AIR POLLUTION Improvements Needed in Detecting and Preventing Violations -_ CL Ji II 142597 RELEASED RESTBI~--Not to be released outside the General Accounting Office unless specifically approved by the Offlce of Congressional Belations. 54% I!@ Resources, Community, and Economic Development Division B-233666 September 27,199O The Honorable John D. Dingell Chairman, Subcommitteeon Oversight and Investigations Committee on Energy and Commerce Houseof Representatives Dear Mr. Chairman: This report respondsto your request that we examine the Environmental Protection Agency’s (EPA) efforts to ensure that major stationary sourcesof air pollution do not exceed air pollution control requirements. The report discussesEPA’S efforts to (1) require the use of the most effective methods to detect air pollution violations at major sourcesand (2) impose appropriate enforcement actions when violations are found. Unless you publicly releaseits contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copiesof the report to appropriate congressionalcommittees; the Administrator, EPA; and other interested parties. We will make copies available to others upon request. This work was performed under the general direction of Richard L. Hembra, Director, Environmental Protection Issues(202) 276-6111.Major contributors to this report are listed in appendix I. J. Dexter Peach Assistant Comptroller General Executive Summ~ Exposure to unsafe levels of air pollution has been linked to incidences Purpose of cancer, lung disease,and other health problems, yet over one-third of the nation’s population live in areas that exceededone or more federal standards for air quality. The Environmental Protection Agency (EPA) estimates that nearly 40 percent of air pollution comesfrom about 33,500 of the nation’s major stationary sourcesof pollution such as elec- tric utilities, oil refineries, steel mills, and large factories. Concernedabout pollution from major stationary sources,the Chairman, Subcommittee on Oversight and Investigations, HouseCommittee on Energy and Commerce,requested that GAO determine whether EPA (1) usesthe most appropriate method for detecting violations at major sta- tionary sourcesand (2) ensuresthat appropriate enforcement actions are taken when violations are found. Each major source has the potential to emit 100 tons or more of pollu- Background tants into the air annually, and somecan emit that much daily. In light of the potential health risk posed by air pollution, the Clean Air Act provides for a federal/state partnership under which states develop, and EPA approves, State Implementation Plans specifying actions to reg- ulate and control these sourcesof air pollution, including the issuanceof permits specifying acceptableemission levels. EPA and authorized state and local agenciesare responsible for detecting, and then abating, viola- tions and for deterring their future occurrence.During 1989, these agen- cies detected violations of permit or other requirements at about 14 percent of the major stationary sources. Two primary methods are used to detect violations: on-site inspections and emission monitors. On-site inspections, often lasting less than half a day, are periodic assessmentsof facility compliance that may include visual observation, review of operating equipment and records, and a check of pollution control equipment. Emission monitors-automated mechanical equipment usually placed in a facility’s exhaust or smoke stack-provide direct, accurate measurementsof emissionsand can operate up to 24 hours per day. EPA estimates that emission monitors are 10 times more likely to detect Results in Brief air quality violations than on-site inspections becausethey measure Y emissionsdirectly, provide nearly continuous coverageof facility opera- tions, and detect violations that inspectors cannot. Although EPA issued a policy aimed at increasing the use of monitors where feasible, it has Page 2 GAO/RCED-90-155 Air Pollution Violations Executive Summary not implemented this policy by developing regulations that (1) establish criteria for determining where monitors are feasible and (2) require monitor use at sourcesmeeting the criteria. As a result, monitors have been installed at only about 1,065, or 11 percent, of the sourceswhere EPA estimates that monitors could be installed. In order to better address the nation’s acid rain problems, proposed Clean Air Act amendments call for the use of monitors at 1,100 sources,principally large utilities that emit large quantities of sulfur dioxide, but someof these sources already have monitors installed and EPA has no plans to require addi- tional monitors beyond those called for in the legislation. When violations are detected, EPA favors cash penalties calculated to offset the economicbenefit that violators gained by not complying with air quality requirements. However, over half of the fiscal year 1988 and 1989 violations defined as significant--EPA’s highest enforcement pri- ority-were concluded with no cash penalty imposed. This occurred becausemost enforcement actions against major stationary sourcesare taken by state and local programs which, for the most part, operate under their own enforcement authority. Proposedchangesto the Clean Air Act should improve EPA'S ability to increasestate and local penalty amounts and encouragestate and local agenciesto adopt economicben- efit penalty policies. Principal Findings More Extensive Use of Conditions have changedsince the 1970swhen the concernsof industry Emission Monitors and regulators over monitor cost and reliability causedEPA to limit its emission-monitoring regulations to the newest and biggest major station- Warranted ary sources.Costs for monitors have becomemore reasonablefor many major sources,to the point where monitor costs are small relative to other pollution control equipment, and monitor reliability has greatly improved. Monitors in Pennsylvania, for example, have collected reli- able data over 90 percent of the time. Becauseof these changedcondi- tions, and the greater detection potential of monitors as compared with inspections, EPA announceda policy in 1988 calling for monitor usage wherever feasible. EPA, however, has not issued regulations implementing this policy nor has it developed criteria for making such feasibility determinations. Instead, EPA has tried to convince state and local programs to Page 3 GAO/RCED-90-155 Air Pollution Violations Executive Summary voluntarily require the use of monitors under their own authorities, but with little success.As a result, only about 1,065 major stationary sourceshad monitors installed by the end of 1989. Senior EPA compli- ance officials estimate that monitors may be feasible at about 9,000 more major stationary sources,but they acknowledge the absenceof clear criteria for evaluating sourcesfor monitor feasibility. Although proposed amendmentsto the Clean Air Act call for EPA to require monitors at about 1,100 major sulfur dioxide sources,someof these sourcesare among those that have already installed monitors. EPA has no plans to require additional monitors beyond those called for in the proposed legislation, citing insufficient resourcesto addressnew and emerging air quality problems and, at the sametime, develop criteria and regulations to implement its emission-monitoring policy. Neverthe- less,becauseemission monitors represent a significant improvement in ensuring compliance, GAO believes that EPA development of such regula- tions is warranted. Penalties Often Both the Clean Air Act’s administrative penalty authority and EPA'S own Insufficient to Deter internal civil penalty settlement policy direct agency enforcement offi- cials to seek cash penalties sufficient to remove the economicbenefit to Violations noncomplying sources.However, in fiscal years 1988 and 1989, no cash penalties were assessedin more than half of the significant violator casesat major stationary sourcesbecauseEPA, in its review and approval of State Implementation Plans, has not required state and local agencies-which conduct the majority of enforcement actions-to assesspenalties sufficient to eliminate the economicbenefit. GAO'S review of eight state and local programs disclosedthat none regularly sought to recover economicbenefit penalties. For example, GAO found that the $15,000 penalty assessedin one casewas $200,000 lower than the estimated economicbenefit gained by the violating source. Not col- lecting sufficient penalties in such casesmay place sourcesthat comply at an economicdisadvantage to violators. According to EPA, the primary reason that state and local programs gen- erally do not assesseconomicbenefit penalties is that they often seek only to correct the causeof the violation. EPA can “overfile” and impose larger penalties in such cases,but has generally not done so, citing unclear authority and insufficient resources.ProposedClean Air Act amendmentswould broaden EPA'S current authority to administratively assesspenalties and would allow it to overfile in more casesto ensure that economicbenefit penalties are sought. Nevertheless,EPA needsto Page 4 GAO/RCED-90.155 Air Pollution Violations Executive Summary take actions to better ensure that state enforcement programs seek eco- nomic benefit penalties when violations are detected so that inequities do not occur between firms that comply and those that do not. Currently proposed amendmentsto the Clean Air Act contain provisions Recommendationto that would broaden EPA'S administrative penalty authority to enable it the Congress to better ensure that appropriate cash penalties are assessed.GAO sup- ports these proposals and recommendsthat the Congressinclude such penalty authority in the final Clean Air Act legislation. GAO recommendsthat the Administrator, EPA Recommendationsto the Environmental l implement EPA'S emission-monitoring policy by developing regulations Protection Agency that (1) establish criteria for determining where monitors are feasible and (2) require monitor use at all sourcesmeeting the criteria and . require EPA enforcement staff to (1) use available enforcement authority to overfile to the maximum extent possible when states assessinade- quate penalties and (2) undertake efforts to include specific standards for assessingeconomicbenefit penalties in the next revisions to State Implementation Plans. GAO discussedthe report’s contents with appropriate EPA, state, and Agency Comments local officials, and their comments have been incorporated as appro- priate. As directed by the SubcommitteeChairman, GAO did not obtain official agency comments on a draft of this report. Page 5 GAO/RCED-90-155 Air Pollution Violations Contents Executive Summary Chapter 1 8 Introduction EPA, State, and Local Agency Rolesin Controlling Stationary SourceEmissions 12 Ensuring Compliance Through Detecting and Deterring 12 Violations Objectives,Scope,and Methodology 13 Chapter 2 17 Regulations Needed to Detection Capability of Emission Monitors Not Enough Emission Monitors in Use 17 20 Require Greater ConcernsRegarding Monitor Reliability and Cost 22 Emission Monitor Use EPA’s Monitor Use Policy Not Implemented 27 at Major Stationary Conclusions 29 Recommendation 30 Sources Chapter 3 31 Penalty Amounts Not Economic Benefit Penalties Established to Promote Compliance 32 Sufficient to Deter Economic Benefit Penalties Not Imposed by State and 33 Violations Local Programs EPA Has Made Little Effort to Increase Penalties 36 ProposedChangesto Clean Air Act May Improve 40 Penalties, but Clearer Direction Needed Conclusions 41 Recommendationto the Congress 42 Recommendationsto the Environmental Protection 42 Agency Appendix Appendix I: Major Contributors to This Report 44 Tables Table 1.1: State and Local Offices in EPA RegionsIII, IV, 14 and IX Included in GAO’s Review Table 3.1: Resolution of State and Local Enforcement 33 Actions for Significant Violator Cases,Fiscal Years 1988and1989 Page 0 GAO/RCED-99-155 Air Pollution Violations Contents Figures Figure 1.1: Contribution of Major Stationary Sourcesto Air Pollution Nationwide Figure 1.2: Major Stationary SourcesReleasingVisible 