American Trucking Associations v. Federal Motor Carrier Safety Administration

United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT Argued March 15, 2013 Decided August 2, 2013 No. 12-1092 AMERICAN TRUCKING ASSOCIATIONS, INC., PETITIONER v. FEDERAL MOTOR CARRIER SAFETY ADMINISTRATION AND UNITED STATES OF AMERICA, RESPONDENTS OWNER-OPERATOR INDEPENDENT DRIVERS ASSOCIATION, INC., ET AL., INTERVENORS Consolidated with 12-1113 On Petitions for Review of a Final Rule Issued by the Federal Motor Carrier Safety Administration Erika Z. Jones argued the cause for petitioner American Trucking Associations, Inc. and intervenors in support of petitioner. With her on the briefs were Adam C. Sloane, Richard P. Caldarone, Prasad Sharma, Richard S. Pianka, Paul D. Cullen, Sr., Paul Damien Cullen, Jr., Joyce E. Mayers, Karyn A. Booth, John M. Cutler, Jr., and R. Eddie Wayland. 2 Richard P. Schweitzer and Craig M. Cibak were on the brief for amici curiae American Bakers Association, et al. in support of petitioner. Scott L. Nelson argued the cause for petitioners Public Citizen, et al. With him on the briefs were Allison M. Zieve, Henry M. Jasny, and Gregory A. Beck. William B. Trescott, pro se, was on the briefs for intervenor in support of Public Citizen, et al. Jonathan H. Levy, Attorney, U.S. Department of Justice, argued the cause for respondents. With him on the brief were Stuart F. Delery, Acting Assistant Attorney General, Matthew M. Collette, Attorney, Paul M. Geier, Assistant General Counsel for Litigation, U.S. Department of Transportation, and Peter J. Plocki, Deputy Assistant General Counsel for Litigation. Erika Z. Jones, Adam C. Sloane, Richard P. Caldarone, Prasad Sharma, Richard S. Pianka, Paul D. Cullen, Sr., Paul Damien Cullen, Jr., Joyce E. Mayers, Karyn A. Booth, John M. Cutler, Jr., and R. Eddie Wayland were on the brief for intervenors American Trucking Associations, Inc. et al., in support of respondents. Before: BROWN and GRIFFITH, Circuit Judges, and RANDOLPH, Senior Circuit Judge. Opinion for the Court filed by Circuit Judge BROWN. In this case we consider two challenges to the 2011 Hours of Service (HOS) rule issued by the Federal Motor Carrier Safety Administration (FMCSA). American Trucking 3 Associations, Inc. (ATA), petitioner in Case No. 12-1092, asserts that the new safety-oriented provisions in the final HOS rule are overly restrictive and costly. By contrast, various public interest organizations and individual truck drivers (collectively “Public Citizen”), petitioners in Case No. 12-1113, claim the rule is insufficiently protective of public safety. The agency comes down squarely in the middle, believing it got everything “just right.” Recognizing that the arbitrary and capricious standard is “highly deferential” and “presumes agency action to be valid,” Am. Wildlands v. Kempthorne, 530 F.3d 991, 997 (D.C. Cir. 2008), we conclude that what remains of the 2003 Final Rule after two remands and three rulemakings are highly technical points best left to the agency. We therefore generally affirm the rule and vacate only the agency’s application of the 30-minute break to short-haul drivers. I. BACKGROUND This protracted rulemaking traces its beginnings to 1999, the year Congress passed the Motor Carrier Safety Improvement Act, Pub. L. 106-159, 113 Stat. 1748, and created the FMCSA. Tasked with making the nation’s roads safer, the new agency’s first rulemaking proposed significant revisions to the regulations that had governed trucking operations since 1962. See Hours of Service of Drivers; Driver Rest and Sleep for Safe Operations, 65 Fed. Reg. 25,540 (May 2, 2000) (2000 NPRM). That effort concluded three years later in 2003 with the promulgation of a final rule that increased the daily driving limit from 10 to 11 hours; reduced the daily on-duty limit from 15 to 14 hours; increased the daily off-duty requirement from 8 to 10 hours; and created a new exception to the weekly on-duty limit known as the 34- hour restart. See Hours of Service of Drivers; Driver Rest and 4 Sleep for Safe Operations, 68 Fed. Reg. 22,456, 22,457 (April 28, 2003) (2003 Final Rule). But as is often the case, the interested public did not go quietly. Trucking industry associations and safety-oriented public interest groups long at odds with each other — and the agency — pushed back against the rule. Public Citizen challenged the 2003 Final Rule as arbitrary and capricious. We agreed. See Public Citizen v. FMCSA, 374 F.3d 1209, 