Over the past few years, a debate has raged in Washington about whether the government should restrict carbon emissions in order to reduce the risk of climate change. In the course of that debate, “cap-and-trade” has made its way into the public vocabulary, even transforming into something of a political catchphrase. Many people now associate the term with activists and politicians who demand far-reaching measures to address what they believe is a looming, and serious, threat to the environment.
Because the legislation stirring this debate is fairly recent, the cap-andtrade concept has come to seem like a new innovation devised solely to achieve reductions in carbon emissions. But in fact, the idea has been discussed by economists for more than four decades. It has long been considered a means of using economic incentives to curb activities that reduce environmental quality — like the emission of harmful particles into the air or water, or unsustainable over-fishing of the seas.
So why has the most recent iteration of cap-and-trade suddenly stirred up such passions, and come to be so vehemently opposed by people who would, in other circumstances, celebrate the use of market forces to bring about a public good? Perhaps it is because this cap-and-trade debate has had more to do with a dispute about the nature of the problem than about the character of the proposed solution. Deciding whether (and how) the government should respond to the risk of climate change requires considering the extent to which anthropogenic climate change is actually occurring; the need for coordination with other countries’ carbon policies; and the ethics of imposing added costs (especially while the economy is weak) in exchange for the uncertain prospect of long-term gains. The ongoing controversy about climate change thus risks preventing many friends of market economics from considering market-friendly government regulation in other arenas of environmental policy.
It is therefore worth stepping back from the particulars of the recent climate-change debate to consider the broader problem of how best to reduce environmental degradation. At its core, the problem comes down to applying market pressures in an arena in which property rights do not operate as they do in the rest of our economy — and in which markets therefore often need some help to function properly.
Under these circumstances, some version of a pollution tax or cap-and-trade system is likely to be the best way to attain environmental benefits while minimizing the costs to American producers and consumers. It is also the form of environmental regulation best able to keep direct intervention by government at a minimum. Given the alternatives, friends of markets and limited government would be wise to move beyond the negative associations caused by the recent debate, and to give market-based environmental regulation another look.
Pollution and Property
Free markets work by voluntary exchange, so that a trade occurs only if both the buyer and the seller perceive it to be beneficial. And prices are the means by which such perceptions are formed and communicated. A market price conveys information about the strength of consumer demand for a good and about the scarcity of supply for that good, allowing for the implicit balancing of buyers’ and sellers’ interests without the need for a coordinating authority. If the cost of some raw material that is needed in order to produce some particular product goes up, for instance, then the price of the product will also go up. As a result of the price increase, the quantity demanded by consumers will go down. These market adjustments, of which prices are the indispensible ingredient, are the means by which both shortages and surpluses are prevented.
This flow of information — and the ensuing coordination among suppliers of raw materials, producers of finished goods, and consumers — occurs without any centralized oversight, and is generally accomplished with greater efficiency than any overseer could ever hope to achieve. As Friedrich Hayek put it, “[t]he marvel is that in a case like that of a scarcity of one raw material, without an order being issued, without more than perhaps a handful of people knowing the cause, tens of thousands of people whose identity could not be ascertained by months of investigation, are made to use the material or its products more sparingly; i.e., they move in the right direction.”
Nonetheless, as Hayek himself noted, the price mechanism does not provide perfect information and does not necessarily lead to a perfectly efficient distribution of resources. For one thing, prices reflect common perceptions, even when those perceptions are based on incorrect information. And recent history demonstrates the dangers of misguided herd behavior — such as when mortgage-backed securities, say, become dramatically overvalued. The result is a severe misallocation of economic resources.
The price mechanism also falls short when a voluntary transaction between two parties affects a third party, producing some sort of harm to an innocent bystander. This is known as the problem of externality, and it is why pollution and environmental degradation pose a challenge to the free-market system.