10 Emissions Figure 1.2: Continued 11 Figure 2.1: Location and Operation of a Typical Opacity 18 Monitor Abbreviations CEM continuous emissionsmonitors EPA Environmental Protection Agency GAO General Accounting Office NfiQs National Ambient Air Quality Standards NESHAPS National Emissions Standards for HazardousAir Pollutants NO, nitrogen oxides NSPS New SourcePerformance Standards RACT ReasonablyAvailable Control Technology SIP State Implementation Plan so2 sulfur dioxide VOC volatile organic compound Page 7 GAO/RCED-90-155 Air Pollution Violations Chapter 1 Introduction As the Congressconsiderssignificant changesto the Clean Air Act, con- trolling air pollution continues to be a difficult task, Although signifi- cant progress has been made in reducing air pollution, the Environmental Protection Agency (EPA) reported in 1990 that more than 120 million Americans lived in areasthat exceededone or more federal standards for acceptableair quality. Increased incidencesof cancer, lung disease,and other illnesseshave been linked to air pollution. Whether the air we breathe is safe or not dependson many factors, but principally, it dependson the amount of emissionsfrom over 170 million mobile sourcesand an estimated 300,000 stationary sourcesof air pollu- tion Mobile sources,such as cars, trucks, buses,and aircraft, are believed to account for nearly half of the emissionsof the six criteria pollutants, * including over two-thirds of the carbon monoxide emissions and nearly one-third of the volatile organic compounds(WCS). Station- ary sourcessuch as electric utilities, factories, and commercial buildings are responsible for most of the remaining emissions.Thesesourcesemit 96 percent of the sulfur dioxide (so,),68 percent of the nitrogen oxides (NO,.., and 62 percent of the particulate matter, as well as contributing about 61 percent of total voc emissions-a major contributor to the for- mation of ozone,generally referred to as “smog.” Emissions from most mobile and stationary sourcesare difficult to mon- itor and control becauseof their large numbers and diversity of opera- tions. However, major stationary sources-comprising about 33,600 of the nation’s largest facilities2 -are more readily monitored and con- trolled. These sourcesinclude electric utilities, oil refineries, steel mills, and large factories and, according to EPA, account for nearly 40 percent of the emissionsof criteria pollutants nationwide and nearly 86 percent of all stationary source emissions.Figure 1.1 shows the relative contri- butions of major stationary sourcesto the nation’s air pollution problems. ‘EPA establishedNational Ambient Air Quality Standardsfor six health-relatedpollutants-ozone, carbonmonoxide,sulfur dioxide, nitrogen dioxides,particulates,and lead.Thesepollutants are com- monly called “criteria pollutants.” ‘The term “facility” is usedthroughout this report to describea majorstationary sourceof air pollution. Page 8 GAO/RCED-90-165 Air Pollution Violations Chapter 1 l.utroduction Figure 1.1: Contrlbutlon of Major Stationary Sources to Air Pollution Nationwide r Mobile Sources Other Sources Stationary Sources Source: GAO illustration based on EPA data. According to EPA officials, control of emissionsat major stationary sourcesis especially important becauseof their ability to releasegreat quantities of pollutants if uncontrolled. Major stationary sourcescan releasemore pollution into the air in one day than many other sources can in one year. Every major stationary source,by definition, has the potential to emit 100 tons of pollutants, or more, annually and some facilities can emit that much pollution daily. For example, so, emissions from one large electric utility in Pennsylvania can reach 76,000 tons annually, or over 206 tons each day. Figure 1.2 shows two major stationary sourcesproducing visible emissions. Page 9 GAO/RCED-90-155 Air Pollution Violations Chapter 1 Introduction The Clean Air Act provides for a federal-state partnership to be used in EPA, State, and Local addressingenvironmental problems. The act requires EPA to set national Agency Roles in standards for air quality and provides for states and localities to assume Controlling Stationary the responsibility of designing and implementing control strategies to meet these standards. These control strategies are documented in the Source Emissions State Implementation Plans (SIPS). EPA is responsible for reviewing and approving the SIPS to ensure that they are adequate to attain and main- tain compliance with national air quality standards, and for overseeing state and local implementation of these plans. In caseswhere state or local plans are not approved, EPA promulgates a federal implementation plan for the affected area. States and localities can propose revisions to SIPS for EPA approval at any time, as long as the revised plan will still attain or maintain the national standards. An essential component of SIPS is the issuanceof permits specifying acceptableemissionslevels to owners and operators of major stationary sources.EPA and authorized state and local agenciesare responsible for detecting violations of these permits and other requirements and, when a violation is discovered, for taking appropriate action to bring the facility back into compliance and to deter future violations. EPA relies heavily on state and local programs to detect and deter violations at major stationary sources.In 1988 and 1989, EPA authorized 107 state and local agenciesto carry out provisions of the act. Under EPA approval, funding assistance,and oversight, these agenciescarry out over 90 percent of the actions designedto detect and deter violations at stationary sourcesof air pollution. For example, state and local agencies performed 34,263 of 37,716 inspections, or about 91 percent of all inspections performed, in fiscal year 1988. Ensuring Complia& EPA, state, and local strategies for controlling stationary source emis- sions rely, in part, on educating facility owners and operators through Through Detecting the issuanceof regulations and guidance, coupled with technical assis- and Deterring tance and site-specific permits that translate and tailor these require- ments to an individual source’soperations. However, once permits Violations spelling out how pollution is to be controlled at individual facilities have been issued, regulators generally rely on two methods to ensure that sourcesmaintain compliance:3 3EPAalsousesstackteststo assesscompliance;however,thesetestsare not considereda primary detectionmethodbecausethey are scheduledby the facility, employcontractorspaid by the facility, and are performedinfrequently, often only onceevery 6 years during permit application or renewal, accordingto EPA. Page 12 GAO/RCED-90-165 Air Pollution Violations Chapter 1 Introduction l periodic on-site inspections, which compare facility emissionsand prac- tices with permit levels and conditions, and l emission-monitoring devices,also known as “continuous emissions monitors” (CEMS), which are mechanical instruments that measurethe amount of pollutants leaving a facility on a near-continuous basis. Section 114 of the Clean Air Act gives EPA the authority to use either or both detection methods, in addition to others, to assessthe compliance of any stationary source.The act doesnot require or prefer one detec- tion method over another, but provides regulators with broad authority to inspect, monitor, test, sample, and review any facility operations that may reasonably be required to assesscompliance. Onceviolations are detected, bringing the facilities back into compliance and deterring future violations largely dependson regulators taking appropriate, timely actions. In 1977, the Congressadded section 120 to the act, which provided EPA with an enforcement mechanism designed specifically to remove financial incentives to noncompliance.In adding this provision to the act, the Congressintended for noncomplying major sourcesto be penalized an amount equal to the economicvalue of the noncomplianceto the polluter. Otherwise, sourcesnot complying were viewed as obtaining a competitive advantage over firms that had already installed the neededcontrol equipment or changedproduction processesto meet their emission limitations. This penalty was to be assessedin all caseswhere EPA found noncomplianceto exist and can be waived only if the violation is de minimis or due to specific causeslisted in the statute. Concernedwith major stationary sources’potential to significantly con- Objectives, Scope,and tribute to unsafe levels of air pollution when operating in violation of Methodology the Clean Air Act’s requirements, the Chairman, Subcommitteeon Over- sight and Investigations, HouseCommittee on Energy and Commerce, requested that we determine l whether EPA usesthe most appropriate method of detecting violations at major stationary sourcesof air pollution and . once discovered,whether EPA takes appropriate action to compel compli- ance at these sites. To accomplish these objectives, in accordancewith agreementswith the Chairman’s office, we performed work at the following EPA headquar- ters offices: Page 13 GAO/RCED-90-156 Air Pollution Violntionn ., Chapter 1 Introduction l Stationary SourceCompliance Division, Office of Air Quality Planning and Standards, Office of Air and Radiation. l Office of Air Enforcement Counsel,Office of Enforcement and Compli- ance Monitoring. l Legal Enforcement Policy Branch, Office of Enforcement and Compli- ance Monitoring. l Office of the Assistant Inspector General for Audit, Office of the Inspector General. We also visited three EPA regions and eight authorized state and local programs within these regions as shown in table 1.1. Table 1.1: State and Local Offices in EPA Regions III, IV, and IX included in GAO’s EPA region State agency Local district Review Region III Pennsylvania Department of Environmental Resources Region IV North Carolina Mecklenberg County Department of Division of Environmental Protection Environmental Management Region IX California Air Bay Area Air Quality Management District Resources Board Arizona Department Maricopa County Bureau of Air Pollution of Environmental Control Quality, Office of Air Quality Pima County Air Quality Control District Wejudgmentally selectedthese regions, states, and localities for geo- graphical coverage,program size, and diversity of sources,and because each had a significant number of the more than 120 million Americans living in areas that exceededone or more of the nation’s air quality stan- dards in 1988. For the agenciesselected,we performed work at each organizational level responsible for establishing, interpreting, and imple- menting EPA policies for detecting and deterring air pollution violations. Region III was selectedbecause,according to EPA, it contains someof the largest industrial sourcesin the nation, someof which are located in Pennsylvania. In addition to program size, Pennsylvania was selected because,according to EPA, it has one of the most advancedemission- monitoring programs in the nation. RegionIV is EPA'S largest region, responsible for overseeing8 states, 17 local programs, and over 5,400 major stationary sourcesin 1988. North Carolina, with 1,462 major sources,has the largest single portion of this universe. Region IX was selectedto provide west coast geographical coverageand also because Page 14 GAO/RCED-99-155 Air Pollution Violations Chapter 1 Introduction California has a large population potentially exposedto unsafe levels of pollutants. To addressour first objective, we reviewed agencyregulations, policies, and selectedsource files, and interviewed air program officials in EPA headquarters and the above regions, states, and localities to determine the methods available for detecting violations, the successfulnessand extent of the use of each, and the benefits and trade-offs of each method. We gave particular emphasisto Pennsylvania’s use of emission- monitoring devicesbecauseEPA points to this state as having one of the most advancedprograms for both detecting and deterring violations using continuous emissionsmonitors, and to Ohio