. With a cap-and-trade system, Congress needs to create a new legal commodity--a fixed number of allowances to emit carbon dioxide. It then needs to design a system for distributing those allowances among regulated entities, rules for transferring and tracking them in the marketplace when emitters buy or sell allowances, and procedures for ensuring that emissions match the allowances a regulated entity has in hand. Cap-and-trade systems may be "market-based," but they must first create a new product and market; carbon taxes add one new component to a traditional, well-established system.
The straightforwardness of carbon taxes makes them economically efficient, as the Congressional Budget Office has recognized. The cost of the carbon component of any fossil fuel--set by the tax rate--is known and predictable, so users can plan accordingly, whereas tradable allowances have fluctuating prices. Cap-and-trade proposals can build in features that limit the price exposure and allow flexibility in annual compliance, but add more layers of complexity.
Why not just keep it simple and impose a tax? A classic and serious argument is that the "cap" will produce a known limit on the quantity of emissions, whereas the tax would allow emitters to decide whether to emit more (or less) based on what is financially rational. But achieving a legitimate ceiling on emissions through a cap-and-trade regime will require accurate monitoring and verification support. Providing allowances for offset projects, which also must be verified, ups the ante even more.
Congress and a new president might be skittish about enacting a carbon tax high enough to drive down emissions. But a cap-and-trade system will also impose new--and uncertain costs--on society, particularly if allowances are auctioned rather than distributed without charge (a hotly debated topic that we will no doubt pursue as this conversation continues). The difference is that the costs of a cap-and-trade system are more politically opaque. Will voters really understand how cap and trade will work and how it will affect their bottom line? The debate about cap-and-trade needs to fully expose the costs to consumers; voters should have truth in lending.
In short, carbon taxes offer cost certainty and simplicity. In 10 years, we may have questions about whether the IRS is auditing enough carbon tax returns, but we won't be wondering about whether middlemen are making too much money from allowance trading, if investors are manipulating the carbon market, or what new laws we need to guard against those risks.
Nevertheless, it's not just about efficiency and simplicity; a broader, related principle underlies carbon taxes. As former Vice President Al Gore said last month when he reiterated his support for carbon taxes, "we could and should speed up this transition [to renewable energy] by insisting that the price of carbon-based energy include the costs of the environmental damage it causes." Cost-internalization is a fundamental plank in the platform of market-based instruments, and it isn't limited to carbon taxes. Carbon taxes could lead to other ways to build environmental costs into prices. What about a climate change tax on other greenhouse gases such as hydrofluorocarbons, or taxes on nuclear power or hydropower that recognize their non-carbon-related environmental impacts?
Congress has two carbon tax bills before it (the Save Our Climate Act of 2007 and the America's Energy Security Trust Fund Act of 2007). It should look closely at the concept of carbon taxes--and beyond.
Cap and trade sometimes seems like the Rodney Dangerfield of environmental policies: It's popular with the masses (or in this case both presidential candidates, members of Congress, major environmental advocacy groups, and the private sector) but gets no respect from most academic economists and other high-minded policy wonks. That's true despite the fact that cap and trade is a proven policy: The emissions trading program for sulfur dioxide from electric power plants, established by the 1990 amendments to the Clean Air Act, has cut emissions by 40 percent since 1990 at a fraction of the expected cost.
A cap-and-trade system differs from a tax in two basic ways. The essential distinction is that a tax fixes the price of pollution, while a cap controls the quantity. From the perspective of the planet, this makes all the difference. Based on what the science tells us, we have to reduce greenhouse gas emissions dramatically in the next several decades--by roughly 80 percent in the United States relative to current levels. The only way to guarantee those cuts is to put a cap on emissions.
A tax only achieves emissions reductions indirectly, through the price. The problem is that we don't really know the appropriate level of the tax that will induce the changes in energy demand and consumption that are needed to bring emissions down far enough and fast enough to avoid dangerous climate change. And we don't have the time (or the political attention span) to try out different taxes in a search for the "right" level. Moreover, as the economy grows, so will emissions, meaning that the tax would need to be raised over time to ensure that emissions continue to fall, as they must. Finally, political pressures will be omnipresent to set the tax low to begin with and to lower it thereafter; just witness the recent calls for a gas tax holiday. All of these factors suggest that we neither know what the right tax is, nor have the political will to set it at that level.
In contrast, a cap controls quantity directly. Congress, not the market, would determine the allowable quantity of emissions--while the market, not Congress, would determine the price. In this way, science can determine the policy goal--while the market figures out the cheapest and fastest way to achieve it. That seems like the right division of labor.