1216 (D.C. Cir. 2004). Because FMCSA had “failed to comply with [the] specific statutory requirement” to “ensure that . . . the operation of commercial motor vehicles does not have a deleterious effect on the physical condition of the operators,” 374 F.3d at 1216 (internal quotation marks omitted), we vacated the rule in its entirety, id. at 1223.1 In response, Congress enshrined the 2003 Final Rule as law until FMCSA could promulgate a new rule, see Pub. L. No. 108- 310, 118 Stat. 1144 (2004), which the agency did in 2005, see Hours of Service of Drivers, 70 Fed. Reg. 49,978 (Aug. 25, 2005) (2005 Final Rule). Nearly identical to its 2003 predecessor, the 2005 Final Rule failed to win over agency critics. Interested groups again challenged the rulemaking as arbitrary and capricious, and this Court once more agreed. See Owner-Operator Indep. Drivers Ass’n, Inc. v. FMCSA, 494 F.3d 188, 206 (D.C. Cir. 2007) (OOIDA). But rather than vacate the contested provisions on broad grounds, we rested our holding on two technical shortcomings: the agency’s failure to (1) timely disclose its methodology for determining its time-on-task multipliers, see id. at 201, and (2) “provide a reasoned 1 While the agency’s obvious oversight was “dispositive” of petitioner’s challenge, certain “troubling” aspects of the rulemaking led us to voice our broader concerns in dicta. Id. at 1217. 5 explanation for a number of the methodology’s critical elements,” id. at 203. FMCSA responded in 2008 by reissuing the 2005 Final Rule with supplemental explanations and analysis. See Hours of Service of Drivers, 73 Fed. Reg. 69,567 (Nov. 19, 2008) (2008 Final Rule). Only after dissatisfied parties sought judicial review of the 2008 Final Rule did the agency agree to undertake a more responsive rulemaking. This most recent effort began with the 2010 notice of proposed rulemaking, Hours of Service of Drivers, 75 Fed. Reg. 82,170 (Dec. 29, 2010) (2010 NPRM), and ended in 2011 when FMCSA promulgated the final rule now before the Court. See Hours of Service of Drivers, 76 Fed. Reg. 81,134 (Dec. 27, 2011) (2011 Final Rule). For our purposes, the 2011 Final Rule resembles the earlier rules in all essential respects save for the addition of several new, safety-enhancing provisions: • 30-Minute Off-Duty Break. The 2011 Final Rule bars truckers from driving past 8 hours unless they have had an off-duty break of at least 30 minutes. • Once-Per-Week Restriction. To prevent drivers from abusing the 34-hour restart, the 2011 Final Rule allows truckers to invoke the provision only once every 168 hours (or 7 days). • Two-Night Requirement. To ensure that drivers using the 34-hour restart have an opportunity to get two nights of rest, the 2011 Final Rule also mandates that the restart include two blocks of time from 1:00 a.m. to 5:00 a.m. See 2011 Final Rule at 81,135–36. Unsatisfied, industry associations and public interest groups promptly petitioned for review. 6 II. JURISDICTION & STANDING Because FMCSA is part of the Department of Transportation, the Hobbs Act circumscribes our jurisdiction to hear only those challenges brought by petitioners “aggrieved” by the agency’s final order. 28 U.S.C. § 2344. “Proof of such aggrievement requires a showing of both Constitutional and prudential standing,” Burlington N. & Santa Fe Ry. Co. v. Surface Transp. Bd., 403 F.3d 771, 775 (D.C. Cir. 2005), the burden of which falls squarely on petitioners, see Grocery Mfrs. Ass’n v. EPA, 693 F.3d 169, 174 (D.C. Cir. 2012), as does the obligation to supplement the record to the extent necessary, see Sierra Club v. EPA, 292 F.3d 895, 900 (D.C. Cir. 2002). Prudential standing can be resolved without much fuss. As interested participants in the notice and comment process below, both sets of petitioners have more than adequately established their party status. It is equally clear that all fall within the zone of interest protected by the statute. See, e.g., Reytblatt v. NRC, 105 F.3d 715, 721–22 (D.C. Cir. 1997). Tougher is the question whether all the parties have satisfied the Article III requirements — a concrete and particularized injury shown to be caused by the defendant and capable of judicial redress. See Lujan v. Defenders of Wildlife, 504 U.S. 555, 560–61 (1992). FMCSA says both sets of petitioners have failed to provide the Court with sufficient evidence to support a claim of constitutional standing, but we think the agency only partially correct. An association “has standing to sue under