since this state’s CEM program was cited as one of the more comprehensiveprograms in EPA’S 1988 Continuous Emission Monitoring policy. We also gave particular emphasis to EPA’S stationary source inspection program becausethe effectiveness of this surveillance method was questioned in an earlier GAO report,4 and becausethis is still the principal method used by EPA, state, and local programs to detect violations. Additionally, we discussed emission-monitoring technology with equipment manufacturers and selectedusers. To addresswhether EPA and the states take appropriate action to ensure compliance at noncomplying major stationary sources,we interviewed EPA headquarters and regional compliance and enforcement program officials, as well as those in the states and localities listed above and obtained documents,where possible, describing their enforcement responsepolicies and practices from January 1986 to the present. We compared and contrasted these policies and practices regarding major sourceswith Congress’intent in adding enforcement provisions to the 1977 act, and with EPA'S national policies for the appropriate, timely res- olution of violations. In addition, we judgmentally selectedand reviewed enforcement casefiles with violations identified from January 1, 1986, to December3 1, 1987, to illustrate EPA and EPA-approvedenforcement policies and practices at work. We also reviewed relevant reports issued by EPA'S Office of the Inspector General. Our work was performed from February 1988 through December1989. In conducting our work, we reviewed EPA’S1987 and 1988 Federal Man- agers’ Financial Integrity Act reports to the Congressand the President for previously reported internal control weaknesses.Where possible, we 4Air Pollution: EnvironmentalProtectionAgency’sInspectionsof Stationary Sources (FADIR6GD _85- 1BR, Oct. 24,1986). Page 15 GAO/RCED#O-155 Air Pollution Violations Chapter 1 Introduction sought to identify the underlying causesof problems we found in EPA'S programs for detecting and deterring Clean Air Act violations and the associatedmanagementcontrols that should have prevented these problems from occurring. The views of EPA officials responsible for overseeingcompliance and enforcement activities at major stationary sourceswere sought during our review and are incorporated into the report where appropriate. However, as requested by the Chairman’s office, we did not obtain offi- cial agency commentson a draft of this report. Except as noted above, our work was performed in accordancewith generally acceptedgovern- ment auditing standards. Page 16 GAO/RCED-M-156 Air Pollution Violations Chapter 2 Regulations Neededto Require Greakr Emission Monitor Use at Major Stationary Sources EPA compliance officials estimate that continuous emission monitors are 10 times more likely to detect air pollution violations than on-site inspections becausethey measure emissionsdirectly, provide near- continuous coverageof facility operations, and detect violations which inspectors cannot. Although concernsexisted over the reliability and cost of CEMS, their reliability has improved and the costs have become more reasonable.Monitors have been shown to be reliable over 90 per- cent of the time, and while costs may continue to be a concern for some, the economicburden of CEMS doesnot appear substantial for many major sources,and somesourcesmay be able to reduce operating costs by using fuel more efficiently. In 1988, recognizing that conditions for CEMSuse had becomemore favorable, EPA announceda policy encouraging the installation of monitors at any major stationary sourceswhere their use is estimated to be feasible. However, EPA has not issued regulations that (1) establish criteria for determining where CEMuse is feasible and (2) require CEM use at sourcesthat meet the feasibility criteria. As a result, monitors have been installed at only 1,066 facilities, or about 11 percent of the 10,000 major sourceswhere EPA estimates that CEMS are feasible, and EPA and states continue to rely primarily on inspections to detect violations. Recentlegislative proposals to amend the Clean Air Act recognizethe value of emission monitors and-in an effort to help addressthe acid rain problem-call for the use of CEMS at about 1,100 of the nation’s largest sourcesof so,. However, EPA compliance officials state that they have insufficient resourcesto develop regulations requiring CEMS at all sourceswhere feasible, and consequently have no plans to require monitors beyond those called for in the proposed legislation. are electro-mechanicalinstruments, usually installed in the Detection Capability CEMS facility’s exhaust or smoke stacks, which sample, analyze, measure, and of Emission Monitors record the amount of pollutants passing through the stack. CEMS have been developedto measure various types of pollutants emitted by stationary sources.For example, one type of cE:w--opacity monitors- generally operates by measuring the amount of light displaced by visible emissions.This monitor, located in the stack, automatically determines when the amount of displaced light (or opacity) exceedsits permit con- ditions. Figure 2.1 shows the location and operation of a typical opacity monitor. Page 17 GAO/RCED-90-156 Air Pollution Violations Chapter 2 , R.egulatio~ Needed to Require Greater EmWon Monitor Use at Major Stationary Sources Figure 2.1: Location and Operation of a Typical Opacity Monitor Similarly, other CEMS measure visible and nonvisible pollutants such as so,, NO,, and vocs by analyzing emission gasesseveral times each hour. Both opacity and gaseousmonitors automatically record emission levels. Detection Potential of Senior EPA compliance officials told us that CEMS are 10 times more likely Monitors Greater Than to detect violations than on-site inspections, unless the violations are blatant and ongoing at the time of inspection. Somestate officials Inspections believe that CEMS are even more effective, with regulators in Penn- sylvania- a state which has made substantial use of monitors since 1984-estimating that CEMS may be up to 60 times more effective than on-site inspections. While precise comparisonsof the effectiveness of these different detection methods are difficult to make, both groups said that the detection potential of CEMS is much greater than that of inspec- tions becausemonitors . measure and record emissionsdirectly and accurately, Page 18 GAO/RCED!WlBS Air Pollution Violations Chapter 2 Regulations Needed to Require Greater Emission Monitor Use at Major stationmy &urMe . provide near-continuous coverageof facility emissions,and . detect violations that periodic inspections cannot. A 1988 EPA study of CEMS at large facilities in Pennsylvania demon- strated that monitors directly measured and recorded accurate opacity readings 98 percent of the time and accurate so,readingsover 92 percent of the time.’ Conversely, in many situations the inspector, using his training and experience,judges a facility’s compliance status without empirical evidenceof the quantity, rate, or concentration of pollutants emitted. Inspectors often assessfacility compliance by visually observing emissionsand by reviewing operating equipment, records, and pollution control equipment. Becauseof the uncertainties involved in making thesejudgments, inspectors often are unable to determine whether sourcesare in compliance. For example, in one inspection report we examined, the inspector reported that although visible emis- sions from a major stationary source appeared to be about 60 to 80 per- cent opacity-well in excessof the 20-percent permit limit-he “could not do a visible emission evaluation becauseof the angle of the sun” and therefore no violation was officially recorded. Unlike inspections, CEMS can operate 24 hours a day and have been shown to measure more than 90 percent of a facility’s annual emissions. Inspections often cover less than 1 percent of a facility’s annual emis- sions, usually take 2 to 4 hours to perform and, becauseinspection resourcesare limited, are normally conducted at major stationary sourcesonce each year, according to EPA compliance officials. Some facilities may be inspected more frequently if resourcesallow for it and indications of problems exist. Consequently, facilities are subject to far less coveragewhen inspections are the only method used to assess compliance. CEMS can also detect violations that inspectors cannot. According to EPA compliance officials, inspectors have difficulty judging visible emissions at night and in adverse weather, whereas CEMS are not affected by these conditions. More importantly, gaseousemissions,such as SO, and NO,, are generally not visible, whereas CEMS consistently measurethese gases directly and reliably. For somegaseouspollutants, inspectors often can only infer compliance by comparing existing processand control system operating conditions with those recorded during stack testing. However, according to EPA, stack test data are collected under finely tuned process ‘According to EPAcrikria, readingsare consideredacceptableif obtainedfrom a quality-assured CEMthat has beencalibratedand verified to provide data within 16percentof the true value. Page 19 GAO/RCED-90-156 Air Pollution Violations Chapter 2 Regulation Needed to Require Greater Emission Monitor Uee at MaJor Stationary Sources and control system operating conditions, and thus may be atypical of tests conducted under normal operations, further adding to the diffi- culty of detecting violations of permit conditions for gaseouspollutants. We noted an example of these difficulties during our review. Inspectors performing routine inspections at a large fertilizer-manufacturing plant in North Carolina did not detect emissionsviolations from 1983 until time for permit renewal in November 1986. The State’s Technical Ser- vices Branch Chief said most inspections during this period had focused on whether the required control equipment was in place, similar to their inspection practices at other facilities. At permit renewal, inspectors dis- covered that the facility had altered the pollution control equipment years earlier by removing the materials inside the facility’s scrubbers. According to state officials, these materials were integral to the proper functioning of the control devices,and their removal allowed untold amounts of so, and fluoride to be releasedinto the atmosphere. EPA'S Region IV field inspector responsible for covering North Carolina told us that, had CEMS been installed in this case,many of the emission viola- tions would have been detected earlier and, with prompt enforcement action, the environmental harm lessened. Senior EPA compliance officials told us in October 1989 that only about Not Enough Emission 1,066 major stationary sourceshad CEMS installed, even though they Monitors in Use estimated that CEMS were feasible at about 10,000 sources.These offi- cials’ estimate of 10,000 facilities is basedon their knowledge of the cat- egories,types, and sizesof facilities that constitute the universe of 33,600 of the nation’s largest stationary sources.Ten thousand is the number of sourceswhere, in their opinion, one or more CEMS would be technologically feasible and economically viable for use in detecting vio- lations. Although no specific criteria have been developed translating these conceptsfor monitor use into further action, according to the National CEM Coordinator, this means that a facility is capable of installing and maintaining a CEM that provides reliable data at a cost that would not be an economicburden to the firm. EPA had not devel- oped more specific criteria delineating the potential for major stationary sourcesto use monitors at the time of our audit. EPA also lacks precise data on the number of major stationary sources with CEMS, but based its 1989 estimate of 1,065 facilities with one or more monitors installed on its 1989 workload model, contacts with EPA regional and state CEM coordinators, and EPA-sponsored studies of CEM use. According to the National CEM Coordinator, little has changedsince Page 20 GAO/RCED-90-156 Air Pollution