Obstacles to Global CO2 Trading: A Familiar Problem

Executive Summary

The creation of an international emissions trading system starts at home. Once a functioning national system of emissions trading is in place, there is every reason to believe that it would extend first to Annex I parties and, eventually, to non-Annex I parties who see the value of participating more fully in the benefits of emissions trading. The real problem-even in a domestic trading system-is not monitoring and enforcement, which would exist in any case, but the allocation of permits, which is fundamentally a political issue. Current concerns about such issues as equity and distributions of wealth reveal a basic lack of consensus on the nature of the climate change problem and suggest action is not imminent.


Global trading in CO2 emissions appeals to policymakers for two reasons. First, U.S. experience with the acid rain program demonstrates that tradable permits are an efficient means of meeting both environmental and economic goals.1 Second, the cost of meeting any domestic target can be lowered by purchasing less expensive emissions reductions abroad.2 This study examines issues that must be resolved in order to develop a global CO2emissions trading program. We start by examining how emission permits might be allocated and how best to handle the rents associated with this newly created scarcity. Second, we analyze how a national tradable permit system might evolve into global emissions trading.

The Problem of Allocating CO2 Emissions Rights

The Kyoto Protocol, if implemented, would impose a national cap for greenhouse gas emissions on each Annex I party to the Framework Convention on Climate Change. In the aggregate, this limit purports to recognize an emerging scarcity of the capacity of the Earth’s atmosphere to act as a repository for greenhouse gas emissions. We may debate whether the atmospheric sink is limited, and, if so, what the limit is from a physical sciences perspective, but as far as climate negotiators are concerned, an interim cap is now in place, subject only to ratification by the requisite number of parties.

In anticipating the cost of keeping within the proposed emissions limit, policymakers have begun examining the merits of a global emissions trading system. However, a global system will not exist until at least one national system proves functional. In short, the greatest challenge lies at home, and the biggest obstacle to implementing such a system-should we decide to-is a crassly familiar one: who gets the rent generated by limiting the right to emit CO2 (i.e., making this commodity a scarcity)? What previously was free would now become scarce, and scarcity presents any society with the decision of who has the right to use what has become scarce and who is to receive the rent associated with that use. Indeed, creating a national system may be even more daunting than assembling an international trading system.

The use of a resource and the receipt of rents attached to that use are often viewed as a single phenomenon. However, these two facets can be separated practically and analytically in a permit trading system. For example, most people would agree, at least in principle, that the atmospheric “sink” should be reserved for the most highly valued uses.

Distributing Revenue From Carbon Ceilings 

Today, many people-and certainly economists-would trust markets, more than government, to allocate use to the highest value. Accordingly, auctioning off permits granting access to the sink is frequently proposed; however, auctioning creates a new problem: what to do with the revenue. There is a rich literature on this subject, and some very appealing arguments on how the revenue might be recycled optimally. But there are two problems with any auction that involves large amounts of revenue. First, most people do not have any more faith in a government’s ability to distribute rents optimally than they have in a government’s ability to allocate use-a suspicion that gave rise to the auction in the first place. As such, the auctioning of permits takes on the appearance of a disguised tax.

The second and more fundamental problem is the assumption that a government owns the rights that are to be auctioned. The inconvenient fact is that these incipient rights are possessed de facto by existing emitters and actively exercised by them. From their point of view, the auction is not just a tax in disguise, it is confiscation of rights established by time-hallowed use. In a society that seeks to be just and equitable in dealing with its citizens, CO2emitters have a legitimate claim to compensation.

There are other ways to allocate use and rents, but as a matter of political economy, taxes and auctioned permits lose out unless there is overwhelming consensus to take action and existing users are poorly organized. When existing users are organized, which is usually the case, and there is sufficient consensus to take action, a deal can be and will be struck. The deal that can be struck is one in which the rights are limited, but given to the existing users, either through conventional command-and-control regulation or through grandfathered permits. If emissions rights are allocated through a grandfathering process, many find the explicit grant of the rent to be objectionable. They may prefer command-and-control regulation, for it does not appear to grant rents, but this is not quite the case. True, the allocated right is not as explicit as with grandfathered permits-and it may not be as secure-but it is still there, just well-hidden. It shows up in the value of the underlying physical asset, the power plant or factory, that is permitted by regulation to emit some amount.