Consider the case of a firm that produces a good that a consumer then buys, to both the firm’s advantage and the consumer’s. The production of this good, however, results in air pollution that damages the health of another person living near the factory. In this case, the free market functions inefficiently; the externality problem arises because the environmental good — in this case, clean air — is priced at zero. The two parties involved in the trade neglect the health cost borne by the third party (or by a great many third parties), and the price they agree on thus does not reflect the true and total cost of producing and consuming the good in question.
The source of this failure of pricing is the lack of clear property rights over the environmental good. If the third party exposed to air pollution owned the clean air in question, then any action that involved polluting the air would require his consent, which would be granted only if the polluter compensated him for the damage caused to his property. This would then establish a price for the environmental good, and the free-market system could account for its value to everyone involved.
As economist Ronald Coase explained in 1960, a price for the environmental good could be established even if the property right applied in reverse. For example, if the third party did not own the right to clean air, but instead the producer owned the right to pollute the air, then the person exposed to the contaminants would need to pay the firm to reduce its pollution. Reversing the property right would bring about a different distributional result, of course, since everyone prefers to be paid rather than to pay. But the process would still take account of the environmental good, and of its worth to the people who benefit from it.
The important point is that, regardless of whether the firm purchased the use of the air or the person exposed to the pollution paid the polluter to reduce emissions, the simple establishment of property rights over the air would allow a price to be set for the environmental good. This, in turn, would meet the essential precondition for the free market to bring about optimal amounts of both production and environmental protection.
Why would property rights and prices help yield an optimal outcome? When pricing can be established for the clean air, it means that the producer and the third party have settled on the point at which allowing any more pollution would create health damages in excess of the benefits to the firm, and at which allowing any less pollution would create economic damages to the firm in excess of the health benefits to the person exposed to the pollution. A free market with clearly defined property rights would prevent both of these inefficient outcomes, because the person suffering the harm in either case would be willing to pay more than enough to compensate the other party to prevent any movement away from the optimal balance.
Note that where such a process of compensation is in place, any further reductions in emissions or production mandated by the government — or any additional fee charged by the government for pollution — would lead to a sub-optimal outcome. The added reductions — either directly mandated or effectively forced through increased fees — would lead to too little pollution, or too little use of natural resources, meaning the firm’s production (and, with it, economic welfare more generally) would decline. Proper pricing by the private market, then, would make heavy-handed government regulation and taxation unnecessary and, indeed, detrimental.
But establishing proper pricing is no easy feat. While the absence of property rights is at the heart of the challenge of environmental regulation, the simple assignment of property rights is not enough to ensure an optimal outcome. Most pollution disputes involve large numbers of polluters, and even larger numbers of people exposed to the pollution. The scope of the matter complicates the assignment of property rights, and renders the normal methods of enforcement unserviceable — as arriving at a price would require a vast (and thus highly expensive) process of multi-party bargaining. A firm that seeks the right to pollute, for instance, would need to negotiate with every single person exposed to the pollution in question in order to arrive at the optimal outcome. And not one of these people could sell his clean-air rights without simultaneously affecting the air quality experienced by everyone around him. The free market does not provide an effective and inexpensive mechanism for aggregating the individual benefits to all parties. And as the number of affected parties increases, the likelihood of voluntary collective action diminishes.
This leaves government action as the principal remedy for environmental degradation that involves many parties. But government action need not take the form of crude bans and quotas. Rather, the government’s role should be to make the environment resemble, as closely as possible, a “private good” in the eyes of its users.
Since the primary advantage of treating the environment as a private good is the ability to set a price for it — and since the lack of a price for environmental goods leads to inefficient overuse — government’s task is thus clear: It should create a price for pollution. Properly carried out, price-setting is the essential characteristic of what economists call market-based environmental regulation. And it can achieve environmental goals at a lower cost (to producers, consumers, and the broader economy) than more prescriptive government regulations that, say, require the use of certain clean technologies or restrict production — the sorts of policies, in other words, that form the core of our environmental regulations today.