Violations chaptw 2 Reguhtions Needed to Require Greater Emission Monitor Use at Mdjor stationary sources the findings of a 1988 EPA study that reported that few monitors were in use relative to the potential for their widespread use. The report stated that only 6 states had more than 100 CEMS installed within their jurisdic- tions, and 26 states had 25 or less CEMSinstalled. According to EPA offi- cials, facilities often have more than one device at the samefacility. Additionally, EPA reported in August 1988 that many installed CEMS were not used to detect violations, and that only 16 states had used CEMS to detect violations. Instead, data from monitors were often used only to target sites for inspections. Extent of Noncompliance BecauseEPA and state regulators continue to depend largely on on-site Understated inspections to detect violations at nearly 95 percent of the major stationary sources,senior EPA compliance program managersbelieve the extent of noncompliance among major stationary sourcesmay be signifi- cantly understated. In this regard, EPA’SStationary SourceCompliance Strategy states that, becauseof the limitations inherent in on-site inspections, it is fair to assumethat compliance data being reported by States do not indicate what is happening at a facility on a day-to-day basis, but rather whether the source has been determined to be in compliance at an announced inspection after it has had the opportunity to optimize the performance of its control equipment. Thus, it indi- cates whether the source is capable of being in compliance rather than whether it is in compliance in its day-to-day operations. Similarly, a September 1987 EPARegion IV memorandum points out that EPA’Spractice of relying on opacity observations and on-site inspection of the source does not ensure that the source is in compliance at all times. The memorandum further states that EPA’Spractice is “hit or miss.” Even with the limited detection potential of on-site inspections, about 14 percent of the major stationary sourceswere detected violating permit or other requirements in fiscal year 1988, with 1,404, or about 30 per- cent, of these noncomplying sourcesin violation one or more times each month for every month of the fiscal year, according to EPAdata. Page 21 GAO/RCJ3D-90-166 Air Pollution Violations chapter 2 l@gulations Needed to Require Greater Emission Monitor Use at Major stationary sources Industry and regulatory officials have expressedconcernsabout the Concerns Regarding reliability and cost of CEMS. According to EPA'S Stationary SourceCom- Monitor Reliability pliance Division Deputy Director, monitor reliability has improved sub- and Cost stantially and the costs are more reasonablethan they were in 1975, when EPA issued its first SIP emission-monitoring regulations. While these concernsmay have been legitimate in the past, the issuesof reliability and cost, according to EPA'S 1984 National Compliance and Enforcement Strategies, have been replaced to someextent with misinformation and inertia. For example, in Pennsylvania-a state which has made increasing use of CEMs since 1984-emission monitors were required at nearly 200 sourcesas of May 1988. The Chief of Pennsylvania’s Division of Abatement and Compliance, Bureau of Air Quality Control, said that using CEMS is largely a matter of mind-set, and that Pennsylvania had decided several years ago that a properly implemented CEM program could produce accurate, reliable data at a cost that was not burdensome to industry. More importantly, he said that, where they can be used, emission monitors are the most effective method of detecting violations and reducing emissions.The Director of Pennsylvania’s Bureau of Air Quality Control said that, while CEMS are not appropriate for every major stationary source,the state’s experience had been so positive that Pennsylvania will require sevenCEMS on each municipal waste inciner- ator it permits in the future. On average,Pennsylvania CEMS have col- lected reliable data over 90 percent of the time. With respect to cost, EPA compliance officials recognizethat cost may continue to be a concern at somesourceswhere EPA envisions using CEMS to detect violations, but these officials said that in relation to facility size and gross annual revenues,the cost of monitors is justified, in their opinion, at about 10,000of the 33,500 major stationary sources.In their opinion, installing and operating CEMS would not constitute an economic burden for these major stationary sources.Further, EPA points out that using CEMS to detect violations is already an important part of a few state programs. For example, California and Pennsylvania officials have used CEM readings to not only detect violations but also to take enforce- ment actions at a number of sourceswithin their jurisdictions. For example, of the 4,076 total violations cited in the San Francisco Bay Area in 1986 and 1987,216 were basedon CEM readings, while Penn- sylvania issued 260 violations on the basis of CEM readings for a similar 2-year period, eventually settling the casesfor $336,600 in penalties. Becausemost state and local agencieshave been slow to implement a CEM regulatory program, EPA sponsoredseveral studies in an attempt to foster their wider acceptanceand use in detecting violations. According Page 22 GAO/RCED-90-166 Air Pollution Violations chapter2 Re@latio~ Needed to Require Greater Emhion Monitor Use at Major statlonluy sourcee to a June 1986 report,2 state agencies’reluctance to implement an emis- sion-monitoring program centers around (1) perceptions of CEMS as unre- liable and (2) concernsthat their use would place additional burdens on the limited resourcesof regulators. The report stated that the perception of state managerswas that EPA had required the states to proceed with a questionable monitoring program, without adequate state agency con- sultation, and then had not provided sufficient technical and policy sup- port to make the program work. After pilot testing CEMS in two states for more than a year, the study concluded that CEMS’performance showed “very high levels of relia- bility,” and the burden of reviewing excessemission reports was not a burden, averaging less than 1 hour per report. Most importantly, the study demonstrated that CEMS were a highly effective method for assessingthe compliance of selectedfacilities, and that the issue of unre- liability can be readily resolved by a managementcommitment to quality assurance. Cost Implications of While an effective quality assuranceprogram can overcomeconcerns Emission Monitors about the reliability of CEMS, resolving the issue of cost has not been as easy for EPA. First, becauseinstalling, operating, and quality assuring data from CEMS’ is an expensethat facility owners and operators must absorb, EPA doesnot envision their use at all sources;instead, EPA has targeted only the newest and largest of the major stationary sources. According to EPA compliance officials, the costsof installation vary greatly, from about $10,000 to $150,000, depending on the type, size, location, and facility design, as well as whether any retrofitting is involved. Annual operating and maintenance costs,including quality assuranceand periodic reporting, range from about $6,000 to $60,000. However, EPA’s National CEM Coordinator points out that relative to the cost of other pollution control equipment that major stationary sources must install, these costs are small, reasonable,and justified, in his opinion, at many major stationary sources. To help alleviate industry concern over the issue of cost, EPA conducted reviews of selectedfacilities. As a result, EPA reported in a December 1988 Federal Register that CEMS constituted lessthan 4 percent of total air pollution costs.Further, as reported by the 1988 Joint Power Gener- ation Conference,in somecasesCEMS can reduce operating expenses. ?+unmary Report:A Pilot Projectto Demonstratethe Feasibility of a StateContinuousEmissionMon- itoring System(CEMS)RegulatoryProgram(EPA,June 1986). Page 23 GAO/RCELMO-155 Air Pollution Violations Chapter 2 Regulations Needed to Require Greater Emission Monitor Use at Major Stationary Sources According to the Deputy Director, Stationary SourceCompliance Divi- sion, the relative cost of installing CEMS at major sourcesis less than the cost of installing a speedometeron an automobile. He pointed out that while using CEMS to detect violations may not result in significant sav- ings in inspection resources,it is the most effective meansof ensuring facility compliance with permit conditions. According to the National CEM Coordinator, in addition to unreliability, industry has used cost as a reason for resisting CEMS when cost is not the reason. The real reason, in his opinion, is that facilities simply do not want their emissionsmoni- tored on a full-time basis. CEMS can tell exactly when a sourceexceedsits permit limits, by how long and how much, he said. He baseshis opinion on his dealings with industry, EPA regional and state CEM coordinators, and EPA'S own studies. For example, in responseto one facility’s high cost estimates for CEMS, an in-depth investigation by EPA found that the facility’s annual costs would be increased less than two-tenths of 1 percent. Our work similarly indicates that cost may not be a principal reason that more sourceshave not installed CEMS. For example, we visited a major coal-fired utility that had installed two opacity and two so, monitors at a total cost of about $250,000, according to the Assistant Plant Manager. To place this amount in perspective, he said gross revenuesof this facility average over $100 million per year through the sale of electric power to sectionsof the northeastern United States. Company repre- sentatives told us that CEMS allow them to more closely monitor fuel combustion efficiency and to better ensure the quality of low-sulfur coal from suppliers. Although they had not evaluated whether CEMS had actually reduced operating expenses,the representatives said that the costs were not unreasonable. Other firms we contacted said that they had reduced their operating expensesby using CEMS. For example, a major manufacturer of wall cov- erings in Ohio reportedly savesabout $50,000 annually in fuel costs becauseits so, monitors more closely gaugefuel combustion efficiency, and a large U.S. automobile manufacturer, which also operates its own power generating facilities, similarly savesat least $12,000 annually in fuel costs at each of its industrial boilers from GEMS it uses. Operational cost savings have also resulted from using CEMS at sources other than large power-generating facilities. Although EPA has promul- gated no regulations on the use of voc monitors, savings have been esti- mated for several Pennsylvania facilities using voc monitors. For example, one inspector reported that one manufacturer reduced its Page 24 GAO/RCED-SO-155 Air Pollution Violations Chapter 2 Regulatlone Needed to Require Greater Jhlssion Monitor Use at Major Stationary Sources annual costs for solvent by $50,000 after equipping its carbon absorp- tion unit with WC monitors and changing its operating practices accord- ingly. Solvent recovery for this firm increased from 68 percent in June 1988 to 87 percent in September 1988 after installing CEMS to detect excessemissions.Another Pennsylvania facility’s solvent usagewas nearly cut in half after installing voc monitors and correcting inefficient operating techniques. According to a 1988 inspection report, this firm decreasedits solvent usageby 46 percent, representing an annual reduc- tion in operating expensesof about $20,000. Monitors Believed Feasible EPA, state, and industry officials told us that CEMS are technologically at More Sources feasible and economically viable, or capable of providing reliable emis- sions data at a cost that would not be an economicburden to the source, at many more sourcesthan those that currently use them. For example, the Coordinator of Ohio’s CEM program said steam generators as small as 10 million Btus in size have been required to install and operate CEMS to monitor so, emissionsin