Good afternoon. Thank you for inviting me here today to testify before you on EPA's proposed changes to the Clean Air Act's new source review ("NSR") requirements.
My name is Bill Tyndall. Since August 1998, I have been Vice President of Environmental Services for Cinergy Services, Inc., and I recently was named Vice President of Federal Affairs as well. Cinergy Services is the service company for Cinergy Corp., one of the nation's leading diversified energy companies. Its operating companies, The Cincinnati Gas & Electric Company and PSI Energy, Inc., serve more than 1.4 million electric customers and 478,000 gas customers in Indiana, Ohio, and Kentucky. Cinergy is active in U.S. power and natural gas markets and maintains a 24-hour-a-day, seven-day-a-week trading operation. The company's international business unit, Cinergy Global Resources, has assets in power generation, transmission, and distribution projects in the Czech Republic, Spain, the United Kingdom, Zambia, Estonia, and the United States. Cinergy's 1999 revenues were $5.9 billion, and its total assets are $9.6 billion. Cinergy's core energy system comprises approximately 11,000 megawatts at 14 baseload stations and seven peaking stations. Its natural gas distribution system is connected to six interstate pipelines.
Before joining Cinergy, I served as minority counsel to the House Commerce Committee and advised Committee Democrats on air quality issues. Before that, I spent five years at EPA, serving first in EPA's Office of General Counsel, where I worked on new source review and other stationary source issues, and later as a senior policy advisor in EPA's Office of Air and Radiation, the office responsible for administering the NSR program.
Today I am here on behalf of Cinergy, a company with nearly three decades of experience under EPA's regulatory treatment of NSR. Thus, I am speaking as someone who has spent nearly ten years working with EPA's new source review program from a variety of perspectives. My testimony also is on behalf of the Edison Electric Institute, an association of investor-owner electric utilities such as Cinergy. I will be addressing what I believe to be the serious ramifications of EPA's attempt to reform the Clean Air Act's new source review program.
A series of summer heat waves and steadily rising consumer demand have forced many utilities to the limits of their generating capacity. With the industry's equipment pushed to the breaking point for extended periods, the nation as a whole faces a risk of electricity shortfalls that is higher than ever before. As these trends continue, the need to keep electric utilities running, and running reliably, is at its apex. The availability of power in America depends on the ability of utilities to continue maintaining their facilities in the manner needed to ensure safe, efficient, and reliable generation on demand. To exacerbate the situation, electric utilities, rural cooperatives, municipal electric systems, and independent power producers are all facing significant obstacles in siting and building needed additional peaking capacity.
In the midst of these potential electricity shortfalls, EPA now proposes changes to the Act's NSR program that could require existing facilities to undergo an expensive and time-consuming permitting process before they undertake any activities intended to maintain safety, availability, and reliability. A close examination of EPA's proposal shows not only that it is contrary to Congress' focus on new sources of pollution, but that it is inconsistent with how the rule has been written and applied for nearly three decades. Moreover, EPA's plan to promulgate an unworkable rule that discourages or delays needed maintenance projects is at odds with maintaining the reliability of the nation's power supply.
Background
Electric utilities occupy a unique position in the industrial world. We arguably are the ultimate example of "just-in-time delivery" of a product to our customers. Because it is not feasible to store significant amounts of power, we must generate electricity at the very instant that our customers consume it. This requires constant and careful maintenance of our generating units, which are comprised of thousands of individual components working together as an integrated system. In this interdependent environment, the failure of a single component is sufficient in many cases to cause an entire generating unit to be shut down and require repair.
Furthermore, utilities are operated under extreme conditions of temperature, pressure, and wear that make such failures particularly likely. As in an automobile, or any other highly-integrated piece of equipment, these various parts wear at different rates, with the result that parts both large and small must be replaced on a periodic basis in order to keep the unit running properly. In contrast, the failure to make such repairs results in rapid and predictably declining reliability and unit availability. At present, Cinergy operates over 60 individual generating units in our three-state system, and the maintenance required to keep these facilities operating smoothly has been, and remains, a daunting task.