Establishing a Price
Once it is clear that government should set a price for pollution, government has basically two options: to set the price directly through a pollution tax, or to set it indirectly through a cap-and-trade system.
While cap-and-trade has a decades-old pedigree, the idea of a pollution tax dates back even further — to the work of economist Arthur Cecil Pigou in the 1920s. Pigou believed that negative externalities caused by market interactions justified government intervention. And he showed that the way to reduce damaging externalities was for government to tax the activities leading to them, based on the amount of damage they caused — thereby giving offenders very real, material reasons to cut back on their harmful behavior.
A pollution tax provides an incentive for a firm to reduce its pollution, and thereby to economize its use of the environment. For each unit of pollution the firm might produce, the polluter must first consider: “How much will it cost to pay the tax on this pollution?” And then, “How much will it cost instead to avert this unit of pollution, by using the means at my disposal to manufacture my product without contaminating the environment?” Guided by these considerations, each polluter will reduce emissions so long as the answer to “How much will it cost to pay the tax?” is greater than the answer to “How much will it cost to avoid the pollution?”
Ideally, a pollution tax should be set at a level equal to the damage caused to third parties by the producer’s next unit of pollution. In practice, however, few environmental taxes achieve this goal. In the United States, the primary purpose of most environmental taxes is not to curb pollution but rather to raise revenue; thus pollution control is instead pursued through regulation. As a result, the taxes tend not to bear any relation to the true costs inflicted by pollution.
But a direct tax is not the only way to price pollution. In 1968, economist John Dales showed that the same goal — minimizing the cost of reducing pollution — could also be achieved by having the government assign polluters transferable rights to their pollution, and by setting the available amount of those rights such that pollution would be capped at the overall emissions goal. Each polluter would be required to submit one allowance (also known as a permit or credit) to the government for each unit of pollution emitted; these allowances would be transferable between polluters, thus allowing variation of emissions levels across firms. This scenario is now known as a cap-and-trade system.
Under such a system, the government could either auction off the allowances — which would turn the system into a source of public revenue — or it could give them away for free to regulated entities. Either way, the firms could then trade the allowances with one another. Much as Ronald Coase demonstrated that a price for a non-market good is established no matter who receives the property right for that good, the same permit price results under cap-and-trade regardless of whethethe government distributes allowances through an auction or gives them away for free. The relative advantage of auctioning depends on how the resulting revenues are used. The more that the revenue is employed to reduce economically harmful taxes or to invest in productive public goods, the stronger the case becomes for auctioning the permits rather than giving them away. (In practice, most cap-and-trade systems around the world have chosen free distribution over auctions.)
Generally speaking, cap-and-trade leads to the same outcome as a pollution tax. The cap creates a scarcity of allowable pollution, which then generates a price for pollution; this price is equal to the one that would have been set by the tax, given the same pollution-reduction goal. Similar to the incentive structure with a pollution tax, under cap-and-trade, each polluter reduces emissions until it costs more to reduce one additional unit than to buy (or not sell) an allowance. In this way, cap-and-trade — like the pollution tax — leads to cost minimization. But under cap-and-trade, rather than setting a price directly, legislators or regulators set a pollution goal. The market is then allowed to arrive at the permit price on its own.
The Right Regulations
This sort of decentralization is the chief advantage of market-based regulations like pollution taxes and cap-and-trade systems. Unlike other, more burdensome regulatory approaches, these policies do not rely exclusively on the controlling hand of government, but rather on the power of dispersed decision-making spread across millions of producers and consumers. Today, unfortunately, most environmental regulations are of a decidedly non-market variety, known broadly as command-and-control regulations — which rely instead on the government to prescribe exactly how pollution reductions are to be achieved.