Setting a cap on emissions is especially important given the growing body of scientific evidence suggesting that climate change is likely to be abrupt and nonlinear, characterized by "tipping points" such as the melting of the Greenland or West Antarctic Ice Sheet, which would slowly but inexorably raise sea level by several meters. A hard cap is the best way to ensure that we cut greenhouse gas emissions enough to avoid exceeding dangerous thresholds in the climate system.
A second (and less fundamental) difference between a tax and cap and trade concerns whether firms pay for every ton of pollution they emit, or receive some tons for free. A carbon tax would likely be applied to all emissions. In contrast, under a cap-and-trade system, once the cap on allowable emissions is set, government could give away some portion of the allowances for free to regulated firms--without affecting the efficiency of the program in terms of achieving emissions reduction. This strikes some people as an unfair break for polluters. But it does offer an important political advantage: Politicians can distribute allowances as they see fit, without affecting the basic performance of the program.
Now, it's worth pointing out that raising revenue does matter for efficiency in the economy at large. That's because government can use the revenue to cut income taxes, which represent a drag on the economy. Moreover, if those tax cuts are targeted wisely, they can also help ensure distributional equity, offsetting the impact of higher energy costs on low-income households.
But if raising revenue is our goal, a cap-and-trade system can be designed to do it just as well as a tax can. Congress could decide to auction off most or all of the allowances, rather than giving them away for free. Whether we use a tax or cap and trade, we will have to negotiate how to divide up the "pie" created by placing a value on the right to pollute. Will we allocate those allowances to polluters or to consumers at large? The key point is that a cap-and-trade system doesn't require doing one thing or the other. Rather, it provides the political flexibility to make those divisions while still accomplishing the ultimate goal, which is reducing greenhouse gas emissions.
I generally agree with Janet Milne that a carbon tax--or even better, a tax on all greenhouse gases--is preferable to a cap-and-trade system. However, if we follow good design principles, either will be effective. The importance of good design outweighs the choice of instruments: A well-designed cap-and-trade system and a well-designed tax look and operate similarly, while a badly designed tax or permit system is a bad idea.
For example, take the issue of where in the production process the regime is imposed: It could be upstream when the fuel is extracted or imported, downstream when carbon dioxide is emitted, or somewhere in between. If it's substantially upstream, we can cover almost all fossil-fuel emissions by imposing the regime on fewer than 3,000 entities: the 149 petroleum refiners, 1,438 coal mines, and 530 natural gas processors in the United States, plus imports at a few locations. Downstream, there are almost 250 million automobiles and millions of heating units throughout the country. Upstream, we could capture a broad base of emissions cheaply. Downstream, we would capture a narrow base expensively. Either carbon taxes or cap-and-trade regimes can make the same choice in this regard, and a bad choice is likely to far outweigh any differences between the systems.
Similarly, taxes raise money, which is nice because other taxes can be lowered or our debt paid down. But if the permits were auctioned, a cap-and-trade regime would raise the same amount. Alternatively, if permits were to be distributed freely to favored industries, the tax system equivalent would be to exempt those industries from an equal amount of tax liability. Either way, it's the policy choice--in the permits world, between auctioning and giving away permits, and in the tax context, taxing and grandfathering existing facilities--that's paramount, not the choice between taxes and permits.
A final example concerns whether we cover other greenhouse gases under a cap-and-trade or tax regime. As Janet pointed out, emissions from fossil fuels are easy to price and cover because of the simple correlation between the input--the fuel--and the output--the carbon dioxide. With other greenhouse gases it's not as simple to impose a cost. For instance, quantifying emissions from farming (where tillage and fertilizer use are factors) or forestry depends upon the local ecology and how exactly it's being manipulated. Nevertheless, it seems clear that some additional emissions--i.e., potent methane from livestock operations or substitutes for ozone-depleting chemicals--should be targeted in any system. Again, both cap-and-trade and a tax would have to find ways to deal with this complexity, and one isn't particularly better suited to it than the other.
Debates about a tax versus a cap-and-trade system are often about flexibility. Tax advocates claim that flexibility is a central virtue of a tax--the market decides which emissions are reduced and when. Cap-and-trade advocates claim inflexibility as a central virtue--by setting a cap, we'll know the total emissions.
I tend to agree with the tax advocates on this issue, but again, perhaps too much is made of it. In both systems, we should expect to see adjustments to the system over time. If, for example, a tax isn't reducing emissions quickly enough, we might expect to see the rate go up; therefore, the lack of an emissions target in a tax system may be less of an issue than it appears. Similarly, a cap-and-trade system can have all sorts of mechanisms to allow flexibility on the total amount--from banking and borrowing to safety valves and price floors. Cap-and-trade systems will not, in reality, provide a fixed target. The underlying policy choices about targets and flexibility are more important than whether they're implemented through a tax system or a permit system.
