If an agency uses cost-benefit analysis (CBA) to inform its decision-making, what costs and what benefits should it consider? A case currently before the D.C. Circuit, White Stallion Energy Center, LLC v. EPA, raises this issue. White Stallion suggests a tension between the incentives created by Office of Information and Regulatory Affairs (OIRA) review and those created by judicial review, such that an agency seeking to insulate itself from review from within the administration may end up exposing itself to increased risk of losing upon external (judicial) review.
White Stallion concerns one of the most important and expensive rules the EPA has ever promulgated: a regulation under Section 112(n)(1)(A) of the Clean Air Act (CAA) that requires oil- and gas-fired power plants to reduce their emissions of mercury and other hazardous air pollutants. The benefits of the regulation are staggering: monetized benefits of $37 billion to $90 billion, plus non-monetized benefits above and beyond that range. But the costs are also significant – approximately $9.6 billion per year. Among the issues before the court is whether the statute requires EPA to consider these costs, as industry challengers to the rule contend. EPA argues that Section 112(n)(1)(A) – which permits it to promulgate regulations that are “appropriate and necessary” to address hazards to public health posed by oil- and gas-fired power plants – does not require the agency to take costs into account.
Regardless of how the D.C. Circuit resolves that question, the way in which EPA did consider costs raises an interesting issue with broad implications. EPA did tabulate costs and benefits in its rulemaking, not because it believed the CAA required it to do so, but because Executive Order 13,563 requires all agencies (to the extent permitted by law) to adopt regulations “only upon a reasoned determination that [their] benefits justify [their] costs.” EPA’s use of CBA in this context may be seen as an example of what Jennifer Nou has identified as agency self-insulation: that is, an attempt to insulate the agency’s rule against review from within the administration at OIRA. The tension, however, is that, in insulating itself from review from within the administration – by demonstrating very positive CBA scores – EPA may have exposed itself to liability in external review.
Under D.C. Circuit precedent, agency rulemaking will not be invalidated for failure to conduct a CBA if consideration of cost is not required by statute. But if an agency relies on a CBA in making its decision (even if it is not required to conduct a CBA), the agency’s analysis must be reasonable to survive judicial scrutiny. Here, EPA explicitly did not rely on a CBA to make its “appropriate and necessary” finding; so the strength of the CBA will be irrelevant if the court affirms EPA’s legal theory that section 112 does not require consideration of costs. But if the court rejects this theory, EPA might be in real trouble. This is because EPA’s CBA estimated benefits from mercury reductions totaling just $4 million to $6 million. The vast majority of benefits from the regulation come from “co-benefits” due to reductions in particulate matter, PM2.5. EPA’s decision to regulate oil- and coal-fired power plants was based on health effects caused by hazardous air pollutants, under a provision of the CAA specifically focused on hazardous pollutants. PM2.5 is not a hazardous pollutant. Thus, there is an argument that EPA’s consideration of the benefits of PM2.5 reduction was arbitrary and capricious because these benefits are statutorily irrelevant for the purposes of Section 112.
In this case, the tension between intra-administration review and judicial review of an agency’s CBA will likely remain below the surface. EPA quite explicitly declined to rely on its CBA to justify its “appropriate and necessary” determination, and it has a very strong argument that this determination should be upheld under Chevron. But even if EPA is vindicated in the D.C. Circuit, the underlying tension between intra-administration and judicial review is unlikely to be resolved anytime soon.
This is because the tension between intra-administrative and judicial review highlights a larger problem: the shortcomings of our current environmental laws. Since the Reagan Administration, cost-benefit analysis has gained an increasingly prominent role in agency decisionmaking, and courts (the D.C. Circuit in particular) are increasingly likely to read cost-benefit balancing into statutes.
But the environmental laws have not been revised to reflect these policy choices. Agencies, left trying to make sense of laws that have not been revised in decades, are left in limbo. To pass muster at OIRA, an agency must justify its decisionmaking in terms of costs and benefits. In a case like this one, where the costs are very significant but the benefits are even greater, EPA has every incentive to insulate its rule with a CBA highlighting those great benefits. But reliance on a CBA of this kind may create a risk that a court will invalidate the rule as arbitrary and capricious.
 No. 12-1100 (D.C. Cir.).
 77 Fed. Reg. 9306 tbl.2.
 See Jennifer Nou, Agency Self-Insulation Under Presidential Review, 126 Harv. L. Rev. 1755 (2013). Nou suggests a “simple theory”: Under an anti-regulatory president, agencies will submit CBAs of poor quality to increase the costs of review for OIRA, making it less likely that OIRA will reverse the agency. Under a pro-regulatory president, agencies will submit high-quality CBAs to reduce the costs of review for OIRA, making it more likely that OIRA will approve the agency’s rule. Id. at 1806–07. The agency-OIRA interaction here suggests a complication to Nou’s “simple theory.” The Obama Administration would surely be considered “pro-regulatory,” but EPA might still feel a need to insulate itself because of the sheer magnitude of this rule and its high political saliency.
 OIRA is not the only body within the Administration that participates in review, but “OIRA” is often used as convenient shorthand for this review process. See Cass R. Sunstein, The Office of Information and Regulatory Affairs: Myths and Realities, 126 Harv. L. Rev. 1838, 1856 (2013).
 See Bus. Roundtable v. SEC, 647 F.3d 1144, 1148–49 (D.C. Cir. 2011) (invalidating agency action as arbitrary and capricious because, inter alia, it “inconsistently and opportunistically framed the costs and benefits” of its rule).
 77 Fed. Reg. 9306, tbl.2.
 See 77 Fed. Reg. 9306; CAA § 112.
 See Clean Air Act § 112, codified as amended at 42 U.S.C. § 7412(b)(1) (listing hazardous pollutants).
 See Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 42–44 (1983) (holding that an agency acts arbitrarily when it considers statutorily irrelevant factors in its decision-making).
 Cf. Richard J. Lazarus, Congressional Descent: The Demise of Deliberative Democracy in Environmental Law, 94 Geo. L. J. 619, 629–32 (2006) (noting the lack of environmental legislation since 1990, with particular focus on the absence of new laws addressing substantive environmental issues, including climate change).