There are two broad categories of command-and-control environmental regulations. The first is known as a technology standard, in which the government prescribes a particular pollution-reduction technology — thus leaving companies little or no latitude in how they achieve the desired pollution reduction. The Clean Air Act, for example, has largely relied on technology standards. As part of this legislation, the Environmental Protection Agency determines, among other things, the “maximum available control technology” that must be installed by major industrial facilities emitting hazardous air pollutants; the EPA also specifies the “best technological system of continuous emission reduction” for new sources of common air pollutants. The problem with laws like the Clean Air Act, however, is that, by mandating a certain technology, such regulations close off the possibility of firms searching for other low-cost options for curbing emissions — such as switching to a cleaner fuel, reducing production, or developing an alternative pollution-control technology.
A somewhat less restrictive type of command-and-control regulation is known as a performance standard, in which the government prescribes a particular level of emissions or an emissions rate — for instance, pollution by unit of production — for each regulated firm. (Sewage-treatment plants in America, for example, are subjected to monthly caps on biochemical oxygen demand.)
With performance standards, the required level or rate of emissions is typically uniform across firms. In theory, these regulations are cheaper than technology standards, because they can allow firms some discretion over how they achieve the prescribed level or rate. But the tendency of legislators and regulators to set emissions rates rather than total emissions levels often undermines this flexibility: For example, a regulation that caps emissions per unit of production limits the incentive to reduce emissions by lowering production, thus effectively taking that option off the table.
Market-based systems — including both pollution taxes and cap-and-trade regimes — are far superior to both of these command-and-control policies. Their greatest benefit is that, even though they do require government to set the price of pollution (either directly or through the establishment of a pollution cap that yields a price), the regulated firms have the flexibility to respond to that price that a command-and-control system denies. And this flexibility exists not only within each firm, but across firms in the entire market.
Consider first the role of flexibility within firms. Under market-based regulation, any unit of pollution reduced by the polluter provides savings equal to the amount of the pollution tax (or the price of the pollution allowance in a cap-and-trade system). The polluter therefore has an incentive to take action to reduce pollution, so long as those pollution-prevention measures cost less than paying the tax or buying (or not selling) an allowance. But just how the polluter reduces those harmful emissions is up to him: He can choose among such options as switching to a cleaner-burning fuel, reducing production, installing a pollution-reduction technology, or exploring the development of a new pollution-control technology. Absent direct government regulations that prohibit specific production techniques, or that mandate specific anti-pollution safeguards, firms will have a wide range of options to consider for meeting their pollution-reduction targets. And the wider the range of options a firm can consider, the lower the cost of reducing pollution will be. Each firm — faced with the same pollution price as other companies, but able to consider its own unique production model, business plan, labor and materials costs, and available cash reserves or credit — can decide for itself which method of reducing pollution makes the most sense for the firm’s bottom line.
Now consider the role of flexibility across firms. In addition to searching for its own low-cost options to reduce emissions, each firm can also implicitly choose from other firms’ possible actions. For example, under cap-and-trade, a firm that finds it very expensive to reduce pollution by one unit can buy an allowance from a firm that achieves the same pollution reduction much more cheaply. As far as aggregate pollution is concerned, the effect is the same as in a system that allows only within-firm flexibility. Costs, however, are lower, because of polluters’ ability to shop among all firms for less expensive allowances. The same holds true for a pollution tax. In this case, the firm that bought the allowance under cap-and-trade will instead opt to pay the tax for the right to pollute more; the firm that sold the allowance, meanwhile, will reduce emissions in order to avoid the tax. The result is the same pollution from each firm, and the same compliance costs under a tax as under cap-and-trade.
Admittedly, this flexibility across firms can create problems. For instance, the damage caused by pollution often depends on where the emissions occur; a factory that pollutes the air upwind of Cleveland, Ohio, will do much more harm to people’s health than a factory that pours the exact same amount of contamination into the air over the Nevada desert. As a result, shifting emissions reductions from one place to another through allowances trading could increase total damages. Ideally, a pollution tax would vary by geography, as would the allowance exchange rates under a cap-and-trade program; each would lead to a higher price for the more damaging pollution occurring upwind of Cleveland than for the pollution occurring in the Nevada desert. In practice, however, such nuanced policies would be difficult to implement, and they have rarely been included in the design of market-based regulations.