The place where one system would be preferable over the other, and where I think a tax dominates, is when we consider implementing the system internationally. Greenhouse gas controls must be implemented internationally: No one nation can do much on its own to slow climate change. The question for cap-and-trade advocates is how is this going to work internationally? Subsidies such as Kyoto Protocol's clean development mechanism, which allows polluters to offset emissions by investing in clean technology development in other parts of the world, have failed. The alternative is allocating permits to many different countries and trying to monitor compliance, but it seems infeasible that we would grant permits to countries that we would be unwilling to give similar cash aid. Taxes are simpler: We don't have to figure out a way to distribute valuable property rights on an international scale.
David Weisbach is absolutely correct: The specific design of either a tax or cap-and-trade program is far more important than the issue of which approach Congress adopts. Either approach, properly developed, can help reduce greenhouse gas emissions. And conversely, either approach, poorly designed, can lead to real trouble.
At one level, there is less difference between the two approaches than it might seem. Janet Milne cites the fact that a carbon tax can be imposed early in the production or distribution cycle of fossil fuels. But as David points out, the same is true of a cap-and-trade program, where an active market could develop among the upstream firms that introduce carbon into the economy. Similarly, contrary to what Janet infers, both systems will require "accurate monitoring and verification," which is a good reason to target the system upstream where there are better records of carbon flows and fewer parties than downstream.
Janet also says that a simple tax on carbon would be more transparent than a cap-and-trade program, but that's only true if you assume Congress wouldn't load a carbon-tax scheme with all of the exceptions and favorable treatments it has included in so many other programs, including both tax systems and environmental programs.
Gernot Wagner and Nathaniel Keohane point out that from a political perspective the genius of a cap-and-trade program is the ability to dole out allowances to regulated firms, presumably to achieve support for the program. But similar favorable treatment could be built into a tax program; firms could be granted tax exemptions for a portion of their emissions without affecting their marginal costs.
This isn't to say that the approaches are the same. At its core, the tax approach protects us from runaway economic costs. That's important if you believe that the costs of emissions reduction could become steep. In contrast, the cap-and-trade approach gives us more certain protection from environmental damage, which is important if you think that environmental risks are daunting and costs of abatement won't rise rapidly.
But there are also political and institutional factors that would suggest that, at least in the case of the United States, a cap-and-trade approach might be superior to a tax approach. One is raised by Gernot and Nathaniel: In the presence of uncertainty about the costs of emissions reduction, and particularly in a growing economy, we'll have a difficult time maintaining a tax aimed at any particular emissions level.
Janet suggests that a market in allowances may exhibit price volatility, making it difficult to plan investments. Certainly that has been the case in the European Trading System and in recent years, in the sulfur-dioxide emissions allowance market. But this volatility isn't so different than that observed in many commodity markets, and businesses manage to cope. The more severe risk isn't market uncertainty, but political uncertainty. Congress has shown an inclination to allow quantity controls to remain in place over long periods of time. It hasn't demonstrated a similar discipline with respect to its treatment of the tax code. This difference in the regulatory stability of the two approaches is particularly important when the government is trying to induce capital investments in a low-carbon economy.
Institutionally, Congress might do better to adopt a cap-and-trade approach simply because other countries and regions are doing the same thing. To allow the integration of the U.S. system with the rest of the world, and to take advantage of the potential gains in trade, it will be necessary to have a similar system.
Finally, although it offends me as a policy analyst to admit this, perhaps the best reason to adopt a cap-and-trade program is because it isn't called a tax. While a well-designed cap-and-trade program would have many of the same revenue-raising features as a carbon tax, the fact is that the U.S. public may not be willing to swallow a pill with the tax label on it.
We seem to have fairly broad agreement that carbon taxes and cap-and-trade programs share similarities, and as David Weisbach and Ken Richards suggest, each is worth its salt only if it's well designed. And I assume we agree that price matters. Just through the market-driven increases in gas prices, Americans traveled 12.2 billion fewer vehicle miles in June than 12 months earlier, a 4.7 percent drop. As others have stated, a fundamental difference between carbon taxes and tradable allowances revolves around certainty of price versus certainty of quantity. So we largely agree on a number of points.