Article III of the Constitution of the United States only if (1) at least one of its members would have standing to sue in his own right; (2) the interest it seeks to protect is germane to its purpose; and 7 (3) neither the claim asserted nor the relief requested requires the member to participate in the lawsuit.” Rainbow/PUSH Coal. v. FCC, 330 F.3d 539, 542 (D.C. Cir. 2003); see also Hunt v. Wash. State Apple Adver. Comm’n, 432 U.S. 333, 343 (1977). We believe ATA, the lone petitioner in Case No. 12- 1092, has made such a showing. “[C]reated to promote and protect the interests of the trucking industry,” the national association has an obvious interest in challenging FMCSA rulemaking that directly — and negatively — impacts its motor carrier members. ATA Br. i. The three public interest group petitioners in Case No. 12-1113, however, have abandoned any claim to associational standing. They have chosen instead to ride the jurisdictional coattails of individual co-petitioner Dana E. Logan, a truck driver “directly regulated by the challenged rule.” Public Citizen Br. 25.2 As a consequence of this litigation strategy, standing for all petitioners in Case No. 12-1113 rises and falls with Logan’s two declarations, the only evidence of its kind in the record. See Pub. Citizen Br. Add. B (Declaration I); Pub. Citizen Reply Br. Add. (Declaration II).3 2 The public interest group petitioners have expressly disclaimed reliance on Mildred A. Ball, a similarly situated truck driver petitioner, see Public Citizen Reply Br. 4 (“Ms. Logan has thus established the three familiar prerequisites to Article III standing . . . . And when multiple petitioners join in a challenge to agency action, the standing of one petitioner is enough. . . .” (citations and internal quotation marks omitted)), and do not claim either Logan or Ball as members of their organizations. 3 In relevant part, those declarations identify Logan as a cross- country commercial driver subject to the HOS rules. Although she tries “to avoid using the 34-hour restart provision to drive more hours” and tries “to avoid driving the full 11 hours allowable per shift whenever possible,” Declaration I, her employer sets a schedule that sometimes requires her to do just that, see Declaration 8 Petitioners contend Logan’s standing to challenge both the 11-hour allowance and the 34-hour reset is self-evident. Not only is Logan regulated, but FMCSA “does not (and could not reasonably) contest that having to work or drive more hours is an injury-in-fact.” Public Citizen Reply Br. 4. The agency concedes Logan’s status as an “individual truck driver[] directly regulated by the HOS rules,” FMCSA Br. 18, but questions the applicability of the traditional wisdom that “there is ordinarily little question” of injury, causation, and redress when “the plaintiff is himself an object of the action (or forgone action) at issue,” Lujan, 504 U.S. at 561–62. The challenged HOS provisions do permit truck drivers to drive or work more hours than would otherwise be allowed, FMCSA reasons, but nothing in the rule affirmatively requires Logan to log those hours. As pled, Logan’s injury stems not from FMCSA’s rulemaking, but from a work schedule that forces her to utilize the hours-increasing provisions — something her employer dictates, not the agency. See FMCSA Br. 18–19. This clever causation/redressibility argument shifts the conversation toward the allegedly injurious actions of a third party, a claim for which “much more is needed” to establish standing, Lujan, 504 U.S. at 562, but we still think it flawed. Cutting against the agency’s argument is the well-established principle that standing will lie where “a plaintiff demonstrates that the challenged agency action authorizes the conduct that allegedly caused the plaintiff’s injuries, if that conduct would allegedly be illegal otherwise,” Animal Legal Def. Fund, Inc. v. Glickman, 154 F.3d 426, 440 (D.C. Cir. 1998), which was obviously the case here. Much in the same way, we think it “a II. In Logan’s view, the 34-hour restart “allows an amount of driving that is unsafe” and the 11-hour period “is too long to drive in a single shift.” Declaration I. 9 hardly-speculative exercise in naked capitalism” to suggest motor carriers would respond to the hours-increasing provisions by requiring their drivers to use them and work longer days. Abigail Alliance for Better Access to Developmental Drugs v. Eschenbach, 469 F.3d 129, 135 (D.C. Cir. 2006). If anything, FMCSA