EPA's New Policy on "Adjacency"
In looking over recent developments, I realized that we have never addressed Acting Assistant
Administrator Anne Idsal's November 26, 2019 guidance entitled “Interpreting ‘Adjacent’ for New Source Review and Title V Source Determinations in All Industries Other than Oil and Gas” (hereinafter Adjacency Policy), in which EPA states that it will “apply the Agency’s original interpretation expressed in the 1980 development of the Prevention of Significant Deterioration (PSD) portion of the NSR program, where we focused exclusively on proximity when considering whether properties are adjacent.” In so doing, the EPA rejected its prior practice of looking at physical proximity and functional relationship or interrelatedness in determining adjacency. The EPA encourages, but does not require, EPA-approved NSR programs to use this definition and reiterates that “source determinations are made by permitting authorities on a case-by-case basis after consideration of the relevant administrative record.” Adjacency Policy, at 1-2.
The definition of source flows from the definition of “building, structure, facility, or installation,” which is defined to mean “all of the pollutant emitting activities which  below to the same industrial grouping,  are located on one or more contiguous or adjacent properties, and  are under the control of the same person (or persons under common control).” 40 C.F.R. §§ 52.21(b)(6) (federal PSD), 51.165(a)(1)(ii) (state PSD), 51.166(b)(6) (NNSR). The regulations do not further define “contiguous or adjacent” and the EPA has long considered guidance from the D.C. Circuit that “source” should be “understood to approximate the ‘common sense notion of a plant.’” Adjacency Policy at 3-4, citing Alabama Power v. Costle, 636 F.2d 323 (D.C. Cir. 1979).
In promulgating the PSD regulations in 1980, the EPA declined to set a specific distance within which facilities would be considered "adjacent," relying on a case-by-case approach and the three factors set forth above, but also stated that “it did not intend that a single source include activities that were many miles apart, as may be the case, for instance, with multiple sources located along the same pipeline or transmission line.” Adjacency Policy at 4 (discussing 45 Fed. Reg. 52676, 52695 (Aug. 7, 1980). The EPA did not change this interpretation when it promulgated the Title V permit regulations and simply adopted the PSD definitions. See 61 Fed. Reg. 34210 (July 1, 1996).
Over time, the EPA has looked at functional interrelatedness, for example, finding two General Motors operations separated by a mile but connected by a dedicated rail line and constituting a shared production line were “adjacent.” The EPA has also required sources, once considered single, to be kept together because they may have shared offsets with each other. Adjacency Policy at 5. The EPA’s policy began to change in 2007, when it issued a memorandum for the oil and gas industry that focused on physical proximity. This memorandum was withdrawn in 2009. In 2012, the Sixth Circuit held, in Summit Petroleum Corp. v. EPA, 690 F.3d 733 (6th Cir. 2012), that the “EPA’s use of interrelatedness in determining whether sources were ‘adjacent’ was unreasonable and contrary to the plain meaning of the term as then used in EPA’s regulations.” Adjacency Policy at 6. Instead, the Sixth Circuit said that adjacent was unambiguous and meant in physical proximity. In response, the EPA issued a memorandum applying the Sixth Circuit’s Summit Petroleum within the Sixth Circuit, but not elsewhere. This memorandum was challenged and invalidated by the D.C. Circuit in National Environmental Development Ass’n’s Clean Air Project v. EPA, 752 F.3d 999, 1009( D.C. Cir 2014), on the grounds that it violated EPA’s regulations in 40 C.F.R. Part 56 that required national uniformity. Those regulations were subsequently revised to allow such inconsistency.
Ultimately, the EPA announced that for industries other than oil and gas, “EPA interprets the term ‘adjacent’ to entail physical proximity, and the perceived ‘functional interrelatedness’ of pollutant-emitting activities is not a relevant consideration in this inquiry.” Adjacency Policy at 7. The EPA explained that “contiguous” means “touching” and that adjacent must mean something further away than “touching” or else it becomes “superfluous.” As EPA stated:
Therefore, in sum, for purposes of making source determinations for NSR and title V, EPA interprets the term "adjacent" to entail physical proximity between properties. From this point forward, EPA will consider properties that do not share a common boundary or border, or are otherwise not physically touching each other, to be "adjacent" only if the properties are nevertheless nearby, side-by-side, or neighboring (with allowance being made for some limited separation by, for example, a right of way). This is inherently a case-specific inquiry where determining the appropriate distance at which two properties are proximate enough to reasonably be considered "adjacent" may vary depending on the nature of the industry involved. Therefore, EPA is not here establishing or recommending a "bright line," or specifying a fixed distance, within which two or more properties will be deemed ( or presumed) by EPA to be in close enough physical proximity to be considered "adjacent." In each case, this determination should ultimately approximate the "common sense notion of a plant." Moreover, importantly, for those properties not in physical proximity to each other, EPA will not invoke the existence of some functional interrelationship to establish "adjacency."
Adjacency Policy at 8. While the EPA stated that delegated programs must use this interpretive approach, see Adjacency Policy at 2, n.4, it stated that EPA-approved NSR and Title V permitting programs are not bound by the decision, so long as they remain within the boundaries of the regulatory language. The EPA also encouraged permitting authorities to apply this interpretation “prospectively and not retroactively” and specifically disclaimed that it would be appropriate for state and local authorities to revisit prior determinations due to upsetting of settled expectations and the burden on agency resources. Adjacency Policy at 9.
The new Adjacency Policy represents a considerable change in prior EPA practice for determining whether sources are co-located. In conjunction with the corresponding changes in the definition of “control” found in the Meadowbrook and Ameresco determinations, the EPA has substantially scaled back the number of situations when facilities will be determined to be a single source and hence forced to “cohabit” in a single PSD source. While environmental groups are doubtless concerned that these changes allow sources to “escape” appropriate regulation under the NSR programs, the policy change reflects some shifts in the American industrial landscape. In the 1970s, primary industry was highly vertically integrated with large companies controlling all aspects of a process from raw material handling through final product distribution. Since the 1970s, vertical integration has fallen from favor, many large industrial complexes have been sold to multiple owners, who specialize in different aspects of the business and who may actually be competing with each other or the apparently "dominant" source. This development makes "integrating" all of these separate entities under a single permit difficult and may leave the nominal permit holder with little practical control over air pollution control decisions. Further, disaggregation brings its own policy challenges, as separate areas of what used to be a single source become "ambient air" and the modeling demonstration to demonstrate that the NAAQS are assured becomes more difficult. Given the evolution in how sources are used and managed, it is unlikely that this statement will be the last by the EPA on this issue.