The EPA Addresses Residual Risk for Hazardous Air Emissions at Refineries

by Victor Flatt

On May 14, 2014, the EPA proposed new rules to control “residual risk” from hazardous air emissions (such as from benzene) at the nation’s petroleum refineries.

The Clean Air Act requires the EPA to calculate whether or not residual risk to human health exists after the agency has put Maximum Achievable Control Technology (MACT) in place to control hazardous air emissions.  Studies have long shown residual risk to the public after MACT was put in place at refineries, and this finding forms the legal basis for this rule.  In particular, the EPA proposes addressing more fugitive emissions, addressing emissions controlled during changes in facility operation, and putting new requirements on storage vessels.

The last EPA rulemaking on residual risk from refineries occurred during the George W. Bush administration (initiated in 2002), and that proposal was controversial in at least three respects.  First, it wasn’t clear that the amount of exposure being measured was accurate, since there were few actual monitors in place.  Second, there was significant disagreement with the EPA’s decision at that time to only reduce residual risk to one excess death in 10,000, though this was legally upheld, and third, the proposed requirements to implement the residual risk controls were all recognized as actually creating profit at refineries because they were failing to recapture valuable chemical during the refining process.

The new proposal takes at least two steps in the right direction by requiring enhanced monitoring at the fence-line of refining facilities, and by proposing that emissions that occur in start-up, shutdown, and disruptions be counted in calculating the risk.

These changes are important because they better reflect actual exposure to the air toxics at issue, and thus provide a better picture of true risk.  Because of improvements in monitoring technology, the ability to actually monitor what is seeping into the environment outside of a petroleum facility is possible, and this rule will require such measurements.  This ensures that given a “residual risk” standard, that there will be a way to better measure and enforce those standards.  The control proposals will cost $40 million on an annualized basis, but given the high number of emissions from start-up, shut-down, and upsets, that were not counted before, this should reduce potential harm.

While industry is already complaining about the burden of adding in these measurements and additional contributions, the EPA is on sound legal footing with the rule, as the CAA gives it great discretion in setting residual risk standards, and the agency has well justified the change in the proposed rule.

However, not all of the proposed rule makes progress.  The biggest failure of the proposed residual risk rule for petroleum refineries is the EPA’s failure to address its “default” residual risk standard of 1 in 10,000 excess cancer deaths per lifetime exposure.  Though this is ostensibly a bigger question than residual risk from just refineries, it is important that the assumption not go unchallenged.  The EPA repeatedly calls the 1 in 10,000 (or 100 in 1,000,000) excess mortality standard a default or benchmark and refers to its use in  prior Benzene cases cited in the 1990 CAA hazardous air pollutant emissions standard. There is nothing that requires the EPA to use this as an acceptable standard.  While it may use such a standard, the 1990 CAA amendments indicate that it could find a lower acceptable risk.  Indeed, in justifying its more stringent requirements for storage tank emissions, the agency cites that it will reduce the risk of cancer incidence further below the 60 in 1 million. (proposed rule, page 308).  But for the rest of possible controls, the agency simply notes that no proposed changes will reduce the lifetime exposure risk overall, leaving 7,000,000 people exposed to an increased cancer risk of 1 in 1 million, and 90,000 people exposed to a cancer risk of greater than 10 in 1 million. (proposed rule, page 301).  While this risk may seem small, it is probably not something the 7 million would be happy with. Moreover, the increased risk of greater than 10 in 1 million falls disproportionately on minority communities.

It is true that over the last several years, we have learned much more about air toxics, and we now know that for some toxics, such as benzene, most exposure comes from small or mobile sources rather than the large petroleum refineries and chemical plants.  (See David Adelman, The Collective Origins of Toxic Air Pollution: Implications for Greenhouse Gas Trading and Toxic Hotspots, 88 Ind. L. J. 273 (2013))  Nevertheless, while this is generally true, it is NOT the case at the most intensive refining areas near Houston, Texas, where a combination of lack of zoning and concentrations of refineries means that most risk comes from these refineries.  For these poor neighborhoods facing such exposure, tightening up the petroleum refinery standards is extremely important, and tightening the residual risk standard would also be helpful.  But even in areas where refineries don’t create the majority of hazardous air pollutant exposure, they still create significant amounts of the health risk (between 10 to 30% of the risk from all emissions).  Simply because some other sources may produce more emissions does not give refineries the right to further degrade the air quality of the general public.

Our basic environmental laws, including the Clean Air Act, are predicated upon the common law principle that we have a right to be free from significant externally imposed harm that is solely for the profit of others.  (Victor Flatt, This Land is Your Land (Our Right to the Environment), 107 W. Va. L. Rev. 1 (2004), ) Because of its change in measuring actual emissions at all times, this rule moves us in the right direction, but the EPA misses an opportunity to protect us more.

© 2016 The Center for Progressive Reform