counted on that fact when it increased driver hours and flexibility to help counteract the imposition of new limits and costs elsewhere. The agency cannot now ignore this reality. That does not mean our inquiry has come to an end, however. Although satisfied with Logan’s standing to challenge FMCSA’s decision to employ an 11-hour driving allowance over her preferred 10, we cannot say the same holds true for Logan’s broad challenge to the 34-hour restart. Crucially, whereas Logan has been consistent in her criticism of the 11-hour allowance, she expressed clear, unqualified support for a circumscribed 34-hour restart in a March 4, 2011 comment submitted as part of the 2010 NPRM. As reproduced in her declaration, Logan’s comment stated: I also like the one time a week re-start. In all of my years on the road I have never used [the 34-hour restart] more than once a week. However I think [the proposed once-a- week limitation on the 34-hour restart] is a good idea so it will not be misused. Declaration I. The parties pay little heed to this concession in their briefing, but we think it significant in determining what constitutes an injury for standing purposes.4 4 In a throwaway sentence, FMCSA takes Logan to task for her about-face, calling standing to challenge the 34-hour restart “particularly questionable.” Respondent Br. 19. Petitioners respond with a brief footnote framing Logan’s comment as “reflect[ive of] 10 Given that Logan expressed support for a once-a-week restriction she believed sufficient to eliminate abuse of the 34- hour restart provision, we fail to see how she can now claim injury from FMCSA’s decision to do as she advised and restrict the use of the restart in the 2011 Final Rule to once every 7 days.5 Is it possible Logan was always opposed to the 34-hour restart? Yes, of course. Is it possible Logan’s views shifted in the intervening period such that she now thinks the entire provision unwise? Probably. But hope as petitioners might that we will take such a view of Logan’s concession, we see nothing in the hopelessly sparse record that would justify doing so. Not only did petitioners’ counsel think it unnecessary to supplement Logan’s terse declarations and attempt to explain these comments, they made no effort whatsoever to demonstrate how Ball, the other individual truck driver petitioner — and presumably someone not on the record as supporting FMCSA proposals — was likewise aggrieved. Perhaps counsel thought we would simply assume standing as we did in the prior decisions, but we are unable do so.6 her view that a limited restart was better than an unlimited one,” which “certainly has no impact on her standing.” Public Citizen Reply Br. 11 n.3. In our view, neither conclusory statement proves to be of much help. 5 At an earlier January 22, 2010 FMCSA listening session, Logan stated that use of the 34-hour restart was “adequate” when used after 8 days on the road but not when she returns home “after 10 days or 14 days.” J.A. 365. Logan did not include these remarks in her declaration, but we think the same result would obtain even if she had. 6 We note that “courts are not bound by a prior exercise of jurisdiction in a case where it was not questioned and it was passed sub silentio.” Indep. Petroleum Ass’n of Am. v. Babbitt, 235 F.3d 588, 597 (D.C. Cir. 2001) (internal quotation marks omitted). 11 For the foregoing reasons, then, we turn now to the merits of Public Citizen’s challenge to the 11-hour driving limit and ATA’s claims in their entirety.7 III. ANALYSIS A. Standard of Review Where petitioners challenge agency decisions as arbitrary and capricious, we apply the familiar touchstones set forth in Motor Vehicle Manufacturers Ass’n of the United States v. State Farm Mutual Automobile Insurance Co., 463 U.S. 29 (1983). Deferring as appropriate to the agency’s expertise and looking only for “a rational connection between the facts found and the choice made,” id. at 43 (internal quotation marks omitted), we remain ever mindful that in performing “a searching and careful inquiry into the facts, we do not look at the [agency’s] decision as would a scientist, but as a reviewing court exercising our narrowly defined duty of holding agencies to certain minimal standards of rationality,” Nat’l Envtl. Dev. Ass’n’s Clean Air Project v. EPA, 686 F.3d 803, 810 (D.C. Cir. 2012) (internal quotation marks omitted). B. 34-Hour Restart Having determined that Public Citizen is without standing to challenge FMCSA’s decision to include a 34-hour restart, we are left only to deal with ATA’s far more circumscribed objections to the two limits on the restart’s use: 7 Pro se intervenor William B. Trescott offers nary an argument in his briefs as to why his lobbying activities would establish standing. For this reason, we need not reach the merits of his arguments. 