Multipliers: The Very Sharp Greg Ip Gets One Wrong
links for 2009-09-26

Greg Mankiw and Company Do Good on Cap-and-Trade (from 2004)

The George W. Bush Administration's CEA's official analysuis of the issues:

2004 ERP:

Market-Based Quantity Regulations: Cap-and-Trade

The main problem with an emission fee is that it is difficult to know beforehand what fee level will achieve the desired amount of pollution reduction. A cap-and-trade regulation addresses this issue and provides market incentives to reduce emissions in a cost-effective way. Such regulations "cap" the amount of allowable emissions and require that a firm own a permit for each unit of pollution emitted in a given period (for example, a year). This permit effectively establishes a legal property right for the air affected by the pollution, so that any emissions must be paid for by the firm. The government allocates the pollution permits to the emission sources and then allows the sources to buy and sell permits from each other.

Under a cap-and-trade system, a source with a high cost of reducing an additional unit of emissions would be willing to purchase a permit from a source with a lower marginal abatement cost. With a well-functioning market for the permits, sources will trade permits until the price for the permits equals the marginal abatement cost. As with the emission fee, the marginal abatement costs will be equal across sources, leading to a cost-effective result. The cap-and-trade system also provides an incentive to reduce emissions because each unit of emissions reduction saves the source the price of another permit. This regulation sends a market signal that there is a price for emissions and any innovative means of reducing emissions will save firms from paying the price. The cap-and-trade system therefore achieves the target level of pollution reduction at the lowest cost.

One consideration for a cap-and-trade system is how to allocate the permits initially. A cap-and-trade system that allocates the permits based on historic emissions or other firm characteristics, known as grandfathering, in essence gives away a valuable asset--the permits. A grandfathering system could establish a barrier to entry for new firms because any new entrant would have to purchase permits from existing firms.

One way to avoid these problems is to auction the permits at some regular interval to the highest bidders. Firms with higher marginal abatement costs would bid more for permits than those that can achieve less-costly emissions reductions. While auctioning the permits would result in lower profits for the regulated firms (compared to giving away the permits), it would not affect the firms' output decisions. Grandfathering versus auctioning the permits is primarily a question of distribution, not efficiency--it is a question of whether a public asset should be given to firms for free or sold as a means of generating public revenues.

A notable example of a cap-and-trade system is the sulfur dioxide (SO2) trading program created under Title IV of the Clean Air Act Amendments of 1990. The program set a goal of reducing emissions by 10 million tons from the 1980 level by 2010. This was to be accomplished in two phases. The first phase, which began in 1995, initially capped the SO2 emissions at 263 individual units which were owned by 110 electric utility power plants in 21 eastern and midwestern states. These plants, which were primarily coal-fired, emitted the greatest amounts of pollution among power plants in these regions. From 1995 to 2000, an additional 182 units were allowed into the program. The second phase, which began in 2000, further decreased the annual emissions of SO2 and required all large fossil fuel-fired power plants in the contiguous 48 states and the District of Columbia to hold permits to cover their emissions.

In both phases, power plants could purchase permits from other power plants in order to meet their emissions coverage. The program also allowed plants to carry over (or bank) unused permits to use in later years, which gives firms even greater flexibility in achieving long-term pollution reduction. In contrast to a command-and-control system, this cap-and-trade system allows plants that find it costly to reduce their SO2 emissions to purchase credits from plants that can reduce SO2 at lower cost.

Evidence indicates that such cost-saving trades did indeed take place as firms took advantage of the system's inherent flexibility (Chart 9-4). Each bar in the following chart represents the emissions rate each plant achieved after trading permits in 1997. The superimposed line in the figure shows the level of emissions each plant would have had to achieve in the absence of trading. Bars below the line indicate plants that reduced their emissions by more than the required amount and sold their excess permits or banked them. Bars above the line indicate plants that purchased permits or used previously banked permits to avoid costly abatement. The figure shows that almost every plant took advantage of the flexibility of the system, suggesting that plant-level costs of reducing SO2 emissions vary greatly.