In short, grandfathered permits alone seem to combine efficient allocation of use to the highest value with the politically expedient granting of the rents to existing sources. Of course the rent will remain obvious and will be charged to consumers. What is not clear is whether consumers would be more willing to pay fees to local utilities or energy companies that, under different circumstances, would be collected as taxes by Uncle Sam.

Grandfathered Permits: What Are the Issues?

Rents and Recycling. Allocation to private corporations lends itself easily to the demagoguery that already-rich corporations will be further enriched. However, it is not at all clear that corporate profits would be higher since charging for the use of the emissions’ permit will reduce demand for that company’s products or services. From this perspective, rents can be seen as compensation. From a societal standpoint, however, the discussion does not end here. Since the corporation is a shell, any revenue will be recycled through a combination of taxes, investment, and dividends. Approximately one-third would go to the government, hopefully to be optimally recycled. Moreover, the new set of prices would create an incentive to retool or replace existing capital stock with less carbon-intensive stock. In the best of all worlds, the increment of revenue here would make the financing of such investment easier, and might even reduce demands for tax credits. Finally, appreciated stock values would benefit stockholders, who are, because of the role of mutual funds, individual retirement accounts, and pension funds in modern industrial societies, a broadly distributed group. In fact, an interesting test is suggested: Would the government recycle its one-third as efficiently and equitably as the two-thirds that would remain with corporations?

New Entrants. Another concern is often voiced by potential new entrants, who say that grandfathered permits favor existing parties with whom they must compete. However, the real issue is whether the grandfathering process affects decisions at the margin. Typically it does not; it is a lump-sum transfer that benefits the recipient but does not change decisions about entry, retirement, or whether to produce more or less.3 Nevertheless, potential new entrants would complicate the development of a domestic emissions trading system, even if their claims are nothing more than disguised entreaties for their share. Behind their complaint is an even more serious concern: Who along the vertical chain of existing users has the superior claim to the right? In the case of the automobile, do the rights to the use of carbon reside in the fuel, or in the car? Or for that matter, with the driver? The same could be said of power plants. Do coal producers have the right to produce carbon, or do electric utilities have the right to emit CO2?

Monitoring. There is no necessary connection between monitoring emissions and permit ownership, but history suggests that monitoring will have a strong influence on who gets the permits. For example, in the U.S. acid rain program, monitoring and the distribution of permits occur at the same point, the power plant. So long as transaction costs are negligible, permits could be distributed to any party along the vertical chain. Thus, to take electricity as an example, carbon permits could be distributed to coal producers or even consumers as well as to power plant operators. All would have assets that would be adversely affected by the new set of prices.

Utility Deregulation. The allocation of rents associated with the SO2 permits in the acid rain program was relatively easy because the recipients (electric utilities) were presumed to be subject to cost-based regulation. Thus, the rent associated with the grandfathered permits was to be passed on to rate-payers. Still, there was plenty of jockeying for permits.4 Utility deregulation will make the allocation of carbon permits harder since it can no longer be argued that utilities act as agents for rate-payers. Furthermore, the other prospective recipients-oil companies, natural gas pipelines, and large industrial users-are not now regulated. Consequently, carbon rents will be as transparent as taxes, but not necessarily more politically acceptable.

Land as an Example. Land is a God-given resource made more scarce by increasing human and economic activity, not unlike the Earth’s atmosphere. It just became scarce a lot sooner. The history of establishing and perfecting the title to land is long, convoluted, and instructive. Today, in most industrialized societies, no one questions private ownership of this limited resource; and any entitlement we may receive comes through private inheritance, not from the state. Although there surely was an original recipient of the right to this land, now lost in the shrouds of history, no one seems to care. For example, it would be nearly impossible today to settle competing claims to all of mid-town Manhattan, but in the seventeenth century the value was low and there would have been little controversy over the rents, and probably active encouragement to settlers to take title to the land.

This would not be the case with the Kyoto Protocol. Emissions targets are so stiff-about 30 percent of 2010 emissions-that the value of the right is high and many parties would want ownership of the rents. This fact argues for gradually creating scarcity, so that discounting of future costs would dampen their present value.