12 the once-per-week (168-hour) restriction and the two-night (1:00 a.m. to 5:00 a.m.) requirement. i. Once-Per-Week Restriction One ostensible problem with the 34-hour restart is that hours-maximizing drivers could strategically employ the restarts to increase the amount of time they spend driving or on duty. A once-per-week restriction forces drivers to wait the full 168 hours before invoking the provision a second time. The math is byzantine, but the net effect of the limit is to reduce the maximum number of hours a driver could spend driving or on duty per 7- or 8-day period. “[T]he maximum driving hours per 7 days,” for example, would drop from 73.9 hours “down to 70 hours, a small but not a trivial reduction.” 2011 Final Rule at 81,158. Since the agency previously argued in favor of an unlimited 34-hour restart, the administrative record reveals some tension between earlier statements and the agency’s present call for additional safety requirements. ATA highlights these inconsistencies as proof the record will not bear FMCSA’s current interpretation (or in the same vein, that the agency has acted arbitrarily in failing to distinguish its “prior positions”). See ATA Br. 36–37, 39. ATA directs its weightiest challenge toward FMCSA’s shifting views regarding the existence vel non of hour-maximizing drivers, a concern FMCSA had initially dismissed as unlikely and unrealistic. See, e.g., 2005 Final Rule at 50,022 (“FMCSA believes the average driver is not, and cannot realistically, drive and work the longer weekly hours, on a regular basis, as described by some of the commenters.”); 2008 Final Rule at 69,570 (“Commenters have not provided nor has the Agency seen any contrary evidence.”). 13 FMCSA responds that “new evidence caused a change in the agency’s view.” Respondent Br. 49. As explained in the 2010 NPRM, drivers and carriers disabused the agency of its previously held views when they “stated at the listening sessions and in their comments that, especially on the road, drivers do indeed take the minimum restart allowed,” with some carriers even acknowledging “that they have used the restart to add one work shift a week.” 2010 NPRM at 82,182; see also Respondent Br. 49 n.12. In light of these new developments, we cannot say FMCSA acted arbitrarily or capriciously in recalibrating the HOS regulation to reflect its changed understanding of how the 34-hour restart is used in practice. Agencies are free to change their views provided they offer reasonable explanations and justifications for their departure, see FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009), which is exactly what FMCSA did here.8 ii. Two-Night Requirement ATA’s broadest criticism operates with the greatest force: the agency has long championed stability (or circadian synchronization) — the maintenance of a consistent, 24-hour daily schedule instead of a constantly shifting or rotating schedule — but this rule is designed to result in circadian de- synchronization. Because a 34-hour restart must now include two 1:00 a.m.-to-5:00 a.m. periods, the rule strongly encourages nighttime drivers who generally sleep during the day to switch to nighttime sleep during a restart. See ATA Br. 43. While FMCSA concedes the dissonance, it notes that it “never championed the maintenance of circadian rhythms above all else.” Respondent Br. 54. The record amply 8 We have given due consideration to ATA’s other, scattered arguments and found them wanting. 14 supports the point. When weighing the costs and benefits of a 44-hour restart that would likewise “encourage drivers to operate on a rotating shift,” 2005 Final Rule at 50,024, the agency observed “there is no conclusive scientific data to guide it in determining which factor (recovery time vs. circadian disruption) is more effective in alleviating fatigue,” id. Even more compelling is the intervening 2010 study concluding “that the 2-night provision works better than 1- night to mitigate driver fatigue in nighttime drivers.” 