4.4. Synthetic Organic Chemical NESHAP
The Regulatory Impact Analysis for the National Emissions Standards for Hazardous Air Pollutants for Source Categories: Organic Hazardous Air Pollutants from the Synthetic Organic Chemical Manufacturing Industry and Other Processes Subject to the Negotiated Regulation for Equipment Leaks
deals with emissions from the Synthetic Organic Chemical Manufacturing Industry, called SOCMI in the document. Petroleum refineries are excluded.
of the Clean Air Act Amendments of 1990 altered the way EPA regulates hazardous air pollutants (HAP):
"Instead of requiring the Administrator to determine which air pollutants ought to be listed and regulated as hazardous air pollutants, Congress provided a list of 189 hazardous air pollutants in the statute itself. … The Agency is now required to develop a list of all categories and subcategories of sources emitting any of the listed pollutants, and develop technology-based standards to control such emissions. Thus, these standards are to be based on the sources of the emissions rather than being set pollutant by pollutant as in the past and are no longer to be risk based." (p. 1-19, emphasis in original)
The regulation is applied to five types of emission points at each source (plant):
"The HON [Hazardous Organic National Emissions Standards for Air Pollutants] regulation covers five types of emission points: process vents, wastewater, transfer operations, storage vessels, and equipment leaks. The regulation is made up of two standards, one covering the first four emission points, and the second covering equipment leaks. The standard for the first four emission points was arrived at by the usual regulatory process, while the equipment leaks standard was developed by regulatory negotiation." (p. iii)
The leak standard also applies to seven other non-SOCMI chemical industries. Ibid., p. 5-80.
Because the 1990 Amendments require quantitative emission regulation but do not set control levels, EPA has determined those control levels for the first four emission points:
"Controls that comprise the best 12 percent of performance for existing sources determine the existing source floor. For new sources, the best controlled similar source is used to determine the floor." (p. 5-81)
The control levels thus set are deemed Maximum Achievable Control Technology (MACT). The regulation requires use of designated reference technologies or an approved alternative. Ibid., p. 5-82.
MACT is the floor level of control consistent with the statute. In fashioning the regulation, EPA also considered more stringent standards for each of the emission points. For example, for large storage vessels, the MACT floor would controls tanks storing HAPs with vapor pressure above 0.75 psia. The next more stringent option would control tanks storing HAPs with vapor pressure above 0.1 psia. The latter was chosen. Ibid., p. 6-118.
The document analyzes costs in a number of dimensions. In one dimension, the compliance cost and cost-effectiveness of controlling each emission point category and several sub-categories are tabulated. Ibid., Tables 6-8 through 6-19. Each table shows cost and cost-effectiveness for an emission point for the MACT option and for the set of increasingly stringent options that were considered for that point. For each option, average cost-effectiveness and incremental cost-effectiveness are shown. Incremental cost-effectiveness tends to increase with more stringent options. For one option, incremental cost-effectiveness was $336,000 per megagram of HAP abated. Ibid., Table 6-15. This option was not the regulatory choice.
In another dimension, the costs of monitoring, record keeping and reporting are tallied. Ibid., p. 6-105. These total $69 million annually and include costs borne by the regulated firms and the federal government.
The document totals the compliance costs for the for the regulatory options chosen and incorporates the monitoring costs. For the fifth year following the proposal, by which time all affected sources will come into compliance, the total annual cost is $227 million. Ibid., p. 6-105 and Table 6-23. This figure includes the cost of abating equipment leaks under the negotiated agreement. In addition to the annual cost, there is a capital cost of $450 million. Both existing sources and new sources constructed in the first five years are included in the totals. The cost-effectiveness is $350 per megagram of HAP and $170 per megagram of VOC (volatile organic compounds).
In addition to the foregoing cost estimates, an attempt was made to project the effect of the regulation on chemical production costs. Chemicals made in regulated plants are made from other chemicals also made in regulated plants. A chemical tree, a kind of input-output matrix, was constructed in an attempt to capture the ripple effect. Ibid., p. 6-100ff. In each instance, the most stringent of the regulatory options considered was incorporated rather than the chosen option. Most chemicals see their costs rise modestly but some small volume chemicals experience such sharp rises that the overall increase averages more than 1,000%! Ibid., Table 6-7.
The regulation provides for averaging within a source, a type of economic instrument:
[The owner or operator of a source may] "average emissions from two or more emission points such that the overall required emission reduction is achieved. With … emissions averaging, the owner or operator does not have to apply the reference control technology to each … point, as long as an equivalent or greater emissions reduction is achieved elsewhere in the source.…"
"Equipment leaks can not be included in the emissions averages…." (p. 2-23)
Because the cost analysis does not examine the possibility of averaging, it may tend to overstate actual compliance cpsts.
Chapter 8 of the document is a discussion of benefits but the authors acknowledge its limits:
"Various limitations prevent a formal, quantitative benefits analysis. A formal benefits assessment requires analysis of the full pollutant path, tracing from the change in emissions to atmospheric dispersion through population exposure, effective dose, physical effect manifestation considering mitigating and averting behavior, and ultimately to economic valuation. Ideally, this analysis should be conducted for each of the air toxic chemicals regulated under the HON. Lack of information on an individual chemical basis, such as the quantity and location of emissions and chemical-specific physical effect and exposure data, as well as a significant time requirement preclude such a detailed benefit analysis." (p. 8-148)
What the document offers instead is an extensive review of the literature on the benefits of air pollution abatement citing more than two dozen works. Conceivably, an attempt at benefits transfer could have made relying on that literature search. However, that was not done, perhaps because the literature does not quantify or monetize benefits “on an individual chemical basis” for 150 toxics but rather tends to do so for broader groups such as VOCs.
Chapter 9 of the document is a two-page discussion of benefit-cost analysis. It begins:
"This chapter provides an illustration of the economic efficiency framework for evaluating regulatory options under the promulgated HON rulemaking. Unfortunately data paucities and time constraints preclude a formal quantification of the allocative efficiency aspects of this rule." (p. 9-161)
There is, however, a suggestion that the HON regulation has the potential to yield real resource economies:
"The average cost of the HON VOC emission reductions for total industry control [the set of the most stringent regulatory options, not the set chosen] in nonattainment areas is estimated at $1,061 per Mg. The average cost of discretionary controls in the moderate, serious, severe, and extreme nonattainment areas is estimated to be $4,540 per Mg. If a megagram of HON VOC emission reductions replaced a $4,540 discretionary control measure megagram, the savings would be $3,479. This is a potential control cost savings of nearly 77 percent." (p. 9-162)
Finally, the document considers the impact of the regulation on small business to determine if a regulatory flexibility analysis (RFA) is required as provided by statute. Ibid., p. 7-142ff. It notes that each of the affected SICs is listed in a reference as “Large-Business Dominated” although firms small by some definition may comprise 15% of the industry. It concludes:
"The standard is not expected to have a significant economic impact on a substantial number of small firms, and a RFA is not required." (p. 7-143)