Summary. My point in making these distinctions is not to argue for any particular rent allocation, but rather to emphasize the range and complexity of problems any government will face if it decides to set up a national emissions trading system. Creating a scarcity of the magnitude envisioned in the Kyoto Protocol raises fundamental issues of equity and of the definition of rights that are preeminently political questions. Analytically, the scarcity and the allocation of carbon emitting rights and rents can be separated, but in practice, the two are fused and there will likely be agreement on the creation of the scarcity only as there is agreement on the allocation of the rents thereby created.

Evolving Into an International System

The evolution of a national system into a global system is most easily described as a matter of trade. When the thing being traded can be produced abroad as cheaply as at home, we can anticipate that trade will develop across national boundaries. Parties at home will seek cheaper supplies abroad, and parties abroad will seek to tap into this new market opportunity. Here, we will discuss two types of international emissions trading: with other Annex I countries, and with non-Annex I countries who reduce carbon emissions and wish to sell this the right to emit thereby created.

Impacts of Annex I Trading

Trade among Annex I parties is not expected to bring the large benefits associated with global trading, but it is a useful point of departure. Suppose the United States and Norway, two Annex I countries, have solved their domestic allocation problems and adopted domestic emissions trading systems.5 Each would have its own autarkic market price for the right to emit CO2, and Norway’s price likely would be higher than the U.S. carbon emitting price. This difference in price would signal an opportunity for U.S. firms to abate more in order to free up permits for sale to Norway at a higher price. Full trading between these two parties would lead to an equilibrium price that would be higher than otherwise in the United States and lower than otherwise in Norway, but of mutual gain to both.

Such trading could be easily extended to other Annex I countries; all that would be required is faith in the validity of the permit, which would be provided by accurate monitoring of emissions and effective enforcement of each domestic permit system. This faith in the value of the permit is identical to what applies for any good or service in international trade. Moreover, the monitoring of emissions that ensures the value of the permits would be the same as would be needed to determine compliance with the Kyoto targets, even if there were no trading.

Trade With Developing Countries

While Annex I trading would be desirable, it would not lead to really large reductions in the cost of meeting the Kyoto targets. Less expensive abatement opportunities will be found in Eastern Europe and the former Soviet Union, both Annex I parties, but even these sources of emission credits will not be sufficient to bring about the really low price that models predict for full global trading.6 Figure 1 illustrates the demand and supply for emission permits, and indicates just how much difference participation by non-Annex I countries makes in these models. Note that the demand curve is the same for both Annex I and global markets, but the supply curves are very different. In the Annex I market, about 350 million tons (or 25 percent of the total reduction needed) would be supplied by the former Soviet Union. However, in a truly global trading system, predicted supplies are so ample that prices fall by two-thirds (or more) and Annex I countries import about 70 percent of the total reduction requirement.

[Figure 1: Aggregated Supply and Demand Curves in 2010 Under the Kyoto Protocol: Annex I Trading/World Trading]

Estimates vs. Measurements. The logic of non-Annex I participation in emissions trading is fundamentally no different than that for Annex I trading. Trade is a response to opportunity, and emissions trading is the mechanism by which emissions reductions would be made available outside Annex I parties. However, the emissions permits must not be counterfeit, and this requirement introduces a critical difference when non-Annex I emissions reductions are involved. When there is no cap, it is much harder to establish that the permit represents a real or “additional” reduction of emissions. Establishing that fact requires measuring a difference, and what would have been (but now is not) can be estimated but never truly measured. Consequently, the non-Annex I permit is only as good as the estimate-and even then the estimate will be costly and subject to challenges by anyone opposed to the proposed trade.

Transaction Costs-And How to Avoid Them. Policymakers can hope for the best in this process, but making these baseline estimates and even more getting them approved will limit this form of trading, as they have all forms of credit-based emissions trading. These transaction costs will limit Annex I access to low-cost reductions, and this will frustrate would-be non-Annex I exporters, but the underlying requirement cannot be avoided. No Annex I party operating under a tight cap-and-trade system can tolerate ersatz goods. What is needed are institutions to help reduce transaction costs, but project costs in the Activities Jointly Implemented program have not been particularly encouraging.7

It will not take non-Annex I exporters long to realize that these costs can be avoided-and are in trading among Annex I countries. If non-Annex I countries were to accept a limit and implement the monitoring and enforcement mechanisms necessary to ensure the integrity of their permits, transaction costs would be avoided and export quantities and earnings increased. The necessary monitoring and enforcement would already be in place, and there would be a certain economy of scope in negotiating the baseline all at once.