2011 Final Rule at 81,156. ATA takes issue with the study’s methodology and the conclusions FMCSA draws from it, see ATA Br. 42, but we must unquestionably defer to an agency’s expertise in weighing and evaluating the merits of scientific studies, see, e.g., Transmission Access Policy Study Grp. v. FERC, 225 F.3d 667, 714 (D.C. Cir. 2000). In sum, we think the agency has acted reasonably, if incrementally, in tailoring the restart to promote driver health and safety. C. 11-Hour Driving Allowance A decade after first proposing the 11-hour driving allowance, the 2010 NPRM floated a possible return to a 10- hour limit, then the agency’s “preferred option.” 2010 NPRM at 82,171. But much to Public Citizen’s chagrin — and ATA’s delight — the 11-hour limit prevailed. In announcing FMCSA’s final decision, the 2011 Final Rule explained how the agency had been “unable to definitively demonstrate that a 10-hour limit . . . would have higher net benefits than an 11- hour limit.” 2011 Final Rule at 81,134 (emphasis added). Public Citizen seizes on this language to contest FMCSA’s decision as one “based on a misunderstanding of its authority.” Public Citizen Br. 36. “By demanding proof of cost effectiveness before adopting a rule that would improve 15 safety,” it reasons, FMCSA “did the opposite of what Congress intended” when it limited the agency to consider “costs only to the extent that is ‘practicable and consistent with [its safety and driver health oriented] purposes.’ ” Id. at 38 (quoting 49 U.S.C. § 31136(c)(2)). The argument has some intuitive appeal, but we think Public Citizen reads far too much into this introductory language. FMCSA’s point is far more mundane than its imprecise language would suggest: the agency ran the cost/benefit analysis with an eye toward adopting the 10-hour limit, but recognized that doing so would have been unreasonable and unfounded on the record before it. FMCSA can be criticized for using improvidently absolute and unqualified language, but we cannot say its approach is irreconcilable with congressional intent. This is clear when one considers the rulemaking in its entirety and does not focus exclusively on the largely perfunctory statements in the opening summary. For the same reasons, we reject Public Citizen’s related claim that FMCSA based its decision on an erroneous view of the law when it suggested elsewhere in the introduction that there were not “adequate and reasonable grounds under the Administrative Procedure Act for adopting a new regulation” because there was an “absence of compelling scientific evidence demonstrating the safety benefits of a 10-hour driving limit, as opposed to an 11-hour limit.” 2011 Final Rule at 81,135. Public Citizen advances two additional challenges to the 11-hour limit that call into question the propriety and sufficiency of FMCSA’s cost/benefit analysis. See Public Citizen Br. 39–41, 47–52. As explained in Section III.E., infra, we think those criticisms misguided. 16 D. 30-Minute Break The HOS regulations cover not only the “long-haul” truck drivers who regularly sacrifice days, weeks, and even months with their families to make the American economy go round, but “short-haul” drivers as well, those who operate within a local area. See 2011 Final Rule at 81,141. “In general,” however, “short-haul trucking work has far more in common with other occupations than it does with regional or long-haul trucking.” 2011 Final Rule at 81,175. As typified by the local FedEx delivery driver, id. at 81,145, short-haul positions tend to be “5-day-a-week jobs” with typical “8 to 10 hour[]” work days and, quite often, an opportunity for overtime, id. at 81,175. “Most of the work” tends to be “regular in character” with drivers going to “basically the same places and do[ing] the same things every day.” Id. The record also tells us that “[s]hort-haul drivers rarely drive anything close to 11 hours, and available statistics show that they are greatly under-represented in fatigue-related accidents.” 