But the two approaches are also fundamentally different in the characters involved--a difference that's important as we evaluate the pros and cons. The carbon-tax mechanism involves two parties--the entities charged with paying the tax based on the carbon content of its fossil fuel holdings and the IRS. The taxpayers may pass the cost of the tax on down the chain of purchasers, and the behavioral effect of the tax will depend on how the price signals influence decisions by the taxpayers and subsequent purchasers, but the mechanism itself is implemented directly just between taxpayers and the IRS.
The cap-and-trade approach incorporates the notion that allowances (whether allocated for free or by auction) will circulate in the economy until they settle out in the hands of those who can use them more efficiently. This involves multi-party relationships in the direct implementation.
The bill debated in the Senate in June requires the EPA to allocate the allowances, in part through quarterly public auctions and in part through distribution at no cost to manufacturers, states, and other entities, and all the allowances can then be traded with anyone who is interested. It's an open, economy-wide trading system for a new commodity created by government. The circle of direct participants isn't limited to the covered industries; it involves traders and the general public; and in the end the costs (as with a tax) may be passed along to consumers.
The proposal also involves the creation of new entities such as a presidentially appointed Carbon Market Efficiency Board to monitor the market and approve "off-ramps" that can offer cost relief, an International Climate Change Commission to deal with allowances for imports, and a government Working Group to recommend protections against fraud and market manipulation.
Ken has suggested that political and institutional considerations tilt the balance toward cap-and-trade, but the picture may have changed since we started this discussion, given these differences in the players. At the risk of drawing premature conclusions from the economic crisis and congressional bailout negotiations, I wonder whether Congress and the public will have as much appetite as before for creating a new market commodity by government fiat.
Whether awarded for free or through auction, an economy-wide cap-and-trade program for greenhouse gases would create a new market estimated at more than $100 billion in 2012. Carbon-backed allowances are different from mortgage-backed securities in most respects, and they bear different risks to society. But it may be politically more difficult now to argue on Capitol Hill that we should trust a trading market or proceed without a careful regime of regulation designed upfront to govern the new trading market, particularly when we're dealing with energy--another crucial element of the economy. Voters understand taxes (perhaps all too well, as Ken implies), but the complexity and market role of a cap-and-trade regime may be just as hard--or harder--for voters to swallow now.
I'm not saying that a well-designed cap-and-trade regime is inherently fatally flawed as a matter of policy. We can watch and learn from the auctions under the new Regional Greenhouse Gas Initiative (RGGI) for 10 eastern states, the first of which was held at the end of September. But RGGI is limited to the regional, electricity-generation sector, and an economy-wide, national trading regime may carry different political calculations at this point in time.
In addition, recent events underscore another political and institutional issue--the cost to the federal budget. Evaluating the cap-and-trade bill that passed the Senate Committee on Environment and Public Works, the Congressional Budget Office estimates that the EPA would need appropriations totaling $1.7 billion for 2009 to 2013 to implement the program, including the cost of hiring up to 400 new employees. At a time of bailouts, will Congress and the public support the cost of putting a cap-and-trade regime in place and the associated growth of government?
If we're heading for tough economic times, any measure that raises the cost of energy will face its challenges, but taxes are traditional and may have a higher level of trust. Much as we may dislike them, they may now seem more like the good old friend.
Any system clearly needs to be well-designed. On that point, we all agree. The question then becomes whether a cap-and-trade system or a tax would allow for the best possible system to be put in place. We've already argued that a cap would be much more preferable on environmental grounds. So now we'll turn to simplicity.
Janet Milne has argued that a tax would be simpler than a cap. We disagree. The reason that a tax seems simple is that no one has proposed the detailed kind of tax that Congress would actually pass. After all, the Lieberman-Warner climate legislation considered by the Senate in June started as a two-page skeleton outline, which, in our estimation at least, seems simple enough. It's the process of getting a bill to the point where it will attract 60 votes that makes it complicated. (The U.S. tax code encompasses nearly 17,000 pages for a reason.)
Another common pro-tax argument is that a tax would be easier to administer and monitor. But cap-and-trade programs impose exactly the same emission-monitoring requirements as taxes. As David Weisbach points out, what affects monitoring costs is rather the "point of regulation." Under a cap-and-trade system, tradable allowances can be defined with respect to the carbon content of fossil fuels, and the allowances collected "upstream" from petroleum refineries and natural gas distribution hubs. In fact, that's exactly what the Lieberman-Warner cap-and-trade legislation proposed. Additionally, the costs of monitoring end-of-smokestack carbon emissions are pretty low--especially since virtually every U.S. power plant already has continuous emissions monitoring technology installed, collecting hourly carbon emissions data. (See for yourself at the EPA's eGRID data website.)