Problems can also arise if the compliance requirements for market-based regulations do not include enough flexibility through time. Without temporal flexibility, a firm cannot adapt to unexpected market conditions. If reducing pollution is suddenly and unexpectedly more expensive for a firm to achieve this year, then a rigid system will require the firm to incur heavy — and, for some firms in some economic climates, insurmountable — costs to meet the regulation.
One way to address this issue in a cap-and-trade regime is to provide firms with across-time flexibility by allowing them to bank unused allowances for future use and to borrow against future allowances for use in the current year. From the firms’ perspective, such banking and borrowing can provide protection against unexpected changes in market conditions. If the cost of meeting a cap unexpectedly rises this year, the firm can borrow allowances from the future, resulting in more pollution than is allowed by the cap this year in exchange for less pollution than is required by the cap in future years. Similarly, a firm can bank allowances in order to take precautions against future cost increases, thus polluting less than is required by the cap this year in order to be able to pollute more than is allowed by the caps in future years. In this way, the multi-year sum of emissions would eventually equal the multi-year sum of the annual caps.
Allowing complete across-time compliance flexibility allows businesses to be more nimble in any given year, and reduces the cost of achieving the emissions cap over a stretch of several years. But, as in the case of flexibility across geography, flexibility across time does have limitations: This system assumes, for instance, that the damage caused by a unit of emissions does not vary much from one year to the next, when in fact changing environmental or demographic conditions could dramatically increase or decrease such damage.
There is one disadvantage to market-based regulations compared to command-and-control rules: They depend on the ability to measure pollutants. This can be problematic when regulating emissions that are very costly — or simply impossible — to measure, such as agricultural run-off of chemicals.
But on the whole, market-based regulations are a far better way to balance the needs of the economy and the needs of the environment. And numerous empirical studies over decades have found market-based regulations to be substantially less expensive than the command-and-control approach. Given the choice between a market approach and strict control by government, then, the economics of environmental regulation gives market-based policies the edge.
Taxing or Trading?
Pollution taxes and cap-and-trade regimes are therefore clearly the best ways to regulate and limit environmental degradation and overuse. But is one of these approaches better than the other?
For any cap level in a cap-and-trade program there is, in theory, a corresponding pollution tax that will yield the same environmental and cost-minimizing results. In practice, however, policymaking is not nimble enough to constantly adjust to changing economic and environmental circumstances. The results of the two approaches will thus inevitably diverge over time.
Assume, for example, that there is an unexpected increase in the cost of reducing pollution one year after a policy is enacted. If there is, say, an unexpected and sustained increase in worldwide demand for electricity one summer, reducing the pollution associated with electricity generation would become more costly than anticipated. In this scenario, how would an electricity producer be affected by the market fluctuations under both cap-and-trade and pollution-tax regimes?
Under a cap-and-trade program, the annual allowable emissions level is set by the cap; even if the cost of pollution control increases, the cap does not change. Rather, the polluters must pay more to meet the cap, thus driving up the price of allowances in inter-firm trading. If there were instead a fixed emissions tax, however, polluters would respond to the cost increase by reducing the amount by which they cut pollution — opting to pay the tax rather than to pay the higher amount to reduce more pollution.