Note that a failure to maintain generating units properly results not only in decreased performance, but also can cause unsafe conditions for our employees, as well as our customers. The early history of steam generation was plagued by equipment failures, with many injuries to plant employees. Since that time, the American Society of Mechanical Engineers (ASME) and other industry authorities have developed detailed codes that guide utility maintenance and repair activities to ensure that generating units may be operated safely and reliably for decades. In addition, various state agencies and insurance underwriters regulate boiler operation, maintenance, and repair practices to ensure utilities maintain their equipment properly.
These generating facilities are subject to a host of Clean Air Act provisions that constrain emissions to levels that protect the public health and welfare. For electric utilities, this includes, but is not limited to, (1) compliance with SIP-based limitations designed to achieve or maintain the national ambient air quality standards ("NAAQS"); (2) restrictions on NOx and SO2 emissions under the Act's Title IV acid rain program, including a more stringent phase two of that program which commenced on January 1, 2000; (3) restrictions on ozone, SO2, and particulate matter under Title I of the Act; and (4) EPA's Title V operating permit program. In this manner, Congress has ensured that all industrial facilities both old and new are subject to extensive and costly pollution control requirements. In addition, state environmental programs impose additional emission limitations that apply to our plants.
In 1970, and again in 1977, Congress enacted significant amendments to the Clean Air Act targeted at new sources of pollution. Specifically, Congress amended the Act to provide that companies that construct new facilities, or make "major modifications" to existing facilities that result in significantly increased emissions, must apply an extra layer of pollution control to these units. As EPA has recognized, Congress targeted new construction and the extensive alteration of existing units because it understood that it is more feasible technically, and less disruptive economically, for companies to install new control technology at the time these events occur than it is to retrofit existing units.
Under the terms of the NSR program, new and "modified" units must satisfy "new source performance standards," and install controls that represent "best available control technology" or "lowest achievable emission rate" requirements. Such units also must undergo review for their impact on ambient air quality either under the "prevention of significant deterioration" program (for areas where air quality is cleaner than the NAAQS require) or under the nonattainment program (for areas where one or more NAAQS is not being achieved). Thus, a "major modification" to an old plant can create a "new" source for regulatory purposes and trigger the congressional mandate for the plant to undergo a permitting process that takes 18 months or longer to complete and results in millions of dollars in control equipment costs. While Congress may have considered this a reasonable process for a "major modification," it is clear that such a process and cost cannot be imposed on routine maintenance and repair activities if the U.S. is to maintain a safe, reliable, and reasonable supply of electric generation to homes and businesses.
EPA's Historical Treatment of the Modification Rule
Given this history, the question of whether a source has undergone a "major modification" for purposes of NSR is a crucial one for older generating units. Historically, EPA has interpreted the modification rule in a manner consistent with Congress' focus on new sources of pollution and its concern about the costs of retrofits. For example, EPA guidance recognizes that Congress "did not intend to make every activity at a source subject to new source requirements" via the modification rule, and that EPA "in no way intends to discourage physical or operational changes that increase efficiency or reliability or lower operational costs, or improve other operational characteristics of the unit." 57 Fed. Reg. 32,327. Similarly, EPA has expressed concern with requirements that would "unduly hamper the ability of any company to take advantage of favorable market conditions." 45 Fed. Reg. at 52,704. For these reasons, EPA has always excluded "routine maintenance, repair and replacement," as well as increases in production rate or hours of operation within a facility's capacity, from the definition of a modification.
EPA's actual application of the modification rule to utilities also has been consistent with both congressional intent and the agency's regulatory pronouncements. Specifically, EPA has allowed utilities to make those repairs that are customarily undertaken in the industry to maintain the availability and reliability of electric generating facilities, and to thereby maximize the useful lives of these units, without any suggestion that such repairs were non-routine modifications subject to NSR. Importantly, EPA adopted this approach while armed with an extensive agency awareness of utility maintenance practices from on-site inspections, reports filed with state and federal regulatory agencies, and countless industry articles.