Finally, tax advocates claim that a cap-and-trade system will require an onerous administrative bureaucracy. It's true that cap-and-trade requires the creation of tradable instruments and a tracking system to monitor trades. But the costs of such a system would be small for two reasons. First, the system is already in place for electric power plants, thanks to the sulfur-dioxide trading system--these plants are a major contributor of carbon emissions. And second, experience from that program shows that the administrative costs associated with the emissions market amounted to an upfront cost of at most $20 million, along with something on the order of $70,000 annually--a tiny amount of money in a market with total compliance costs of more than $1 billion. Conversely, the bells and whistles tax advocates suggest to patch the holes in a carbon tax--i.e., refund systems to credit carbon sequestration, something that would be unnecessary in a cap-and-trade system--would only increase the tax's administrative costs.
Indeed, keeping things simple should be a prerequisite--although certainly not the most important one. The most important goal is preservation of the environment. That is, after all, the reason we're debating this issue in the first place. Yet, even if simplicity was the only criterion in this debate, a cap might still come out ahead.
If I'm reading the arguments made here correctly, the lay of the land is as follows: We agree that in theory a cap-and-trade system and a tax can be designed to be similar with respect to their coverage, their effects, and, to some extent, their administrative costs. For example, both can be equally imposed upstream or downstream, equally grandfather existing emissions, raise the same revenue, and, at least roughly, provide the same level of price certainty and emissions certainty. But this all depends on how well the systems are designed. The disagreement seems to be largely about which one will be susceptible to bad design decisions, or more crudely, which one will Congress screw up more. Congress is good at writing bad tax laws; would it also bungle a cap-and-trade regime? And will the administrative costs of a cap-and-trade regime as actually implemented exceed those of a tax as actually implemented?
It's unclear how to determine the answer to these sorts of questions. We cannot perform an experiment, nor do we have natural experiments or historical data series to analyze. Existing carbon tax regimes aren't very good, but neither are the existing carbon cap-and-trade regimes--most significantly the EU Emissions Trading System (ETS). However, the existing carbon taxes and the EU ETS were enacted at different times with different information, so it's not clear that they can be directly compared. At best, we have intuitions about likely effects. My intuition tends to lie with Janet Milne's--a tax would be simpler because the institutions already exist. Perhaps one thing we might get out of the current financial crisis is how difficult it is to design well-functioning markets. On the other hand, I'm a tax lawyer by training and have plenty of experience with bafflingly complex tax laws.
Let me suggest two different lines of argument that might help resolve the issue in the absence of information about likely administrative costs. The first is that a cap-and-trade system must have a price ceiling (often called a safety valve) and a price floor. The goal of putting a price on emissions is to make polluters face the full costs of their actions. A price is necessary because the market price doesn't reflect the external harms that come from pollution. In a cap-and-trade regime, if the price spikes or tanks, polluters will very likely no longer face the true external costs of their actions. For example, if the price drops precipitously as happened in the EU ETS, industry would be able to pollute even if the external costs remain high. Similarly, if the price spiked because of short-term demand as happened in the Californian RECLAIM system, the regulatory cost would exceed the true external costs.
This lack of tracking between the trading price and the external harm is the major deficiency of a cap-and-trade system. A price floor and ceiling remedy the problem by ensuring that permits trade within a reasonable band. My guess is that many cap-and-trade advocates wouldn't accept a system with a ceiling and floor. Those who want absolute certainty on total emissions wouldn't accept a ceiling. Industry, which argues for a ceiling, wouldn't want a floor. To me, this means that we should start with a tax rather than risk getting a flawed cap-and-trade system.
Second, I believe that the central issue for choosing a pricing instrument, an issue that has unfortunately received scant attention in our discussions so far, is how each of the systems would be implemented internationally. A climate regime must involve all major emitters and the choice of instrument can affect whether we achieve broad participation and whether a treaty is enforceable. The claimed advantage of a cap-and-trade regime is that it increases participation because countries can be bought off through extra emission allowances, as was done in Kyoto. In theory, exactly the same thing could be done with taxes by transferring a portion of the tax receipts, but the transparency of such transfers may make this infeasible. On the other hand, buying off governments would likely involve giving billions of dollars of permits to badly behaved governments, something that most of the world would not agree to. And if our current foreign aid budgets reflect how much we're currently willing to give to foreign governments, there's no reason to think that we would spend orders of magnitude more in a climate treaty. Moreover, governments have little incentive to enforce caps on their own industries: allowing domestic cheating helps local industries and imposes externalities on the world. Taxes, I think, dominate in the international context.