In other words, cap-and-trade regulates the amount of emissions rather than the price of compliance, whereas a pollution tax caps the price of compliance rather than the level of pollution. As Harvard economist Martin Weitzman demonstrated in 1974, the relative economic benefits of capping pollution (cap-and-trade) versus capping the price of pollution reduction (a tax) will depend on how much both environmental benefits and compliance costs are affected by the resulting change in pollution levels. When the environmental benefits of reducing an additional unit of pollution are relatively constant — as is the case with carbon emissions — then an unexpected change in cost has a big effect on the optimal amount of pollution reduction. In that case, a cap-and-trade system (which fixes the amount of allowable pollution, not the price) would sometimes deviate substantially from the optimal level, and a pollution tax would be preferable. But where the environmental benefits of reducing an additional unit of pollution or environmental degradation are relatively responsive to the level of pollution or degradation (as in the case of the overuse of a scarce resource, like timber or fish for example), cap-and-trade is often preferable to a tax.
A related advantage of a pollution tax over cap-and-trade is that a tax provides certainty about the price of polluting. Market participants prefer price stability, which allows them to better plan their capital-investment decisions (including, for example, long-term investments in low-pollution technologies). Furthermore, a pollution tax might provide more regulatory certainty (which market participants also crave), since any large, unexpected spikes in the cost of cap-and-trade allowances might lead certain business interests to complain to lawmakers, prodding future Congresses to act impulsively to change the cap.
It is possible to avoid some of these concerns by pursuing a hybrid approach between cap-and-trade and a pollution tax. For example, a cap-and-trade program could include a “safety valve.” Under such a system, if the cost of reducing pollution were to spike above a certain price ceiling, the government would offer additional allowances at a pre-established “safety valve” price. In other words, as soon as pollution-reduction costs exceeded a pre-set price threshold, the cap-and-trade program would turn into a pollution tax — and the cost of compliance would be contained.
Similarly, a price floor can be set so that, when the price of reducing pollution falls lower than anticipated — signaling that the original pollution-reduction target was not ambitious enough, and meaning that the cap-and-trade limits will provide too little incentive to reduce pollution — the government lowers the number of allowances available for trading. (Note that if the high-end safety valve and the price floor were set at the same price, we would be left with a pollution tax.)
Another problem policymakers might have to find a way around applies to both a pollution tax and certain forms of cap-and-trade. Remember that both approaches — taxes and cap-and-trade systems — work by increasing the price of pollution. This price increase is then reflected in higher prices for pollution-intensive goods, such as electricity. These price increases amount to a reduction in consumers’ real income, and so, in effect, act as an additional tax on income. Standard economic theory says that the economic inefficiency of taxation rises disproportionately as the tax rate rises. So the pollution tax (or cap-and-trade regime) magnifies the economic inefficiencies in our broader tax system. This compounding of tax inefficiencies — known as the tax-interaction effect — can substantially enlarge the overall cost (to the government and to individual consumers and taxpayers) of any environmental regulation.
The way to address this problem is to use public revenues from a pollution tax (or from a cap-and-trade program that auctions allowances, rather than distributing them to producers for free) to offset inefficient taxes or deficits. In a sense, environmental policy can be a type of tax reform, with the goal of using the pollution revenue to reduce economically harmful taxation. For example, the revenue from a pollution tax (or from a cap-and-trade auction) can be used to fund a reduction in marginal income-tax rates or corporate tax rates, or to fund deficit reduction (which amounts to a reduction in future tax rates).
The carbon cap-and-trade bill passed by the House of Representatives last year, however, would not have meaningfully offset other taxes in this way. Rather than auctioning all pollution allowances, it would have given away about 40 to 60% of the total permits for free. The government would have auctioned the remaining portion, but it would not have used the resulting revenue for tax or deficit reduction: Indeed, only 0.2% of the total revenues for 2016 would have gone toward reducing deficits.
Despite these concerns, a pollution tax and cap-and-trade have far more similarities than differences, especially compared to the more burdensome command-and-control approach. The advantages of one over the other will largely depend on the nature of the environmental pollutant being regulated. As a general matter, however, a pollution tax is more appealing than a cap-and-trade system. It is not susceptible to price volatility, and it offers a better chance of providing revenue that can be used to offset economically harmful taxes and deficits.
Wherever one falls in the debates over carbon emissions and climate change, there is no question that we as a soc