On April 25, 1997, the Environmental Protection Agency (EPA) officially released for comment two chapters of a regulatory impact analysis (RIA) for the Compliance Assurance Monitoring (CAM) rulemaking. (These two chapters are included in our FTP Library as seciv.zip and secv.zip.) One of these chapters consists of a cost impact analysis, estimating the costs industry will incur nationwide in order to comply with CAM regulations. The cost impact analysis was based primarily on information provided by five State environmental agencies, a survey conducted by EPA in these five States, and projected CAM cost information provided by an EPA contractor in 1995.

On behalf of a client representing industrial sources nationwide, RMB evaluated EPA's CAM cost impact analysis and prepared a report that discussed concerns regarding EPA's cost analysis approach and the accuracy of costs projected by EPA. RMB's report focused on the estimated labor and expenses required to comply with CAM, as calculated by EPA for the 5-State industrial sample. Since nationwide cost projections were directly proportional to the 5-State sample results (a factor of 10 greater), all errors and concerns identified in the report proportionally affected EPA's nationwide projections. For the purposes of estimating realistic costs for each CAM activity and providing specific examples when evaluating EPA's cost estimates, the report presented information and cost estimates applicable to four representative add-on control equipment categories -- multiclones, carbon adsorbers, thermal/catalytic incinerators and wet scrubbers. RMB's projected costs presented in the report for each CAM activity represented the average estimated costs for these four types of add-on controls.

Throughout the cost impact analysis, EPA over-simplified the efforts that will be required of industrial sources, particularly given the link between CAM and "credible evidence" (CE) enforcement that EPA's proposal would create. For many cost estimates, EPA assumed that minimal process changes and minimal other steps would be required to meet CAM requirements. Information obtained by RMB, however, indicates that many States do not intend to accept "minimal approaches." For example, a document issued by Ohio EPA describes monitoring, operation, maintenance and training requirements for a variety of add-on pollution control equipment types. This document clearly indicates that, in Ohio, only a combination of EPA's CAM approaches will be considered acceptable for ensuring compliance with emission standards. Furthermore, industrial sources will tend to avoid simplistic CAM approaches in order to minimize inferred compliance problems that might arise due to CAM system excursions.

EPA seemed unaware of the amount of coordination and interaction that typically occurs in industry between plant managers, plant technicians, engineers, and corporate personnel during the development of regulatory documents such as a CAM plan. EPA also failed to acknowledge that CAM activities will affect some "minor" sources. Sources that are not currently considered major sources may still be involved in the CAM planning process due to anticipated future increases in uncontrolled pollutant emissions.

RMB expressed serious concerns regarding the representativeness of the 5-State database and the 375 emission point 5-state survey conducted by EPA. EPA discusses the representativeness of 38 of the 86 source industrial classifications (SIC), without identifying the 38 SIC, and does not address the remaining 48 SIC. Obviously, if 48 of the source categories are not well represented in this 5-State sample, significant errors probably exist regarding the number of 40 CFR Part 64, Subpart B affected sources nationwide. Furthermore, considering the number of control equipment types and SIC codes included in CAM, RMB does not believe that EPA's 375 emission point (75 points per state), random survey could have provided accurate information regarding the applicability of CAM that could be applied to industrial sources nationwide.

Other concerns expressed by RMB included (1) none of the States had a complete list of Title V sources, (2) the percentage of sources subject to New Source Performance Standards (NSPS) that were considered to have add-on controls was probably bias low, and (3) some of the control equipment classifications and proposed monitoring requirements will be interpreted differently by both State and industry representatives, which will necessarily preclude interstate implementation of a particular monitoring approach for a specific type of control device -- increasing compliance efforts.

Based on an evaluation of the CAM cost impact analysis, and not including the possible impact of any errors associated with EPA's approach, RMB estimates that EPA's projected costs to develop initial CAM plans for the 5-State sample database (and consequently, nationwide) are probably more than ten times lower than the costs industry can realistically expect to incur. EPA's projected initial CAM plan preparation costs are low by an order of magnitude. Annual recordkeeping, reporting and certification activities are low by a factor of six. Annualized monitor response costs and periodic CAM revision/renewal costs are low by factors of two and ten, respectively. EPA's 5-state total projected annual cost to industry for CAM compliance is $13,356,000; RMB's projections are significantly higher, totaling $67,929,000 plus any annual incremental costs associated with quality improvement plans (QIP) and permitting fees required to cover the permitting authorities' implementation costs.

EPA argues in their cost impact analysis that the cost of coming into compliance should not be attributed to CAM requirements, and consequently, it is unclear if any Quality Improvement Plan (QIP) activities should be considered CAM-related. RMB disagrees. There are potentially significant costs associated with QIP development and documentation that are attributable to CAM requirements and that would not otherwise be incurred. For example, CAM requires notification of an excursion to the permitting authority; submittal of an initial QIP with procedures that are adequate for evaluating the reasons for excursions; submittal of a modified QIP that includes recommended actions to resolve the problems; summary reports in all subsequent semiannual reports that describe QIP activities until the QIP is completed; documentation of test results showing that pollutant point emissions are in compliance; and submittal of a modified CAM monitoring approach -- which EPA indicated may require a permit revision -- or a rationale for using procedures previously developed. If each CAM monitor triggers one QIP per 5-year permitting cycle, the estimated annual incremental cost to industry due to QIP implementation will be between $9,705,000 and $16,176,000. One QIP per facility would result in annual incremental costs between $1,410,000 and $2,349,000.

Based on an assessment of one permitting agency program that is similar to CAM, EPA's projected costs for permitting authorities to implement CAM are low by a factor of 40. Even though the CAM program is basically new and untried, EPA has often cited the State of Oregon's experience with a CAM-type program as an example of how CAM can work. The Oregon program was highlighted in EPA's recent nationwide video conference on CAM presented to permitting agency staff across the country. Based on the video conference and a follow-up phone conversation with a member of Oregon's permitting staff, Oregon's labor requirement per CAM plan was 61 hours, a factor of 40 greater than the EPA estimate of about 1.5 hours per CAM plan.

With the direct cost impact to industry conservatively estimated to be a factor of five too low, the omission of potentially significant QIP costs, an estimated cost to State agencies that is low by a factor of 40, and potentially enormous errors resulting from EPA's assumptions, the CAM cost impact analysis appears to be flagrantly inaccurate. So inaccurate that RMB recommended that EPA not proceed with the CAM rulemaking until the true costs of these regulations can be properly assessed. To reassess the existing CAM cost impact analysis, RMB recommended that EPA (1) more accurately determine the number of affected sources, including a representative number of major and minor sources in each SIC, (2) provide cost estimates that reflect realistic costs to industry, including QIP activities, true CAM development and testing costs, costs incurred negotiating CAM approaches following initial submittals, recordkeeping and reporting activities, and costs associated with interstate differences in acceptable approaches, (3) use the State of Oregon's actual CAM-type labor efforts as a basis for estimating true CAM-related costs to State agencies, and (4) account for the impact of credible evidence regulations on industry's CAM activities.

| Home | News | CAM | Training | FTP Library | Projects | Links | Contact | Services | Feedback |

RMB Consulting & Research, Inc.
Last Revised: May 22, 1998