However, can an acceptable cap be negotiated for non-Annex I countries? Within the framework of the Kyoto Protocol it is possible to discern a principle concerning a non-Annex I country cap during the first commitment period: whatever emissions would have been absent any trading-related emissions reduction activity. Put differently, no constraint would be imposed on a non-Annex I party’s economic growth and there would be no “hot air.” All exported permits would reflect real emission reductions.

“Headroom.” The reality is not so neat. No one will know what emissions would have been without emissions trading, and given the possibility that economic growth, and therefore counterfactual emissions, might be greater than expected, prospective Annex I countries are likely to want to protect themselves by claiming the need for some “headroom.”8 Unfortunately, granting “headroom” increases the probability of “hot air.” On a practical level, both sides will have to compromise and some element of an enabling myth will be required. However, raising the negotiation from the project to the national level involves more than establishing a multi-project baseline for the first commitment period. Other considerations, discussed below, provide flexibility, but they also raise issues of allocation on the international scale.

Making Global Allocation Work 

To stabilize atmospheric concentrations, limits on developing country emissions will be needed eventually. This means extending limitations on the use of the atmospheric sink to all countries, or at least all major emitters. Accordingly, there is some advantage in having non-Annex I countries accept limits earlier, even if some headroom were involved.

For instance, if the ultimate allocative principle for global access to the atmospheric sink had been agreed to and were enforceable, and that principle allowed headroom now for some countries, then there should be no objection. The country would be free to bank or to sell currently according to its judgment of its interests, and cumulative emissions would be no greater. Even without such an ultimate principle, there is a strategic interest in establishing the precedent and the procedures of monitoring and enforcement, as would be required if Russia and the Ukraine are to export emissions reductions including their hot air. Moreover, if concern about hot air is sufficiently strong, nothing prevents any or several of the Annex I countries from transferring some of their limit to the non-Annex I applicant. For example, a very small part of the U.S. limit of 1.27 billion metric tons would provide comfortable headroom for Mexico. Indeed, if the reduction of transaction costs associated with accepting an Annex I limit were to increase supply sufficiently, the United States might lower costs by opening up these new supplies.

In fact, the ability of Annex I parties to use Article 4 of the Kyoto Protocol to transfer a portion of their limit to others raises new possibilities of bringing non-Annex I countries to accept limits. But this opportunity also returns us to the very issue that will have to be solved at the national level: how would the new scarcity be allocated? Theoretically, a global tradable permits system would allow separation of the assignment and actual use of emission rights. However, there is no more agreement on the global allocation of rights than at the national level, and perhaps less. The discussion will not even become serious among non-Annex I parties until there is some advantage in accepting an Annex I limit.

Flexible Mechanisms

So far, little has been said about how the Clean Development Mechanism (CDM) and Joint Implementation may assist the development of a global trading system. And for good reason; both are essentially transitional institutions. There would be no place for either in a truly global emissions trading market. However, their current role is very important. These institutions could keep project transaction costs as low as possible by locating cheap abatement possibilities and demonstrating the advantages of trade to both exporter and importer. In addition, such activities could provide measurement and enforcement experience and help host countries feel more comfortable in making the transition to a cap.

However, the creation of the CDM as the sole intermediary between Annex I buyers and non-Annex I sellers does raise important questions. At times, the Clean Development Fund appears to be a vehicle for North-South resource transfer. At other times, parties express concern about avoiding price-reducing competition among host countries for projects. Finally, the CDM sometimes is presented as a mechanism to facilitate trade by providing valuable recording, certification, and verification services. The CDM could become very bureaucratic and costly; if so, it will increase the incentive for those interested in emissions trading to bypass it by becoming Annex I signatories.


The creation of an international system of global trading starts at home. While very large reserves of cheap carbon abatement may lie abroad, these reserves will remain untapped until demand is created and value is imparted to them. Once a functioning national system of emissions trading is in place, there is every reason to believe that emissions trading would extend beyond national borders, particularly to other Annex I parties. Trade with non-Annex I parties faces difficult problems regarding transaction costs that diminish the prospects of abundant supply, but these same problems create an incentive for non-Annex I parties to accept Annex I limits and enjoy the full benefits of emissions trading. The evolution from a national to an international trading system will not come quickly, nor easily, but it is the most likely path of development because it is based on self-interest.