2005 Final Rule at 49,980. It should come as no surprise then that FMCSA has historically distinguished between long-haul and short-haul operations. Under 49 C.F.R. § 395.1(e)(2), for example, short- haul drivers who operate vehicles that do not require a commercial driver’s license and who operate “within a 150 air-mile radius of the location where the driver reports to and is released from work,” and, among other things, “return[] to the normal work reporting location at the end of each duty tour,” had greater flexibility in scheduling. Two days a week they could drive between the 14th and 16th hour after coming on duty, see 49 C.F.R. § 395.1(e)(2)(iv), whereas long-haul drivers could never drive beyond the 14th hour after coming on duty, see 49 C.F.R. § 395.3(a)(2). Qualifying short-haul drivers were also exempted from the requirement that drivers 17 keep logs recording their “duty status for each 24 hour period.” Id. § 395.8(a). In most other respects, however, short- and long-haul drivers were bound by the same HOS provisions It was against this backdrop that FMCSA’s 2010 NPRM proposed expanding the twice-a-week 16-hour extension to long-haul drivers as well. Because doing so would render § 395.1(e)(2) largely superfluous, the agency also broached the idea of eliminating the exemption altogether: In order to simplify the HOS regulations, FMCSA is considering rescinding paragraph (e)(2) and requiring the drivers who now use it to comply with the standard HOS limits. Although we have not formally included such a proposal in this NPRM, the Agency seeks comments on the effect of eliminating paragraph (e)(2). . . . FMCSA has little hard information about operations currently conducted under paragraph (e)(2); we invite drivers and carriers that utilize this provision to explain how a decision to remove it would affect them. 2010 NPRM at 82,184. When all was said and done, however, the 2011 Final Rule opted not to disturb the status quo. See 2011 Final Rule at 81,136 (“Because FMCSA has dropped the proposed 16- hour provision, the concerns about confusion are moot.”). The § 395.1(e)(2) exemption thus survived, but short-haul drivers did not emerge from the rulemaking unscathed. FMCSA imposed the requirement of a 30-minute off-duty break on both long-haul and short-haul truckers. ATA did not take kindly to the new requirement, which it now challenges on three grounds. 18 ATA first attacks the 2011 Final Rule as “procedurally invalid,” ATA Br. 50, claiming the final rule was not, as this Court requires, a “logical outgrowth” of the proposed rulemaking, see City of Portland v EPA, 507 F.3d 706, 715 (D.C. Cir. 2007). Although ATA gets the standard right, we think its conclusion wrong. The substance of the NPRM was clear: the proposed change to the existing HOS regulations would impose an off-duty break of “at least 30 minutes.” 2010 NPRM at 82,171. That FMCSA would apply that requirement to short-haul truckers was obviously a logical outgrowth of the NPRM, which proposed eliminating the § 395.1(e)(2) exemption and requiring short-haul drivers to “comply with the standard HOS limits.” Id. at 82,184. The only way ATA can circumvent this reality is to manufacture ambiguity in the proposal’s scope, and that is exactly what the association attempts to do by relying heavily — if not exclusively — on the NPRM’s discussion of the § 395.1(e)(2) exemption. See supra. As best we understand the contention, ATA believes interested parties could not have reasonably anticipated FMCSA’s decision to apply a 30-minute off-duty break to short-haul drivers because doing so would constitute a narrowing of § 395.1(e)(2), and the NPRM proposed only the elimination of that exemption, not a narrowing. See ATA Br. 50. We fail to see how any reasonable commentator would have read the NPRM to suggest the agency would not do what it was otherwise free to: apply the 30-minute break requirement to short- and long-haul truckers alike. That several interested parties expressly recognized this possibility only confirms as much. See, e.g., JA 742 (comment opposing “a 30-minute driver break within the first seven hours of driving time” for “short haul heating oil and propane drivers”); 2011 Final Rule at 81,145 (referring to comment by 19 FedEx that “a 30-minute rest break by the 7th hour after coming on duty would . . . hinder local package pickup and delivery drivers”). Changing gears, ATA next argues FMCSA acted arbitrarily and irrationally by failing to explain the agency’s decision to apply the 30-minute break requirement to short- haul drivers. See ATA Br. 50–51. This claim has far more traction. Despite the many paragraphs scattered throughout the multiple rulemakings distinguishing short- and long-haul trucking both in degree and in kind, see, e.g., 2011 Final Rule at 81,141; 2010 NPRM at 82,175; 2005 Final Rule at 49,995– 96; 2003 Final Rule at 22,482, the 2011 Final Rule contains not one word justifying the agency’s decision to apply the new requirement to the unique context of short-haul operations.9 The agency responds meekly, suggesting that the general explanation for