As we wrap up the second round of our discussion, it might be useful to provide a way of framing our dialogue. Each of us has offered observations about the relative advantages and disadvantages of taxes versus cap-and-trade. But how do these differences relate to the (often implicit) goals in choosing a strategy for a climate change program?
When contemplating policy implementation issues, as we are here, it can be helpful to think of the problem as a "constrained cost-minimization problem"--i.e., we are trying to minimize the cost of the environmental program, but our choice of strategy is also subject to practical constraints. So what are the costs? And what are the constraints?
The costs include, of course, the actual costs of lowering emissions (i.e., the abatement costs), the negative effects--if any--of the program on our system of public finance, and the costs of implementing and maintaining the programs. But our choice of programs might be constrained by any number of factors--including politics, law, and the need to meet a specific environmental goal.
The abatement costs include, for example, the costs of adding carbon capture and storage onto coal-fueled power plants, substituting renewable energy sources for fossil fuel, and aggressively pursuing energy efficiency. The good news is that by presenting all parties with the same price for carbon, both taxes and marketable allowances can minimize the costs of abatement, as long as Congress doesn't toy with the price signal.
Generally when Congress imposes taxes to raise revenue, it creates what economists call "deadweight loss," inefficiencies (read social costs) that result from changing prices in the markets. But in the case of carbon taxes and auctioned marketable allowances, these distortions are minimized, at least in theory, because Congress is raising revenue by pushing parties to do something that needs to be done anyway--reduce emissions. As David Weisbach points outs, the key question is whether Congress is less apt to screw up one approach or the other? Certainly, the leading bills in the Senate suggest that legislators can't resist giving away a major portion of the allowances. However, there's no obvious reason to expect them to do better when it comes to resisting tax breaks.
The third area of costs--implementation and maintenance--includes the costs of establishing the program, setting up the bureaucracy, educating covered parties, monitoring emissions and compliance, and taking enforcement action against violators. It also includes the government's costs of making credible commitments to maintain the system so that long-term investments are protected. Janet Milne has argued that by taking advantage of the existing IRS infrastructure, carbon taxes will minimize implementation costs. I find Janet's argument compelling, but as I mentioned in my previous note, I worry that taxes are less stable than emissions caps. Thus, under a tax system, decision makers may be more reluctant to invest in energy-efficient capital if they think that Congress is more likely to change the system.
The primary constraint that we face in our choice of implementation strategy is that the system needs to provide some assurance of meeting the environmental goal, assuming that there is a clear environmental goal. The Intergovernmental Panel on Climate Change has provided us with a target--a 50 to 80 percent reduction in emissions relative to 2000 levels by 2050. On this count, the cap-and-trade system dominates. It isn't that the tax approach cannot reach any given target, only that we don't have a clear map of which carbon tax rates achieve a particular level of emissions reduction. And if we get it wrong on the first pass, it will be extremely difficult to get Congress to adjust the tax again.
While there are no clear legal obstacles to either the tax or cap-and-trade approach, the two do differ with respect to their political feasibility. For better or worse, there seems to be much more resistance to the tax approach than to a cap-and-trade system. This largely may be due to a misunderstanding by the public, but the impact on political feasibility is almost certainly real.
We're developing a good list of criteria for assessing the pros and cons--environmental effectiveness, the relative risks of bad design, costs, simplicity, political feasibility, and implications for an international regime. David Weisbach raises a good point about the need to look at what might work internationally, which I hope we can discuss more. For now, though, I'd like to focus on environmental effectiveness and political feasibility, where carbon taxes may be getting short shrift.
Environmental effectiveness is a sine qua non. Nathaniel Keohane, Gernot Wagner, and Ken Richards all question whether it's possible to identify the tax rate necessary to reduce emissions by the desired amount. That challenge should be an economist's dream. It will require making assumptions about behavior, technology, and economic conditions, but I suspect that a well-chosen team could provide a credible answer.
I also wonder whether a blunt instrument might be as effective as a scalpel. We're dealing with a multi-decade time span and commitment, and it's difficult to predict how technologies and purchasing patterns will evolve. A tax that sends a loud price signal could have real effects, just as the market-driven price increases in gasoline have triggered real changes. Stated differently, perhaps precision in the tax rate may not be as important as we think in producing fundamental, structural changes in our economy and lifestyle over the next two generations. A tax also allows us to go below the cap.
As we've agreed, effectiveness will depend on design. Ken speculates that legislators would give exemptions from carbon taxes just as they give away allowances in a trading regime; I have no doubt that lobbying would be intense. But a carbon tax worth supporting has a built-in brake. If one gives too many exemptions--or "grandfathering," to use David's term--there's not much left to tax. Too many activities escape the price signal that was intended to reduce emissions. And once the carbon tax proposal has lost its raison d'être, it's not worth enacting.