Ohio, whereas EPA'S regulations require CEMS only on steam generators that are larger than 250 million Btus. A Ford Motor Company Regional Environmental Engineer agreed,noting that Ford has used CEMS extensively since 1981 becausethe company found that CEMS help Ford reduce operating costs.He said they are even used on power boilers not covered by Ohio’s CEM requirements. Furthermore, California and Pennsylvania local air compliance program officials told us that in selectedcases,voc monitors are feasible and already used in their states to detect violations, even though EPA has no regulations requiring WC monitors at any major sources.Pennsylvania officials told us they also plan to use CEMS on hazardous, municipal, and infectious waste incinerators. Other indicators also support the position that CEMS are potentially fea- sible for many more sourcesthan currently use them. For example, according to the National CEM Coordinator, EPA has an effort underway to improve its emission-monitoring data base for so, monitors at major power-generating facilities as part of EPA'S increasedemphasis on acid rain. Airborne so, is a major component of acid rain. Data from this effort indicate that, as of January 1988, only 366 of 2,414 major SO, sourcesin the United States used so, monitors. According to EPA'S National CEM Coordinator, CEMS are economically feasible at nearly all of these facilities. Page 25 GAO/RCED-90-156 Air Pollution Violations , Chapter 2 Regulations Needed to ReqnIre Greater Fhnission Monitor Use at M&w stationary som.ees EPA Regulations Focus on Sections 110, 111, and 114 of the Clean Air Act collectively provide EPA Newest and Biggest with the authority to require any stationary source to install, use, and maintain emission-monitoring equipment. Under this authority, EPA has Sources issued regulations requiring CEMS to be installed at 23 categoriesof new sourcesand 4 categoriesof existing major stationary sources.However, becauseof the reliability and cost concernsdiscussedabove, EPA'S regu- lations governing the use of CEMS have largely targeted only the newest and biggest sources,and have allowed even these sourcesalternatives and exceptions to the use of CEMS, according to EPA compliance officials. For example, in August 1988 EPA reported that only 36 sourcescould be categorized as electric utility steam generators of greater than 250 mil- lion Btu-per-hour capacity that had begun construction after September 18,1978-one of the New SourcePerformance Standard (NSPS)" catego- ries required to install CEMS under section 111 of the act. Similarly, under another subpart of the new source regulations, only those indus- tries that began construction of large steam generators (greater than 100 million Btus per hour capacity) after June 19, 1984, are to install CEMS. The number of CEMS EPA requires under other sectionsof the act is sim- ilar to its new source requirements, targeting only the biggest of sources. Under section 110 of the act, EPA is to ensure that state and local pro- grams’ SIPSrequire certain existing facilities to monitor emissions. Although states are free to target additional sources,EPA has designated only four narrowly defined categoriesof existing sourcessubject to these requirements. These categoriesinclude (1) fossil fuel-fired steam generators with a rated capacity of greater than 260 million Btus per hour, (2) nitric acid plants with production capacity of more than 300 tons of nitric acid per day and which are also located in nonattainment areas for NO,,, (3) sulfuric acid plants with production capacity of more than 300 tons of sulfuric acid per day, and (4) fluid bed catalytic cracking units at petroleum refineries with production capacity of more than 20,000 barrels per day. “Msjor stationary sourcesconstructedafter specificdatesidentified in EPA’sregulationsare subject to NSPSstandards,which generally require installation and useof morestringent controls to attain lower emissionlevels and, in the caseof selectedNSPSsources,continuousmonitoring of emissions. Page 26 GAO/RCEIMO-156 Air Pollution Violations Chnpter 2 Regulatlo~ Needed to Require Greater Emhadon Monitor Ulle at Major stationary sourcee Efforts to ensure that major stationary sourcesmaintain compliance EPA’s Monitor Use have becomeincreasingly important, according to EPA, now that most Policy Not facilities have installed the basic pollution control equipment. In recog- Implemented nition of this growing need and becausemonitor reliability has improved and, according to EPA, the costs are more reasonable,in March 1988 EPA'S Office of Air Quality Planning and Standards issued an emission-moni- toring policy calling for CEM installation and use where feasible. According to the policy, if it is technically feasible, CEMS should be required in all new facility permits, operating permits, and resolutions of enforcement actions, including consent decreesand administrative orders. The policy states that CEMS should be used to ensure continuous compliance of sourcesin both attainment and nonattainment areas. The policy also points out that at least five states-Indiana, Ohio, Penn- sylvania, Tennessee,and Washington-already have well-developed emission-monitoring programs, and that EPA'S review of these programs has shown CEMS to be valuable tools for ensuring compliance. EPA states in its policy that it is committed to using CEM data in assessingfacility compliance. However, EPA has not fully followed through in implementing its policy, citing resource limitations and higher priority activities. For example, EPA'S Stationary SourceCompliance Division Deputy Director pointed out that funding for compliance activities has declined 35 percent (in noninflationary adjusted dollars) since 1979, while the major stationary source population has grown from about 23,000 to over 30,000 sources in the sametime period. Although EPA'S air program work force has remained about the samesize nationally, compliance program staffing has been reduced by 14 percent, from 339 to 293 staff, since 1980. He also pointed out that, while these reductions were taking place, even minimally acceptableinspections have becomemore complex and time- consuming, often taking 60 percent more time than previously required. In addition, efforts expended on a host of new problems such as asbestos,air toxics, radon, hazardous and municipal waste incineration, and smaller sourcesof vocs have taken ever increasing amounts of inspectors’ time. Thus, while EPA has the authority to require CEMS at a broad range of sources,according to the Deputy Director, the Stationary SourceCompli- ance Division’s limited available resourceshave forced headquarters managersto make difficult choices,effectively making EPA efforts to require more extensive use of CEMS a lower priority. The National CEM Coordinator explained that, in order to issue regulations requiring CEMS at a broad range of sources,the Stationary SourceCompliance Division Page 27 GAO/RCJXHO-165 Air Pollution Violations Chapter 2 Regulations Needed to Require Greater Emission Monitor Use at Major stationary sources would need to validate the various measurementmethods involved, develop quality control and quality assurancemethods and manuals, develop and propose regulations for each industry affected, answer and incorporate commentson the proposed regulations, and then promulgate final regulations. The Deputy Director explained that undertaking these activities is not possible with their current staffing if they are to also continue compliance and enforcement actions at sourcesand address new and emerging environmental issues.Consistent with the Deputy Director’s statements, EPA'S 1984 ComGmce Monitoring Strategy for Major Stationary Sourcespoints out that without additi&.l res&rces, EPA’S air compliance program cannot addressboth the need for CEMS and the need to capture previously unregulated smaller sourcesof harmful pollutants such as vocs, or to effectively control asbestosdemolition and renovation projects. EPA decided the latter was a higher priority need. While resourcesand work load are legitimate factors, EPA continues to approve and fund state and local inspection programs that make little use of CEMS. In 1988, EPA, and state and local agenciesconducted over 37,000 inspections. Furthermore, EPA has been slow to ensure that state and local programs comply with the Agency’s 1976 regulation requiring them to incorporate CEMprovisions in any SIP revisions after that date. For example, two states in EPA Region III-Maryland and West Vir- ginia-have had SIPS approved without CEMprovisions becauseCEMS have not been a high priority consistently over the years, according to EPA Region III officials. EPA Region IV officials explained that inconsis- tent guidance and direction from EPA headquarters over the years regarding the need for adequate CEM provisions in state SIPS had also causedthem to question the Agency’s commitment to CEMS as a detection method. Consequently, they told us in June 1988 that their regional approach to CEMS had largely been one of assistancerather than leader- ship. They explained that if a state or local agency solicited their help in establishing an active CEM program, they would work with them in doing so. Otherwise, it was not a priority. In addition, headquarters officials told us that EPA has not developed a strategy for promoting CEMS at more facilities, nor do they have a plan for validating existing CEM technology and measurementmethods-two critical first steps to regulatory use of CEM data in enforcing emission limits. Furthermore, in both 1988 and 1989, EPA reduced funding for its CEM program, according to the National CEM Coordinator. Even an EPA- Page 28 GAO/RCED-!bO-165Air Pollution Violations Rq~I~~tions Needed k, Require Oreater Emimion Monitor Use at Major Stationary Sources contracted study found that lack of managementsupport and commit- ment to the CEMprogram has seriously hampered its implementation. Specific problems identified in this 1988 report4 included the lack of or inconsistent responsesto Regional initiatives and inquiries, failure to pro- vide training and in-house technical support, and failure to resolve important data quality problems for the CDS/CEMSsubset. Recent EPApolicies and program priori- ties have in somecasesbeen inconsistent with continued support of CEMSprogram implementation. In a least one Region, a decision to discontinue GEMS-related efforts was reportedly made in response to commentsfrom a visiting OAQPS[Office of Air Quality Planning and Standards] Headquarters representative that the Agency was moving away from the NSPSprogram. Clean Air Act In an effort to control acid rain, the administration’s proposed amend- Amendments Recognize ments to the Clean Air Act recognizethe value of CEMSand would require their use by the most significant contributors to the nation’s acid Value of CEMs rain problems. More specifically, under SenateBill S.1630,opacity, so2, NO,, and volumetric flow monitors would be required at 111 of the nation’s largest coal-fired electric utilities within 36 months of enact- ment, and at another 800 to 1,000 of the nation’s largest major indus- trial sourceswithin 6 years of enactment, according to EPA.While these proposals are a step in the right direction, they include somefacilities that already have monitors installed and stop short of legislating the extensive use of CEMSalready called for in existing EPApolicy. According to senior EPAcompliance officials, CEMSare technologically feasible and economically viable at an estimated 10,000 major stationary sourcesand could enhancethe likelihood of detecting violations at these sourcessub- stantially over current practices. Despite the significant detection potential of CEMS,becauseof cited resource limitations, EPAofficials said that they will continue to rely pri- marily on their encouragementof state and local agenciesto develop CEM programs, but will not be able to require their broader use under federal regulations. According to senior compliance program managers,while EPAhas the authority to require broader