Testimony of Randall S. Kroszner Member, Council of Economic Advisers Before the Clean Air Subcommittee of the Committee on Environment and Public Works United States Senate

June 5, 2003

9:30a.m.

Introduction

Mr. Chairman, and members of the Committee, I am pleased to appear before you this morning to discuss the Clear Skies Act of 2003. At this time, it is valuable to pause and reflect on this piece of landmark legislation. Strict enforcement of environmental rules can be dated as early as 1306-when a man was allegedly executed for burning coal in London. In the United States, concern for air quality dates back to the mid-nineteenth century, when many municipalities issued smoke ordinances. The responsibility of regulating air polluters rested almost exclusively with states and localities until 1970. The early 1970s marked an unprecedented increase in environmental awareness. During these years, the Federal government began to take a more active role in environmental regulation with passage of the National Environmental Policy Act and the Clean Air Act. Later in the 1970s, the Clean Air Act Amendments of 1977 modified these air quality regulations.

One common thread over time has been that the United States’ air quality regulatory policy, indeed environmental regulation in general, typically relies on command-and-control regulation. This type of regulation generally mandates technologies or processes, does not take advantage of the power of markets and is, therefore, by its very nature more expensive and less efficient than is necessary.

In contrast, the Bush Administration has crafted an initiative that will clean our air using a proven, market-based method. Announced on February 14, 2002, the Clear Skies Act would be the most significant and aggressive step America has ever taken, if enacted, to cut power plant emissions of three harmful pollutants – sulfur dioxide, nitrogen oxide, and mercury. The proposal, which builds upon the highly successful 1990 acid rain trading program, will cut emissions by approximately 70 percent over the next 16 years.

Clear Skies employs a dynamic approach to regulation that mandates specific emission reduction caps while providing managers with the flexibility to reduce emissions in the most efficient and least costly manner possible. Through a market-based cap and trade program, Federal emissions limits, or caps, are set and emissions permits are distributed to electricity generators. Managers then have the advantage to determine the most efficient means of action – whether it is the sale or purchase of unused allowances or banking of credits for later use. Clear Skies provides regulatory certainty and lays out the timeframes necessary for managers to design a cost-effective strategy tailored to both their current budgets and their future plans. With this structure, we uphold a principal feature of the President’s initiative – improving air quality more cost-effectively – so that Americans can continue to rely on clean and affordable electricity.

To improve air quality, Clear Skies will achieve faster reductions than the current Clean Air Act by creating incentives for “overcompliance” and innovation – power plants that develop means to reduce pollution more than or earlier than required can generate and sell extra credits. The Clear Skies Act will improve human health, visibility, and diverse range of ecosystems by reducing emissions and deposition of NOx, SO2, and mercury. In short, Clear Skies will result in dramatic progress towards solving our nation’s persistent air quality problems.

At What Cost?

As you are well aware, a crucial element of any regulatory policy is not only recognition of the benefits received from emissions reductions, but also the resource costs associated with the policy. These resource costs, it must be emphasized, are ultimately borne by citizens, whether stockholders of companies making the reductions or consumers, or both. Therefore, the Administration takes the economic modeling of Clear Skies quite seriously. In this respect, over the past several years we have gained a better understanding of the costs to abate NOx and SO2. Yet, our understanding of the removal costs associated with mercury is in a nascent stage.

The goal of Clear Skies is to reduce mercury emissions by approximately 70% from current levels by 2018 with an interim cap reducing emissions by approximately 50% by 2010. That is, mercury emissions would be reduced from current levels of approximately 48 tons to 15 tons in 2018 with an interim cap of 26 tons in 2010. Consistent with the principal of improving air qua lity cost-effectively, Clear Skies is designed to meet the Clean Air Act goal of reducing mercury with a trading program that is more cost-effective than the program currently required by the Clean Air Act. The interrelationship of cap levels for NOx, SO2, and mercury is also a key feature of Clear Skies for providing regulatory certainty, flexible capital planning cycles, and the co-benefit of mercury reductions from NOx and SO2 emission controls.

The Administration has been examining, among other things, the total resource cost of achieving the mercury reductions required under Clear Skies, the marginal cost of mercury removal, and the level of mercury co-benefits that could be expected from the NOx and SO2 limits in Clear Skies. We have also addressed what additional mechanisms and technologies will be needed to meet the 2010 mercury cap, using different assumptions and models. Major assumptions in our models have been extensively reviewed and, if necessary, updated over the past several months.