In order to be effective, the price signal of a carbon tax or trading system also needs to be reliable over time. Ken suggests that a tax might not be as stable as a cap-and-trade system, deterring carbon-reducing investments. But as David wrote earlier, both are susceptible to legislative change. The cap-and-trade proposal debated by the Senate in June called for periodic studies and legislative recommendations. In addition, cap-and-trade proposals often contain "off-ramps" that allow for flexibility. I'm not convinced that a cap-and-trade system would be built with less stable parts than a carbon tax.
In terms of political feasibility, the assessment may depend on how far one looks beneath the surface. Tax is a politically toxic word, but the Senate debate over the cap-and-trade bill this summer is sprinkled with arguments that the proposed cap-and-trade regime is the same thing as a tax. With a more prolonged debate, will trading programs really look that much more politically feasible?
The relative political feasibility of a cap-and-trade program turns in part on how the allowances are allocated. Giving them away will win more support from companies subject to the trading caps; they won't have to buy them at auction. But allocating all or many of the allowances at no cost may put a trading program in a politically weaker position for two reasons:
Auctioned allowances and carbon taxes avoid these two problems. When we compare the political feasibility of allowances and taxes, the two options may not be as far apart as one might think.
We gladly take on David Weisbach and Jane Milne's challenge to focus on the international dimension. David argues that a tax dominates on this count. We strongly disagree. A cap-and-trade system allows for the creation of a global carbon market. Such a market would provide the mechanisms and flexibility necessary to achieve the environmental goals at the lowest cost and the incentives for other countries to join. A tax does neither, while requiring much more harmonization across countries.
Since environmental integrity is the starting point in our debate, let's also agree that the United States or European Union (EU) won't be able to go it alone. Therefore, any policy solution needs to include China, India, and a cadre of other significant developing countries. Developing countries wouldn't be expected to decrease emissions as steeply as the United States and the ramp-up can be slower, but they need to be part of the solution nevertheless.
We also know that many other countries provide cheaper abatement opportunities than can be found in the United States or EU. Similarly, it will be necessary to provide financial incentives for developing countries to join because of ability to pay and the principle of "common but differentiated responsibilities" enshrined in UN documents (and agreed to by Washington) that says that industrialized countries ought to shoulder much of the burden.
How would such a system of international participation work under a carbon tax? The U.S. government collects taxes from businesses domestically. The Treasury then turns around and sends some of the collected money to its counterparts in China, India, or other developing countries, which, in turn, spend it on local abatement opportunities. That process, as David says, may be "infeasible." More directly put, it's simply not going to happen, and many economists and policy makers agree--former Treasury Secretary Lawrence Summers among them.
How would this process work in a global carbon market? As a first pass--and admittedly, we're offering a rosy picture here--it will be like any other market transaction. Instead of the U.S. Treasury, it would be a private transaction between U.S. and foreign companies. It's in the self-interest of the U.S. company to seek out least-cost compliance options. If that option can be found halfway around the world, so be it. Money will follow the market opportunity.
Concrete numbers are still hard to come by, but these financial flows would be in the low tens of billions of dollars. That alone shows that the market will be the only way to go. These sums are significantly more than current official development assistance or other nonmarket flows. Yet, they're small compared to overall market transactions. In 2007, industrialized countries sent, on net, in the order of $450 billion in foreign direct investment to developing countries. (The gross figure is closer to $1.7 trillion.)
For this reason, a cap-and-trade system could promote broad international participation. Developing countries would almost surely be net sellers in a global carbon market--both because they have ample low-cost abatement opportunities and because they're likely to receive more generous emissions targets than industrialized nations under an international agreement. As a result, emitters in the developing world could expect to earn substantial profits from abating emissions and selling allowances. Meanwhile, because advanced economies such as the United States and EU can set the terms of access to their own markets, they would have considerable leverage to persuade those other countries to take on binding emissions targets. A tax provides neither an incentive to join nor much leverage.
Moreover, if we think it would be difficult to find the right tax level domestically and regularly update it to achieve emission-reduction goals, this feat will be virtually impossible to do internationally. Yet it would be necessary to achieve a harmonized tax structure across countries. Environmental integrity quickly goes out the window.
Of course, we cannot jump from zero to a full-fledged global carbon market. We need several intermediate steps and safeguards along the way that ensure oversight and compliance--excellent topics for another debate. For the question at hand, though, the conclusion is clear: A U.S. cap-and-trade system sets us on the path toward a global market. It enables the kind of international monetary flows necessary to attract participation, and it can more easily be harmonized internationally. A tax does neither, and would even hinder the development of a global carbon market.