use of CEMS,their installation and use beyond the sourcesalready identified in the proposed legislation will continue to be slow. Major stationary sourcesare one of the biggest contributors to our Conclusions nation’s air pollution problems. These facilities are responsible for 4Status of EPARegional CEMSProgramImplementation (EPA,Aug.1988). Page 29 GAO/RCED-90-166 Air Pollution Violations chapter 2 Regulatlona Needed to Require Greater Rmlmlon Monitor Urn at Major Stationary Sources nearly 40 percent of the nation’s total emissionsof criteria pollutants. EPAhas sought, through regulations, permit requirements, and its review and approval of SIPS,to control major stationary sourceemissions.How- ever, ensuring compliance at major sourceshas been, and continues to be, a significant problem. GEMSrepresent a significant improvement in ensuring that major sources comply with emission requirements. The concernsabout CEMSthat existed in the past-reliability and cost-have been eliminated or reduced and, in our opinion, do not remain an impediment to their wide- spread installation and use. Further, in contrast to traditional inspec- tions, the use of monitors provides solid evidenceof a facility’s emission levels and whether it complies with permit conditions. CEMSprovide far more coverageof facility operations than do inspections and can detect violations that inspectors cannot. EPA'sefforts to expand the use of CEMSat existing stationary sources, however, have been inadequate. Although EPA'Spolicy for encouraging the use of CEMSis a strong acknowledgementof the benefits of this tech- nology, installation of this equipment will not occur on a widespread basis until it is required by EPA.This is becauseonce CEMSare installed, major sourcesthat are not operating within their permit limits face a much greater likelihood of detection. However, the regulations neededto transfer this policy into requirements have not been forthcoming. In our view, regulations that (1) establish criteria for determining where CEMS are technologically and economically feasible and (2) require CEMuse in all situations where these criteria are met would significantly increase the use of this technology. The proposed Clean Air Act Amendments would increasethe number of facilities required to install and use CEMS.The amendments,if enacted, would mandate the use of monitors at approximately 1,100 facilities. However, on the basis of EPA'Sestimate that CEMSare feasible at about 10,000 facilities, considerably greater use can and should be made of CEMSin light of the continued air pollution problems. In order to more effectively implement EPA'Semission-monitoring policy Recommendation and achieve more widespread monitor use, we recommendthat the Administrator, EPA,promulgate regulations that (1) establish clear cri- * teria for determining where CEMSare feasible for major stationary sourcesand (2) require CEMinstallation and use at all major sources meeting these criteria. Page 20 GAO/RCED-90-156 Air Pollution Violationa Chapter 3 Penalty Amounts Not Sufficient to Deter Violations Failure to comply with emission requirements can result not only in additional air pollution, but can also be of economicbenefit to major sta- tionary sources.To deter such noncompliance,the Clean Air Act con- tains penalty authority that allows enforcement officials to administratively assessnoncompliancepenalties sufficient to remove any economicbenefit a violator can gain from failing to comply with the act’s requirements. EPAhas also established a civil penalty settlement policy used in its civil litigations that requires agency attorneys to accept only those settlement offers that are calculated to offset the eco- nomic benefit violators obtain through noncompliance. State and local programs, however, which handled over 80 percent of the enforcement actions in fiscal year 1989, have not been required to seek economicbenefit penalties. These state enforcement programs operate under state law as embodied in the various SIPS,not under the administrative penalty provisions embodied in the Clean Air Act or the EPAcivil penalty settlement policy, and EPAhas not insisted through the SIP approval processthat states commit themselvesto imposing eco- nomic benefit penalties. As a result, penalties imposed on significant vio- lators--EPA’s highest enforcement priority-were often not sufficient to remove the economicbenefit of noncompliance.Specifically, l fewer than half of the significant violators in fiscal years 1988 and 1989 paid cash penalties and l when cash penalties were assessed,the penalties often were not based on the violator’s economicbenefit. Further, although EPAcan seek to impose its own penalties in those instances where state and local penalties are inadequate, it has seldom acted to collect additional moneys becauseit views its own enforcement processas cumbersomeand too resource-intensive. To help EPAtake enforcement actions more efficiently, proposed amend- ments to the Clean Air Act contain provisions that grant EPAadditional administrative penalty authority for enforcing air pollution violations. However, EPAstill needsto clarify its authority over state and local agencies’enforcement actions to ensure that fair and equitable penalties are consistently imposed, including taking action in state and local cases where penalties are not sufficient. Otherwise, sourcesthat comply will continue to be at an economicdisadvantage to those that do not. Page 31 GAO/RCEB90-166 Air Pollution Violations Chapter 3 Penalty Amounts Not Sufficient to Deter Violations Failure to comply with emission requirements can result in significant Economic Benefit economicbenefit to major stationary sources.Among other things, these Penalties Established facilities can reap dollar savings by (1) not purchasing pollution control to Promote equipment or maintaining equipment already in use, (2) using less expensive but higher polluting raw materials, (3) not changing their pro- Compliance duction processes,or (4) not applying for a permit to conduct operations controlled under the Clean Air Act. For example, according to a 1986 industry study of emission control technology, steel drum manufac- turers and reconditioners could save from $240,000 to $530,000 per plant in annualized costs by not purchasing the equipment neededto control voc emissions. To addresssuch financial disincentives to compliance, the 1977 Amend- ments to the Clean Air Act gave EPA administrative authority to impose substantial cash penalties for pollution control violations. Section 120 of the act provided EPA with administrative authority to make a finding of noncompliance and assesscash penalties against noncomplying firms sufficient to remove any economicbenefit derived from the violation. The intent of this provision was to forestall noncompliancewith the act’s requirements by removing the economicadvantage to be gained by postponing investment in pollution control. At the sametime, the Con- gressincreasedthe civil penalties that can be imposed under section 113 of the act and expressedan intent that court-imposed civil penalties “be assessedin amounts which are adequate to assurecompliance will result, rather than permitting continued noncomplianceto be economi- cally profitable.” EPA established regulations for imposing economicbenefit penalties that are consistent with the intent of section 120. In 1984, EPA also developed the Clean Air Act Stationary SourceCivil Penalty Policy which requires, among other things, that when EPA settles an enforcement lawsuit out of court, all such EPA settlements seekto remove the economicbenefit derived by not complying with the act’s requirements.’ This civil penalty settlement policy provides that, except in extraordinary circumstances, “the lowest possible settlement penalty will be the calculated economic benefit of noncompliance.” The policy directs that delayed and/or avoided expenses,such as the costs of control and monitoring equip- ment, modifications to production processes,operation and mainte- nance, and the employment and training of pollution control personnel, ‘The EPApolicy doesnot restrict the decisiona judge might render if a particular caseshould go to trial, and the economicbenefit that may be calculatedis subjectto the statutory maximumof $26,000 per day of violation. Page 32 GAO/RCED-90-166 Air Pollution Violations Chapter 3 Penalty Amounta Not Sufficient to Deter Violation9 be included in the calculation of economicbenefit. Furthermore, the civil penalty settlement policy provides that additional amounts be obtained, if warranted, to reflect the gravity or seriousnessof the violation. Becausethe act is structured as a federal/state partnership for the con- Economic Benefit trol of air pollution, state and local agenciesare authorized to carry out Penalties Not Imposed enforcement activities against violators, including major stationary by State and Local source violators. In fiscal year 1989, state and local agencieshandled over 80 percent of the enforcement actions against stationary sources. Programs These activities are required to be performed in accordancewith EPA- approved SIPS, which establish the requirements that authorized state and local programs must adhere to in order to avoid the act’s sanctions and obtain federal funding. However, EPA has not required authorized state and local programs to calculate or impose economicbenefit penalties to correct violations. The last major revisions to SIPS were in 1982-about 2 years before EPA final- ized its Stationary SourceCivil Penalty Policy. Although EPA has the authority to call for unscheduledSIP revisions whenever it is apparent that attainment is in jeopardy, and although individual SIPS have been acted upon during the intervening years, EPA has not required state and local officials to revise their SIPS to upgrade administrative penalty pro- visions to recover economicbenefit. As a result, state and local pro- grams often do not impose cash penalties and at other times impose penalties that do not remove the economicbenefit of noncompliance.EPA data for fiscal years 1988 and 1989 show that over half of the signifi- cant violators-54 percent-identified by state and local programs paid no cash penalties. Table 3.1 summarizesthe penalties assessedin state and local enforcement actions in these years. Table 3.1: Resolution of State and Local Enforcement Actions for Significant 1998 cases 1989 cases Total cases Violator Cases, Fiscal Years 1988 and Type of resolution Number Percent Number Percent Number Percent 1989 Cases resolved with cash penalty assessed 183 37 336 53 519 46 Cases resolved with no cash penalty assessed 315 63 300 47 615 54 Total cases resolved 498 100 636 100 1,134 100 Page 33 GAO/RCED-90-155 Air Pollution Violations Chapter 3 Penalty Amounts Not Sufficient to Deter Violations Our review of eight state and local programs found that none regularly sought to recover economicbenefit penalties. Three of the eight pro- grams used EPA'S civil penalty settlement policy occasionally as a refer- encein assessingbut not collecting fines, four programs did not consider economicbenefits in imposing cash penalties, and one program-Ari- zona’s-did not have the authority to enforce civil penalties against vio- lators, according to Arizona Office of Air Quality officials. In order to assessmonetary penalties for violations in Arizona, criminal charges must be brought by the state attorney general’s office, and the case must be tried in criminal court. Otherwise, the state program can only issue violation notices, which carry no monetary penalties. Our review of enforcement actions in the eight programs included cases that were resolved without penalties to remove the economicbenefit of noncompliance.As shown in the following three examples, although the severity of the violations varied, violations that appear to have econom- ically benefitted the violator were settled with either no cash penalty assessedor, when cash penalties were imposed, no economicbenefit calculated: l A North Carolina facility emitting particulate@ and vocs from the