Before I share our latest results with you, I should highlight that any modeling of the effectiveness of mercury control technology is uncertain since mercury is not currently regulated in the power sector. Current modeling assumptions for mercury are based on data collected during the Environmental Protection Agency’s (EPA) Mercury Information Collection Request (ICR), pilot-scale testing, and some full-scale testing. Because the data set we are working with is evolving, uncertainties exist in how to interpret the data. For example, emissions test data collected for EPA’s ICR often reflect a large variation in mercury reduction on units with identical emissions controls and coal type burned. These differences most likely were associated with the operation of the control equipment, but additional testing continues to be conducted to understand these differences.

In general, there is agreement that selective catalytic reduction (SCR) technology provides enhancement of mercury reduction for bituminous coals. For subbituminous coal, however, there is some disagreement on whether SCR technology also provides this enhancement of mercury reduction. With only one set of test data on a subbituminous-burning unit currently available and more tests currently scheduled, this issue continues to be unclear, but more work is being done. For one of the most common coal plant configurations, a plant with a cold-side electrostatic precipitator for particulate control, the Energy Information Administration (EIA) and EPA agree that adding a SCR for NOx control and a scrubber for SO2 control will result in 90 percent of the mercury being removed from bituminous coals. For subbituminous coals, however, the assumed percent removed ranges from 27 percent for EIA to 66 percent for EPA. There is an ongoing dynamic research process sponsored by EPA, the Department of Energy (DOE), the Electric Power Research Institute (EPRI), and vendors specifically aimed at furthering our understand ing of mercury control, with new data being made available on a continuous basis.

With these uncertainties in mind, I will briefly highlight some of the empirical estimates of interest. As you may recall, the first phase mercury reduc tion cap in Clear Skies is designed to take advantage of the interrelationship of NOx, SO2, and mercury emissions. More specifically, in addition to considering economic consequences and benefits of this multiemission approach, we relied on an estimate of mercury removal achieved through installation of NOx and SO2 controls (SCR and scrubbers, respectively). This removal estimate is commonly termed “co-benefits.”

Concerning our updated empirical estimate of co-benefits, when the NOx and SO2 limits in Clear Skies are modeled without a mercury cap (i.e., without a market signal promoting mercury removal), estimates of annual mercury emissions in 2010 after installation of NOx and SO2 controls vary between 34 tons and 46 tons. An important point to understand in this context is that the mercury emissions remaining after installation of NOx and SO2 controls are most sensitive to assumptions regarding emission modification factors, or EMFs, which is the amount of mercury removal assumed when particular combinations of NOx and SO2 controls are installed. As discussed earlier, mercury reduction is dependent on coal type burned as well as the existing particulate matter, and NOx, and SO2 control devices. For example, in the Administration’s modeling, we assumed that a bituminous-burning unit with a SCR and wet scrubber can achieve 90% mercury removal. Other key assumptions including electricity demand growth, natural gas prices, and coal distribution patterns and prices, however, have not in isolation materially changed projected 2010 mercury emission levels.

While differences exist in this “co-benefit” figure, the Administration estimates the incremental costs of complying with the 2010 cap to be $650 million to $700 million per year. A key feature of understanding this cost is the safety valve mechanism in Clear Skies. This safety valve sets a maximum price of $35,000 per pound. Reducing mercury emissions to the level at which the “safety valve” would be activated – between 27 tons and 30 tons – is projected to cost between $650 million and $700 million in 2010. These costs reflect some units adding NOx and SO2 controls to enhance mercury reductions, the addition of supplemental fabric filters with activated carbon injection (ACI) (approximately 6 GW of about 300 GW of coal- fired generation), and fuel switching between coal types. Little fuel switching to natural gas is projected as a result of the incremental costs of meeting the 26 ton cap.

In sum, the President’s Clear Skies legislation calls for a 70 percent reduction in power plant emissions of NOx, SOx and mercury in the next 15 years. This legislation will meet the required health-based standards laid out under the Clean Air Act -- but it will achieve those results sooner than required and at a much lower cost to consumers. We look forward to working with the Committee and Congress to create a market-based system that will provide early reductions and affordable energy prices for consumers.

Thank you, Mr. Chairman. I look forward to answering any questions you or the members of the Committee may have.