The trading program has achieved its pollution-reduction goals at great cost savings. By the end of the first phase, emission reductions were almost 30 percent below the required level. The flexibility of this approach has been estimated to provide cost savings of approximately $0.9 billion to $1.8 billion a year compared to costs under a command-and-control regulatory alternative; other tradable-permit markets have had significant cost savings as well (Table 9-1).


Emission Fees Versus Cap-and-Trade

As mentioned previously, one problem with emission fees is that it is difficult to know beforehand at what level to set the fee to achieve the desired pollution reduction. This might require periodic adjustments of the fee level, and such adjustments would introduce uncertainty that could interfere with firms' planning decisions. The emissions fee does, however, allow the government to set with certainty the marginal cost of emissions reduction. For each emission fee there is a corresponding allocation of permits that would achieve the same results; however, it is difficult to know beforehand what the market price for permits will be once trading actually takes place.

One way to reconcile these issues is to offer a cap-and-trade system with a safety valve. The safety valve sets a maximum price for a permit, which guarantees that the price of reducing emissions does not exceed the expected benefits. The regulatory agency issues and sells extra permits on request from any firm at this fixed safety valve price, thus guaranteeing that the market permit price does not exceed this level. A cap-and-trade program with a safety valve achieves the target level of emission reductions in a cost-effective manner, while protecting the regulated firms against unexpected short-term price increases in emissions reduction.

The President's [Bush's] Cap-and-Trade Program

An example of a well-designed incentive-based regulatory approach is the President's Clear Skies proposal for reducing emissions of sulfur dioxide, nitrogen oxides, and mercury from electric utility generators by approximately 70 percent by 2018. Clear Skies would cost-effectively reduce emissions by establishing a cap-and-trade system for each of the three pollutants. The EPA has estimated the benefits of the Clear Skies Act at $113 billion annually by 2020, compared with $6 billion in projected annual costs. These include $110 billion in annual health benefits (including the prevention of 14,100 premature deaths and 30,000 hospitalizations and emergency room visits) and $3 billion in annual benefits from increased visibility at national parks. Under the existing Clean Air Act, the EPA issues national air-quality standards for certain pollutants, including particulate matter and ozone. The EPA projects that compared with existing programs, the Clear Skies Act would lead 35 additional eastern U.S. counties to meet the particulate matter standard by 2020, leaving only eight counties not meeting the standard. The EPA expects that the remaining counties not meeting the standards would move closer to achieving them due to the Clear Skies Act.

To mitigate the effects of market shocks that potentially affect the costs of emissions reduction, Clear Skies would establish a safety valve price for permits of each pollutant. It would also provide regulatory certainty by achieving the reductions of all three pollutants in two phases. Firms would therefore plan their reductions of the three pollutants together and over the long term. Indeed, because the Clear Skies plan allows the banking of permits for future use, it provides an incentive for firms to achieve reductions quickly. Additionally, Clear Skies would provide revenue for the government because it phases in an auction system for the permits.

Clear Skies demonstrates the lessons learned from past regulatory experiences: instead of imposing an inflexible, command-and-control regulation to achieve emissions reduction, it offers a market-based, cost-effective, cap-and-trade program to achieve large reductions in emissions from electric utility generators.


Economic growth and environmental improvements are at times incorrectly seen as competing aims. Increased economic production can indeed lead to greater environmental degradation. However, an increase in economic resources provides more options (most notably, technological advancements) for addressing environmental problems. Moreover, a growing economy can also lead to increased demand for environmental improvements. It is therefore important to weigh the direct environmental benefits of a regulation against its economic costs. The goal should be to maximize the net benefits to society, while also giving due consideration to distributional issues. Maximizing net benefits is best achieved in a free-market setting unless there are spillover costs to third parties.

Spillover costs are best addressed by establishing property rights that will lead the affected parties to negotiate a mutually-beneficial outcome. If the costs of such negotiations are prohibitive, however, government should respond carefully and always keep in mind the possible government spillover costs. To make effective regulations, the government must first assess the environmental problems using sound, unbiased estimates of the hazards and then craft incentive-based regulations to address them. Such regulations can address the spillover costs of environmental problems at lower costs to society than the traditional command-and-control regulatory methods. These principles, and the lessons learned from our past regulatory experiences, as described throughout this chapter, should guide our future regulatory endeavors to achieve environmental improvements coupled with economic growth and efficiency.