annual production of about 30,000 tons of pipe (1) violated its permit limits for particulate emissions5 times between 1984 and 1988, with 1 inspection observing opacity emissionsat nearly 100 percent-almost 5 times above its limit of 20 percent, (2) conducted pipe grinding and fit- ting operations without a permit, and (3) used noncompliant paint coat- ings that exceededvoc emission levels by 13 percent. Local program officials acknowledged that the firm may have benefitted financially from the avoided and delayed cost of timely compliance, but the eco- nomic benefit was not calculated by the local agency and no penalty was assessed.According to the manager of the local agency’sAir Quality Branch, the firm agreed in July 1988-nearly 5 years after problems were first discovered-to do a better job of operating and maintaining its control equipment, not operate any new processeswithout a permit, and have its paint reformulated to meet the standard. . A major refinery located in an ozonenonattainment area in California was cited for 10 violations of voc emissionsstandards in 1986 and 1987, resulting from defective tank seals,failure to use proper seals,and leaking valves. At the time of our review, three of the violations were resolved with no cash penalty and four were resolved for a total penalty of $1,300. The remaining three violations were still unresolved at the “Particulatcs include dust, dirt, soot,smoke,and liquid dropletsemittedinto the air. Page 34 GAO/RCED-!bO-166 Air Pollution Violations Chapter 3 Penalty Amounts Not Sufficht to Deter violations time of our review. No calculation of economicbenefit was performed, although enforcement records show that the violations resulted from preventable operational failures and were not the result of equipment breakdown or malfunction. . A major drum-reconditioning facility in North Carolina, producing about 900 reconditioned steel drums daily, emitted excessvocs into a voc nonattainment area from 1980 to 1986 in violation of EPA regulations. This facility failed to install required control equipment by July 1980 and, when discovered in 1986, was using paints that emitted vocs 57 percent above the standard set for coating steel drums. Negotiations with the facility to resolve the violations took over 2 years, resulting in a penalty of $15,000. No calculation of economicbenefit was performed, but the local air program director told us that this penalty amount was consistent with historical amounts for other violations in the local area. These penalties, however, may not approach the benefit the violator obtained nor serve as a deterrent to future violations. At our request, the EPA headquarters official responsible for the economicbenefit pro- gram calculated the estimated economicbenefit for the violation at the drum-reconditioning facility discussedabove. According to his estimate, the economicbenefit the firm derived was over $231,000-about 15 times larger than the $15,000 fine imposed. Furthermore, in this situa- tion, the size of the penalty may not have deterred future violations. During a June 1988 inspection- 2 months after paying the $15,000 pen- alty-this facility was found conducting unpermitted operations and had noncompliant materials on-site that would emit excessvoc emissions. EPA’S Office of the Inspector General has also found that penalties are insufficient. In a March 1988 report on EPA RegionV and the states of Michigan and Wisconsin,the Inspector General reported that penalties are largely basedon what EPA and state regulators believe they can obtain from the company rather than the economicbenefit. The report further stated that most collected penalties did not recover the savings the violator received from delayed and/or avoided costs and that cases were found with no support as to how the penalty was calculated. State and Local State and local agenciesgenerally seekto use their enforcement actions Enforcement Philosophies to securea source’sfuture cooperation rather than to penalize past vio- lations. Someofficials from state and local programs we contacted told Differ From EPA’s us that their penalties are often directed toward getting the attention of Page 36 GAO/RCED-W-166 Air Pollution Violations Chapter 9 Penalty Amounts Not Sufficient to Deter Violationa the violating facility’s management,including threats of adversepub- licity, rather than removing the economicbenefit of noncompliance, They said their programs attempt to achieve compliance by working cooperatively with facility owners and operators to correct the causeof the violation, in lieu of assessingpenalties. For example, the Chief of North Carolina’s Air Quality Section said that their approach doesnot include calculating, assessing,or collecting penalties using economicben- efit as the meansof deterrence. Similarly, one local program official said that compliance, rather than penalizing violators to remove the economicbenefit of noncompliance,is the goal of his program. His goal is compliance through field presence and technical assistanceto those firms having pollution control problems. He also said that other concernsmust be consideredin imposing penalties for violations, one of which is the financial impact of the penalty on the violating facilities, He expressedconcern that placing economicbenefit penalties on significant violators could put facilities in his area at a competitive disadvantage relative to facilities in other areas that do not impose such penalties. According to somestate and local officials, penalties need only be large enough to get the attention of facility management.As a result, they said their enforcement philosophies will continue to emphasizecompli- ance through field presence,technical assistance,and fines on a case-by- casebasis. could increase cash penalties if, through its review and approval of EPA Has Made Little EPA SIPS, it were to require state and local enforcement agenciesto adopt Effort to Increase penalty policies similar to the section 120 regulations and, in situations Penalties where the state and local agenciesdo not obtain economicbenefit penal- ties, by initiating its own action to impose a penalty. However, EPA has been reluctant to specify that state and local programs seek economic benefit penalties becausethe Agency is uncertain of its authority to do so. Further, EPA’S ability to otherwise influence state and local programs’ penalties is viewed as limited becauseof the time, expense,and resourcesrequired to initiate its own enforcement actions. EPA Reluctant to Require EPA officials said that they have been reluctant to mandate that author- Programs to Adhere to ized programs commit themselvesto imposing economicbenefit penal- ties. According to these officials, requiring programs to follow EPA’S civil Penalty Policy penalty settlement policy on recovering economicbenefit’could be Page 36 GAO/RCED-90-165 Air Pollution Violations Chapter 8 Penaky Amounta Not Suf’flcient to Deter Violation.9 viewed as an infringement on the states’ authority to operate their own programs. This could result in the states giving programs back to the federal government to operate rather than complying with enforcement standards basedon EPA’S economicbenefit regulations and the related civil penalty settlement policy. EPA compliance officials pointed out that in 1984, EPA attempted to get state and local programs to adopt EPA’S policy, but terminated this effort in 1986 after failing to reach con- sensuswith the State and Territorial Air Pollution Program Administra- tors and the Association of Local Air Pollution Control Officials- associationswhich represent state and local air pollution control offi- cials. Since 1986, EPA has not attempted any similar efforts. EPA compliance officials also said they are unsure of the Agency’s authority to require authorized programs to upgrade their penalty poli- cies. They pointed out that state and local programs implement state laws. Consequently, for state and local programs to be required to impose economicbenefit penalties, state laws would have to be amended,and the EPA officials were uncertain of their authority to require this. EPA’S earlier attempt led them to believe that states would not be willing to amend their laws to conform to a more stringent pen- alty policy unless such conformance was specifically required as part of the act. We found that EPA has not requested a legal ruling from EPA’S Office of General Counselon this matter. EPA compliance officials said that such a request could further impinge on their already deteriorating relationship with many state and local programs and that a negative ruling from EPA’S own counsel could place them in a worse position for interceding in problem casesthan current practice allows. In our view, however, EPA has the authority to require that SIPS be revised to incorporate whatever enforcement procedures EPA deemsade- quate, including the nondiscretionary imposition of economicbenefit penalties for violations. Specifically, section 110 of the act sets out the minimum requirements for the content of a SIP and chargesEPA with SIP review and approval. According to this section, a SIP must include a pro- gram to provide for the enforcement of emission limitations and neces- sary assurancesthat the state will have adequate authority to carry out the SIP. Additionally, section 172 of the act, which deals with nonattain- ment areas,mandates that EPA ensure that SIPS include all the necessary requirements to achieve and enforce compliance with the plan. Page 37 GAO/RCED-90-156 Air Pollution Violations Chapter 8 Penalty Amounts Not Sufficient to Deter Violationa Concernssuch as those that EPAofficials expressedhave been raised before in the context of vehicle inspection and maintenance programs. Although these programs were a sourceof much contention, the courts ultimately held that states could be required to enact legislation to carry out the Clean Air Act’s requirement to develop such programs or accept the sanctions provided under the act. Under the sametheory, a state could be required to seek and collect economicbenefit penalties for emis- sions, permit, and other violations by stationary sources, EPA Ability to Impose EPA is authorized to initiate a civil suit under section 113 of the act or an Additional Penalties H18s administrative action under section 120 to increasestate penalties. EPA refers to this procedure as “overfiling.” Although sections 113 and 120 Been Limited provide EPA with the legal authority to issue orders, commencecivil actions, institute criminal proceedings,and assesseconomicbenefit pen- alties, EPA compliance officials said their ability to remedy insufficient state and local penalties is limited becauseeach of these sections cur- rently has drawbacks that hamper EPA from ensuring that state and local penalties are adequate. In addition to other sanctions, section 113 authorizes EPA to seek court- imposed penalties for air pollution violations. However, EPA compliance officials contend that this is an expensive,time-consuming process, sometimescosting more to prosecute than the eventual court-imposed penalty. According to EPA'S Assistant Enforcement Counselfor Air Enforcement and Compliance Monitoring, undertaking court action can cost EPA from $50,000 to about $400,000 on civil actions it takes against polluters, depending upon the complexity of the caseand whether the caseactually goesto trial or is settled before a trial begins. He stressed that EPA has no way of knowing which caseswill actually go to court, so all casesmust be handled as if they will eventually end up before a judge. Although overfiling, if successful,can have significant deterrent effects, the high cost of using section 113 to overfile in large numbers of caseswhere penalties are insufficient is impractical, according to the Assistant Enforcement Counsel.The Chief of EPA'S Region IV Air Com- pliance Branch concurred, noting that in somecases,EPA collects less penalty money from overfiling in state and local casesthan it spendsin taking the enforcement action. EPA cannot recoup its costs of litigation, and