There is one case where EPA has found that repair and replacement activity crossed the line between those routine projects that keep a facility operating, and a non-routine expansion of the facility subject to NSR. In a 1989 administrative decision, EPA ruled that a plant-wide reconstruction project at Wisconsin Electric's ("WEPCo") Port Washington facility was a non-routine change because it involved a "massive" and "unprecedented" replacement of major components, some of which had never been replaced before by WEPCo or other utilities. By pursuing the WEPCo project as a non-routine modification, after years of allowing less ambitious repair and replacement projects that extended unit life, EPA confirmed that the majority of utility maintenance projects to extend life qualified as routine maintenance and did not trigger the modification rule. EPA later confirmed this to Congress, leading GAO to report later in 1990 that:
According to EPA policy officials, WEPCo's life extension project is not typical of the majority of utilities' life extension projects, and concerns that the agency will apply the ruling it applied to WEPCo's project are unfounded. (emphasis supplied)
In keeping with its statements to Congress, EPA has continued to allow utilities to undertake repair and replacement projects as needed to maximize unit life in the ten years following the WEPCo decision, without any finding that such projects triggered NSR.
EPA's Proposal to Revoke the WEPCo Rule
However, in Federal Register notices issued in July 1996 and 1998, EPA has proposed a major revision to the NSR modification rule. I refer you to 61 Fed. Reg. 38,250 and 63 Fed. Reg. 39,857. In these notices, EPA essentially proposes to revoke parts of a 1992 rule EPA issued to formalize its WEPCo decision (the so-called "WEPCo Rule"). EPA would replace this WEPCo Rule with a new interpretation of the modification standard under which nearly all activities intended to maintain or restore the reliability or efficiency of a generating unit would trigger NSR.
EPA's original modification rule is fairly straightforward, and defines a "modification" as a: (1) physical or operational change (2) that results in (3) a significant net emissions increase. The rules then provide examples of activities that do not constitute a physical or operational change, including routine repair, replacement and maintenance, pollution control projects, fuel switches, and so on. See 40 C.F.R. 52.21(b)(2), 60.14. As I mentioned earlier, this rule historically has resulted in a regulatory program where utilities are allowed indeed encouraged -- to maintain and operate their plants as they were designed to operate.
EPA now seeks to change how each of these criteria is applied to determine whether a modification has occurred. First and most important is EPA's new approach to what constitutes a "physical or operational change." As described above, EPA has always excluded traditional utility maintenance activities, whether performed singly or grouped together, from the definition of a physical or operational change under the exclusion for "routine maintenance, repair and replacement."
EPA now proposes an interpretation of the "routine maintenance, repair and replacement" exclusion under which all activities that restore deteriorated capacity and availability must be treated as "non-routine" repairs. This would mean that all utility maintenance projects would be "non-routine," since even the simplest utility maintenance project is intended to restore lost availability and reliability. This again is in direct conflict with the history discussed earlier. It also is logically inconsistent with the language of EPA's exclusion for "routine maintenance, repair and replacement." If all repairs are non-routine, there can be no such exclusion. EPA would, in effect, read this exclusion out of the NSR regulations.
Another major change is EPA's elimination of the causation requirement of the modification rule, and its removal of the so-called "demand growth" exclusion, which excludes increases in production rate and hours of operation. EPA created this latter provision based on its understanding that Congress did not in any way intend for the modification rule to punish utilities for responding to changes in demand growth within the design capabilities of their facilities. Thus, under the rule as written, a physical or operational change cannot be considered a "modification" unless the change "results in" a significant increase in emissions. In other words, the mere fact that there is an increase in net annual emissions after a particular change does not necessarily trigger NSR. If the increase was caused not by the change but by something else for example, by an increase in production rate or hours of operation, occurring in response to an increase in product demand or some similar market force -- a "modification" has not occurred. In this regard, EPA stated in the preamble to the 1992 WEPCo Rule that it "declines to create a presumption that every emissions increase that follows a change in efficiency is inextricably linked to the efficiency change." 57 Fed. Reg. at 32,327.
EPA now proposes a regulatory regime under which any non-routine (or otherwise non-excluded) change that is followed by a significant increase in annual utilization of the facility would be presumed to be a modification, even if that change does not affect a facility's overall capacity to burn fuel (and hence its emission rate). According to EPA, the Agency's "experience":
leads to the conclusion that sources generally make non-routine physical or operational changes which are substantial enough that they might trigger NSR in order to increase reliability, lower operating costs, or improve operational characteristics of the unit and do so in order that they may improve their market position. A proximate cause for making such changes may be to respond to increased demand, or to more efficiently compete for share of a market that has flat, or even decreasing, demand.