the necessity of a 30-minute break “applies, on its face, to short-haul as well as long-haul drivers.” Respondent Br. 79. Because such conclusory, post- hoc rationalization falls far short of what is required under State Farm, we think the offending requirement must be vacated. See 463 U.S. at 50. Finally, ATA challenges as arbitrary and capricious FMCSA’s decision to impose an “off-duty” break. In the association’s view, there is no support for the agency’s claim that “off-duty breaks yield greater reductions in crash risk” than on-duty, non-driving breaks. ATA Br. 46. We find no merit to the claim. FMCSA has more than adequately supported its choice by referencing an intervening 2011 study concluding that “off-duty” breaks provided the “greatest 9 Nor do we read the agency’s 2010–2011 Regulatory Impact Analysis, JA 822–1061, as having adequately considered the cost the 30-minute off-duty break would have on short-haul operators. 20 benefit.” 2011 Final Rule at 81,154; see also JA 790 (“[A]ny break was better than no break, though an off-duty break provided the most benefit.”). ATA is certainly free to quarrel with the study’s methodology, but given our deference to the agency in such matters, we have no reason to doubt that FMCSA made a “reasoned decision based on reasonable extrapolations from some reliable evidence.” NRDC v. EPA, 902 F.2d 962, 968 (D.C. Cir. 1990) (internal quotation marks omitted). E. Cost & Benefits To better account for uncertainty, the agency developed a cost/benefit model that toggled multiple variables — including baseline sleep assumptions (low, medium, or high), baseline percentage of crashes caused by fatigue (7%, 13%, 18%), and discount rates (3% or 7%) — to “calculate[] the net benefits for . . . 18 different scenarios.” Respondent Br. 69. FMCSA repeated this process for each particular constellation of regulations under consideration and then compared the different results. Predictably, Public Citizen and ATA take issue with nearly every aspect of the analysis, beginning with the variables used. In evaluating petitioners’ challenges, however, we keep in mind our obligation to review an agency’s cost/benefit analysis deferentially. It is not for us to undertake our own economic study and substitute the Court’s views for those of the agency. See, e.g., Nat’l Ass’n of Home Builders v. EPA, 682 F.3d 1032, 1040 (D.C. Cir. 2012); Nat’l Wildlife Fed’n v. EPA, 286 F.3d 554, 563 (D.C. Cir. 2002). Thus, while a “serious flaw” or otherwise arbitrary and capricious reasoning can crash an agency’s cost/benefit analysis, Nat’l Ass’n of Home Builders, 682 F.3d at 1040, petitioners’ “burden to show error is high,” id. (internal quotation marks omitted). 21 Having carefully considered both the briefing and the record, we cannot say that burden has been met. Where the petitioners themselves disagree on something as fundamental as the percentage of crashes that can be classified as fatigue-related, compare ATA Br. 20–21 (at or around 2.2%), with Public Citizen Br. 44 (at least 13% with some studies placing the figure at over 30%), we hardly think it proper to second-guess the agency’s decision to employ a 7%-to-18% range. That deference translates equally to FMCSA’s choices regarding baseline sleep assumptions, a determination dependent on parsing studies far beyond our ken, as well as discount rates, see Ohio v. U.S. Dep’t of the Interior, 880 F.2d 432, 465 (D.C. Cir. 1989) (deferring to agency’s decision to adopt a particular discount rate, calling it “first and foremost a policy choice”). To be sure, petitioners are not without their broader challenges to the agency’s cost/benefit analysis, but these largely superficial arguments establish at most that FMCSA made unwise policy decisions, not that the agency acted irrationally or contrary to law. IV. Conclusion It is often said the third time’s a charm. That may well be true in this case, the third of its kind to be considered by the Circuit. With one small exception, our decision today brings to an end much of the permanent warfare surrounding the HOS rules. Though FMCSA won the day not on the strengths of its rulemaking prowess, but through an artless war of attrition, the controversies of this round are ended. For the foregoing reasons, we grant ATA’s petition in part and vacate the rule insofar as it subjects short-haul drivers to the 30-minute off-duty break requirement. In all 22 other respects, the petitions of both ATA and Public Citizen are Denied.