It's good to see the discussion turning to international issues, where it seems we have some disagreements and the most difficult issues arise. So far, the discussion has been noteworthy more for consensus than anything else. Although we learn from consensus, focusing on where we disagree is a better way to make progress. I suspect that part of the reason for our broad agreement is that none of us represents the major effected industries. In the political arena where these ideas ultimately have to pass muster, things won't be so easy. I also suspect that we do have some deep disagreements about the design of a carbon price even purely within the domestic arena. In particular, if we're to have a cap-and-trade system, I don't think we agree on whether a cap-and-trade system should have a pricing ceiling and price floor. I believe it must. Perhaps if we have space in another round of discussion, we can address this issue.
But back to the point at hand, which system would work better internationally--a carbon tax or a cap-and-trade system? As Gernot Wagner and Nathaniel Keohane point out, a key issue is finding a way to engage developing countries while still ensuring that developed countries participate. They suggest that we would be willing to buy off developing countries through permit allocations when we wouldn't be willing to make the same size direct cash transfers within a tax system. Under a permit system, U.S. businesses would have to purchase permits from developing countries, giving the developing countries the same cash that they would get through a tax-and-transfer system, except implicitly through the permit allocation rather than explicitly through cash transfers.
There are two possibilities for such a system. The first is that nobody is fooled: People would understand that a generous permit allocation would result in large transfers to various countries, transfers that vastly exceed our current foreign aid budget. In this case, if we would be unwilling to transfer those sums to developing countries explicitly, we would be unwilling to do so through permit allocations. The second is that people are fooled: Somehow because of the complexity of the system, people don't understand that we have increased our transfers to foreign countries by an order of magnitude through permit allocations. This seems to be what Gernot and Nathaniel hope for. But this is neither stable nor democratic. It isn't stable because you can only fool people for so long. Once U.S. businesses have to start paying foreign governments for permission to operate, people will quickly figure out what has happened. It isn't democratic: If people understood the size of the transfers, they wouldn't agree. If we're to enter into a long-term path toward a climate solution, people must understand and agree with the mechanism. Hiding the ball isn't a good strategy for a policy that must endure for decades to centuries.
The other major problem with an international cap-and-trade regime that includes developing countries is enforcement. Countries have an incentive to sell their permit allocations for foreign entities to generate cash and then not enforce the resulting emission reductions in their local industries. A permit allocation to a corrupt government is like free money, something we try not to do in the development context. The alternative is to only allow sales of permits by countries with good enforcement records, but if permit buyers are held hostage to local enforcement by governments of various sorts around the world, the permit market will lose many of its benefits and possibly even fail.
Would a tax work any better? International participation in a climate treaty is one of the thorniest issues in climate change policy, so I can't argue that it would be easy or simple. But yes, I believe a tax would work better than permits. Countries have an incentive to enforce taxes, so the enforcement problem is much smaller with a tax than with permits. To some extent, developing countries could be induced to participate through lower initial taxes plus transfers in the form of technology assistance. Harmonization would be imperfect but it wouldn't be hard to have the same basic components of the tax across countries. After all, international trading of permits effectively requires perfect harmonization while taxes would do reasonably well with less than perfect harmonization. Most countries now have value added taxes that operate in a roughly similar manner. There is no reason this couldn't work with a carbon tax.
Let me close by turning to the issue of environmental integrity, which is something people use to argue for permits over taxes. There's no difference in terms of environmental integrity. Taxes would be able to provide the same emission reductions and the same ultimate carbon concentrations as permits. Just as the number of permits would decline over time to reflect some target, tax rates would go up over time to achieve the target. Just as the number of permits actually issued in any time period would ultimately have to be adjusted to reflect current conditions and new learning about the climate, tax rates would be adjusted. The potential to lower taxes in response to political pressures is no greater than the potential to increase the number of permits. And permits would, in their actual implementation, likely provide flexibility about when to reduce emissions, so to the extent that people believe that their inflexibility is a virtue, it is unlikely to be realized. Moreover, the effect of any remaining uncertainty about final carbon concentrations with a tax as compared to a permit system are swamped by the uncertainty concerning the ultimate effects of carbon concentrations on the climate.
Ultimately, I go back to where I started: It's all about design. If we are to have a carbon pricing system, the critical issues are about designing it to work well--broad coverage, auctions in the case of permits (or no grandfathering of taxes), flexibility, and broad international participation should be the central goals. A tax can achieve these goals more easily than a permit system. But assuming the underlying policy goals are met, it may not matter much what we call it.