63 Fed Reg. 39,860.
"For these reasons," the Agency continues, "EPA now seriously questions whether market demand should ever be viewed as a significant factor in answering the relevant regulatory question of whether an emissions increase results from a physical or operational change at an existing source." Indeed, according to EPA, "in a market economy, all changes in utilization and, hence, emissions might be characterized as a response to market demand." Id. In this manner, EPA's proposal appears to require that any change that is followed by a significant increase in annual utilization is a "modification" and, therefore, subject to NSR.
EPA's final change is to redefine how an emissions increase is calculated. EPA claims that "[u]nder current regulations," non-routine activity at a unit must be "deemed to be of such significance that 'normal operations' are deemed not to have begun" at the facility. In other words, EPA would treat a company that undertakes a non-routine change at a unit as if the unit had never been operated before and had no emissions history. This change is significant because the NSR rules provide that, for a unit that has not begun "normal operations," post-change emissions must be calculated based on the assumption that the unit will be operated at full tilt, 24 hours a day, 365 days a year after the change. Under this test, any change will show an increase in emissions, even if the change would not truly result in any new pollution.
But the Seventh Circuit specifically rejected this view of the emissions increase test in its review of EPA's WEPCo determination. See Wisconsin Elec. Power Co. v. Reilly, 893 F.2d at 917-18. Moreover, EPA rejected this interpretation in its 1992 WEPCo Rule, explaining that the Agency is required to compare actual emissions before and after a change, unless the unit is brand new or has been entirely rebuilt by spending 50% of the cost of a new facility. See 57 Fed. Reg. at 32,317, 32,323.
When one considers these changes together all repairs are non-routine, all non-routine repairs are deemed to result in an emissions increase, regardless of whether they actually did so it becomes clear that EPA plans to subject the entire electric utility industry to a new level of controls not contemplated by the original NSR program.
Summary
If EPA's goal is to obtain new source levels of emission reduction from existing sources even though those sources have already reduced emissions to address the public health and welfare requirements of the Clean Air Act then let's address this directly as a policy issue. The new source review program was never meant to require such sweeping reductions from existing utilities, and is uniquely ill-suited as a vehicle for obtaining them. In particular, EPA's interpretation of NSR would require all utilities to delay repair, and possibly shut down their facilities, for 18 months or more every time there is a minor equipment failure in order to go through the NSR process. There simply isn't enough capacity in the nation to allow for so many units to remain inactive for such extended periods, nor would consumers appreciate the jump in rates that would accompany such a change.
EPA's proposed revision to NSR is problematic for other reasons as well. EPA's proposed NSR approach would force utilities to guess on when to install new pollution controls according to the unpredictable failure of minor pieces of equipment, rather than choosing a time that minimizes outages and technical difficulties the very problem Congress sought to avoid in creating NSR. Utilities faced with the prospect of undergoing NSR every time they replace broken turbine blades or boiler tubing would be discouraged from making such changes for as long as possible, with serious ramifications for the reliability of the nation's power supply and for the safe operation of the units. Degradation in reliability is counter to the expectations of state public utility commissions, which can and have financially penalized companies for substandard performance. Finally, for many utilities, EPA's NSR proposal also would mean foregoing the adoption of new technologies that could increase the efficiency of generating units by reducing the amount of coal needed to produce power. By discouraging such opportunities, EPA would be losing a chance to obtain significant reductions in pollutants not currently regulated under the Act, in particular, greenhouse gases.
In sum, the utility industry already faces many challenges to its ability to provide safe, reliable, and affordable power to the American public. EPA's proposed changes to the NSR modification rule should not be among them. The ongoing rulemaking process needs to be conducted in a manner that will truly reform the NSR program, rather than the piecemeal approach currently being taken by the Agency. It is now time for EPA to abandon such impractical approaches to NSR in favor of straightforward negotiations with industry, and for Congress to address the important policy issues raised in the current debate. Regardless, EPA's proposed NSR rule should not proceed on its current track.
I thank you for the opportunity to present these comments, and would be happy to respond to any questions.