The main obstacle to global emissions trading lies at home. It is not monitoring and enforcement, which would exist in any case, but rather the distribution of the rents, which is fundamentally a political and even philosophical concern. Moreover, the issue is not simply public vs. private good, but also one of deciding which private claimants among many are most suitable. Unfortunately, making rents transparent also makes reaching agreement more difficult, yet a transparent domestic system lends itself most easily to expansion abroad.

Perhaps more than anything, expressions of concern about the fundamental issues of equity raised by the allocation of the proposed scarcity rents reveal a basic lack of consensus on the nature of climate change and the need for action. Those who view the problem as one of allocating a scarce resource are typically not much concerned about the allocation of rents. These individuals focus mostly on adapting to new circumstances and consider assigning rents to existing users a convenient way to provide compensation. The opposing view is that interested parties are not so much allocating a scarcity as they are limiting socially undesirable activity. To these individuals, there is no issue of preexisting rights or of compensation, and no other place for the revenue generated than the government.

Currently, no one view prevails, and there is little willingness to accommodate differing points of view. Indeed, the political solutions required to put a domestic emissions trading system in place-including passage of permit legislation and allocation of rents-are not readily apparent. In the meantime, it is worth working on the many details of an international system, because those details will facilitate the development of a market once national action has been taken.


  1. Richard Schmalensee, Paul L. Joskow, A. Denny Ellerman, Juan Pablo Montero, and Elizabeth M. Bailey, “An Interim Evaluation of Sulfur Dioxide Emissions Trading,” Journal of Economic Perspectives 12(3)(Summer 1998): 53-68, provides a quick summary of the essential features of the initial experience with emissions trading under the U.S. acid rain program.
  2. This point was emphasized in the testimony of the chairman of the Council of Economic Advisers, Janet Yellen, and illustrated in the recently released supporting analysis (Council of Economic Advisers, The Kyoto Protocol and the President’s Policies to Address Climate Change: Administration Economic Analysis, July 1998). Other models of global emissions and economic cost arrive at similar conclusions, when comparable assumptions are made.
  3. A fine distinction must be made here between the initial allocation, which determines receipt of the rent, and the use of the permits. In a functioning market for the permits, any party willing to pay the price has access to the permits, and every party using a grandfathered permit to cover emissions forgoes the opportunity cost of selling the permit.
  4. Paul L. Joskow and Richard Schmalensee, “The Political Economy of Market-based Environmental Policy: The U.S. Acid Rain Program,” Journal of Law and Economics, forthcoming, provides a good discussion of the allocation of SO2 permits.
  5. Norway has in fact decided to implement a system using grand-fathered permits to meet its Kyoto commitment; see “Norway Opts for Emissions Trading over CO2 Tax Expansion,” Global Environmental Change Report, 10 July 1998, p. 5.
  6. At Massachusetts Institute of Technology, we estimate a market price of about $120 for Annex I trading but only $25 with full global trading. See A. Denny Ellerman and Annelene Decaux, “Analysis of Post-Kyoto Emissions Trading Using Marginal Abatement Curves,” Report No. 40, MIT Joint Program on the Science and Policy of Global Change, 1998.
  7. The United Nations Conference on Trade and Development, “Greenhouse Gas Emissions Trading: Defining the Principles, Modalities, Rules, and Guidelines for Verification, Reporting, and Accountability,” Geneva, Switzerland, Draft of July 1998, is unequivocal on this point. For instance, “Under the pilot program for AIJ, [the] verification process has led to the rejection of many proposed trades, and can take one to two years, creating high transaction costs and uncertainty” (p. 6). Or, “In general, past programs that impose emission caps coupled with allowance trading have performed well, whereas credit trading systems have generally not performed to expectations” (p. 1).
  8. Jonathan Baert Weiner, “Designing Markets for International Greenhouse Gas Control,” Climate Issues Brief No. 6, Washington, D.C.: Resources for the Future, October 1997.

This paper was prepared for the September 23, 1998, policy conference sponsored by the ACCF Center for Policy Research, and will be published in the Center’s forthcoming book, Climate Change Policy: Practical Strategies to Promote Economic Growth and Environmental Quality.