all penalty money collected goesto the U.S. Treasury, not to EPA. In contrast to section 113, section 120 authorizes EPA to assesseconomic benefit penalties administratively and to avoid court action unless a source appeals EPA'S administratively determined sanction. However, Page 38 GAO/RCED-90-155 Air Pollution Violations Chapter 3 Penalty Amounts Not Suffkient to Deter Violations EPA views this provision of the act as having other problems which make it undesirable to use in many cases.For example, section 120 doesnot provide EPA with the ability to collect a penalty for any period of viola- tion that occurs before EPA has officially notified the owner or operator of the noncompliance, a limitation not found in section 113. Addition- ally, the penalty calculation formula for section 120 was promulgated in EPA regulations and requires collection of the full economicbenefit, pro- viding little flexibility, according to EPA,for regulators to settle casesfor less. Also, section 120 seeksto gain compliance through economicpenal- ties alone and provides no ability to order compliance or to seek injunc- tive relief. Although the act allows EPA to use section 113 and 120 together to achieve an appropriate remedy for a violation, EPA officials said dual actions such as this are an extremely costly way to achieve compliance, and thus such dual actions are officially discouragedby EPA'S enforcement program guidance. Becauseof the high cost, difficulty, and inflexibility of using their cur- rent authorities to increasestate and local penalties, EPA overfiles rela- tively infrequently. According to the Deputy Director, Stationary Source Compliance Division, the air compliance program’s enforcement resourcesare limited, and any EPA region is able to sustain casework on only 12 to 15 enforcement actions a year, or a total of 120 to 150 casesa year nationwide. Consequently, EPA headquarters and regional officials told us they are able to overfile only in precedent-setting casesand those where the initial penalties are grossly deficient. Region IV staff told us that becauseso many penalties are deficient, a regional rule of thumb- described in a March 1988 EPA Office of the Inspector General report on Region IV as the “laugh test”- is used to determine which caseswill be challenged. According to Region IV staff, a proposed penalty is not chal- lenged if they do not laugh too loudly. A Region III enforcement official told us that, while other judgmental factors may be considered,his region also usesthe laugh test in deciding which penalty amounts to challenge. Senior officials from EPA'S Stationary SourceCompliance Division and the Associate Enforcement Counselfrom EPA'S Office of Enforcement and Compliance Monitoring believe that EPA needsa flexible, easy to use administrative penalty authority to improve the Agency’s ability to overfile in caseswhere assessedpenalties are insufficient to recover the economicbenefit the violator obtained and to deter future violations. According to these officials, such authority could enable EPA to impose economicbenefit penalties against polluters expeditiously, thereby Page 39 GAO/RCED-90-156 Air Pollution Violations , Chapter3 PenaltyAmounteNot Sufficientto DeterviolatioM allowing it to more frequently increasestate and local agencies’penal- ties when warranted and to get maximum deterrent effect from EPA’S limited resources.Substantial resourceswould be saved and more cases handled if EPAhad greater administrative penalty authority, according to the Assistant Enforcement Counsel,who estimated that EPAwould be able to close many casesfor approximately $1,000 to $25,000 each. In June 1989, the administration announcedproposals for amending sec- Proposed Changesto tion 113 of the Clean Air Act, including providing EPAwith the authority Clean Air Act May to administratively penalize sourcesup to $200,000 for violations of the Improve Penalties, but act. Proposedsection 113(e) specifically requires the economicbenefit of noncompliance to be consideredin determining the penalty amount and Clearer Direction also allows EPAto assesspenalties under section 120 back to the first Needed provable date of the violation. In July and August 1989, legislation was introduced in the House and Senateto carry out the President’s pro- posa1s3Adoption and passageof these proposals appear to give EPAthe enforcement tools and the desired flexibility and would negate limited enforcement authority as a reason if penalties continue to be insufficient to achieve deterrence. The proposed revisions, however, do not addressthe issue of whether state and local enforcement programs can be required to collect penal- ties basedon the economicbenefit violators have obtained. According to senior EPAcompliance officials, such legislative direction is neededto changethe state and local agencies’practice of not adhering to EPA’S civil penalty settlement policy. They pointed out that since the act makes no specific reference to the issue of state and local penalties, these agenciesare likely to contest any EPAeffort to require that eco- nomic benefit penalties be imposed. The EPAofficials said that requiring these groups to assesseconomicbenefit penalties appeared to be the most appropriate and practical remedy because(1) state and local pro- grams will continue to carry out most enforcement actions and (2) EPA’S enforcement resourceswill continue to be limited for the foreseeable future. The EPAofficials said that additional legislative authority and/or direction would provide the necessaryleverage to compel state and local programs to impose economicbenefit penalties. 3TheCleanAir Act Amendmentsof 1989was introducedaaH.R.3030in the Houseof Representa- tives on July 27,1989,and asS.1490in the Senateon August 3,lQSQ.The Senatepassedits version of the cleanair legislation,S.1630,on April 4,1990, and the Housepasseda revisedversion of H.R. 3030on May 23,1QQ0. Page40 GAO/RCED-!JO-165 Air Pollution Violations Chapter 3 Penalty Amounts Not Sufficient to Deter Violations Both the Congressand EPA have recognizedthat one of the most impor- Conclusions tant actions that can be taken to control air pollution from major stationary sourcesis to penalize violators in amounts sufficient to elimi- nate the economicbenefit derived from the violation. Such penalties, if appropriately imposed, remove the economicand competitive advantage that the violator, through his actions, may have gained over facilities that complied with pollution control requirements. More importantly, economicbenefit penalties can serve as strong incentives to major stationary sourcesto ensure that proper measuresare taken to achieve and maintain compliance with pollution requirements and that viola- tions do not occur. However, state and local agencieshave not always appropriately imposed penalties against major stationary sources.BecauseEPA has not used the SIPS or any other vehicle available to it to get state and local programs to commit themselvesto economicbenefit penalties, signifi- cant violators have paid penalties that were not basedon the economic benefit derived and more than half of the violators have not paid any penalties at all. Further, the Agency has taken little action to overfile when penalties were inadequate. As a result, there appears to be little incentive for major sourcesto take all actions necessaryto maintain compliance. The proposed amendmentsto the Clean Air Act, if enacted, will provide an appropriate first step in solving this problem. Both the Senateand Housebills provide EPA with the authority to impose economicbenefit penalties administratively for violations back to the first provable date of the violation, This authority, if properly used, should allow EPA to overfile more efficiently in situations where insufficient penalties have been imposed by state and local programs and to better ensure that vio- lators do not profit from polluting. However, in the long run, the most effective method for ensuring that economicbenefit penalties are appropriately imposed and collected is to have such penalty practices be a part of all state and local air pollution enforcement programs. To achieve this, standards for assessingeco- nomic benefit penalties need to be included in each SIP. The consistent requirement for such penalties in all programs will best ensure that vio- lators are treated equally in all areas of the country and that firms that comply with air pollution controls are not placed at an economic disadvantage. Page 41 GAO/RCED-90-166 Air Pollution Violations , Chapter 3 Penalty Amounts Not Sufficient to Deter Violations Although we believe EPA has the authority to require that economicben- efit penalties be imposed, we recognizethat EPA has to rely on state and local programs with differing enforcement philosophies to identify pol- luters and assesspenalties, and that requiring rigid adherenceto eco- nomic benefit enforcement standards may strain EPA’S relationships with these programs. However, EPA’S increased ability to overfile will, in our opinion, also improve EPA’S ability to persuade state and local pro- grams to impose penalties that reflect the economicbenefit that the vio- lator obtained. Oncestate and local programs realize that EPA will act to increase insufficient penalties- and that the federal government will retain the additional penalty amounts- it is likely that the states will be more willing to amend their SIPS to include economicbenefit penalty pro- visions. Nevertheless,if EPA believes it needsgreater statutory leverage in guiding state penalty practices, it should pursue additional legislative authority. Currently proposed amendmentsto the Clean Air Act contain provisions Recommendationto that provide EPA with administrative penalty authority that would the Congress enable it to take more expeditious actions to ensure that cash penalties are appropriate. We support these proposals and recommendthat the Congressinclude such penalty authority in final Clean Air Act legislation, To better achieve the objective of ensuring that those who violate the Recommendationsto act’s requirements do not gain financially from their actions, we recom- the Environmental mend that the Adminstrator, EPA Protection Agency . use the enforcement authority available now and in the future under the amendedClean Air Act to overfile to the maximum extent possible to increaseinadequate state and local ,penaltiesand . undertake efforts to include specific standards for assessingeconomic benefit penalties in the next round of SIP revisions and, if necessaryand desirable, seek specific legislative endorsementfor such action. Page 42 GAO/RCED-W-166 Air Pollution Violations Page 43 GAO/RCFDfMW56 Air Pollution Violations Appendix I Major Contributors to This Report Peter F. Guerrero, Associate Director Resources, William F. McGee,Assistant Director Community, and John R. Schulze,Assignment Manager Philip G. Farah, Economist Economic Development Division, Washington, D.C. Margie Armen, Senior Attorney Office of the General Counsel, Washington, D.C. Paul A. Latta, Regional ManagementRepresentative Norfolk Regional James R. Beusse,Evaluator-in-Charge Office Robert E. Martin, Site Senior San Francisco Regional Office (OOL1417) Page 44 GAO/RCEDW-166 Air Pollution Violations Ordering Infortttation ‘I’ht~ first. five copies of each GAO report are free. Additional copies art’ J2 etch. Ordthrs should be sent to the following address, accom- 1)anied by a check or money order made out to the Superintendent, of I)ocuments, when necessary. Orders for 100 or more copies to be mailed to a single address are discouttt,ed 25 percent. 1J.S. General Accounting Office I’.(). 130x fiO15 Gail hersburg, MD 20877 Ordt~rs may also be placed by calling (202) 275-6241. Chapter 1 Iutroduction Chapter 1 Introduction Figure 1.2: Continued . . Source: EPA. Page 11 GAO/RCED-90-155 Air Pollution Violations
Air Pollution: Improvements Needed in Detecting and Preventing Violations
Published by the Government Accountability Office on 1990-09-27.
Below is a raw (and likely hideous) rendition of the original report. (PDF)