Opinion
Nos. 14–1009 14–1010.
2015-06-05
ALLIANCE OF NONPROFIT MAILERS, et al., Petitioners v. POSTAL REGULATORY COMMISSION, Respondent American Forest & Paper Association, et al., Intervenors.
Paul D. Clement argued the cause for petitioner United States Postal Service. With him on the briefs were Jeffrey M. Harris , Barbara A. Smith , and David C. Belt , Attorney, U.S. Postal Service. Stephen T. Boardman , Attorney, entered an appearance. David M. Levy argued the cause for petitioner Mailer Petitioners and Supporting Intervenors. With him on the briefs were Matthew D. Field , John F. Cooney , David F. Stover , William B. Baker , Matthew D. Field , Ian D. Volner , James E. Anderson , Tonda Rush , Donna Hanbery , William J. Olson , Jeremiah L. Morgan , John S. Miles , and Keith Kupferschmid. Daniel Tenny , Attorney, U.S. Department of Justice, argued the cause for respondent. With him on the brief were Stuart F. Delery , Assistant Attorney General, Michael S. Raab , Attorney, David A. Trissell , General Counsel, Postal Regulatory Commission, R. Brian Corcoran , Deputy General Counsel, Richard A. Oliver , Robert N. Sidman , and Anne J. Siarnacki , Attorneys. Opinion for the Court filed by Circuit Judge MILLETT.
On Petitions for Review of an Order of the Postal Regulatory Commission.
Paul D. Clement argued the cause for petitioner United States Postal Service. With him on the briefs were Jeffrey M. Harris, Barbara A. Smith, and David C. Belt, Attorney, U.S. Postal Service. Stephen T. Boardman, Attorney, entered an appearance. David M. Levy argued the cause for petitioner Mailer Petitioners and Supporting Intervenors. With him on the briefs were Matthew D. Field, John F. Cooney, David F. Stover, William B. Baker, Matthew D. Field, Ian D. Volner, James E. Anderson, Tonda Rush, Donna Hanbery, William J. Olson, Jeremiah L. Morgan, John S. Miles, and Keith Kupferschmid.
Daniel Tenny, Attorney, U.S. Department of Justice, argued the cause for respondent. With him on the brief were Stuart F. Delery, Assistant Attorney General, Michael S. Raab, Attorney, David A. Trissell, General Counsel, Postal Regulatory Commission, R. Brian Corcoran, Deputy General Counsel, Richard A. Oliver, Robert N. Sidman, and Anne J. Siarnacki, Attorneys.
Before: BROWN, MILLETT and WILKINS, Circuit Judges.
Opinion for the Court filed by Circuit Judge MILLETT.
MILLETT, Circuit Judge:
“Neither snow nor rain nor heat nor gloom of night stays these couriers from the swift completion of their appointed rounds.” But a bad economy might. Or so the Postal Service worried when the recent recession caused mail volumes—and thus Postal Service income—to plummet precipitously. Citing exigent economic circumstances, the Postal Service sought a 4.3% rate increase from the Postal Regulatory Commission.
United States Postal Service, Unofficial Motto, https:// about. usps. com/ who- we- are/ postal- history/ mission- motto. pdf (Oct. 1999).
The Commission agreed that the recession that started in 2008 was an “extraordinary or exceptional circumstance” that warranted some rate increase, but the Commission only permitted the Postal Service to recover $2.8 billion in lost revenue. The Commission reasoned that, by 2011, the Postal Service should have adjusted to a “new normal” business environment in which mail volumes appeared to be permanently lower than their pre-recession levels. The Commission also concluded that lost mail volumes could only be counted in the first year they occurred, even before the “new normal” arrived.
The Postal Service says the Commission's decision did not go far enough; mailer industry groups say the Commission went too far; and the Commission says it got the order just right. We hold that the Commission's “new normal” determination is reasonable, but its rule that lost mail volumes should be counted only once makes no sense on this record. We therefore grant the Postal Service's petition for review in part. Finally, because the Commission's econometric analysis was well within the wide bounds of agency expertise, we deny the separate petition for review filed by representatives of the mailing industry.
I
Statutory Framework
Since the founding of the Republic, the Postal Service has been charged with “bind[ing] the Nation together through the personal, educational, literary, and business correspondence of the people.” 39 U.S.C. § 101(a). The Postal Service does so by providing “prompt, reliable, and efficient service to patrons in all areas and * * * all communities,” id., while charging “uniform [prices] throughout the United States, its territories, and possessions,” id. § 404(c).
Since 1970, the Postal Service has been a government-owned corporation, which Congress expected to be largely self-sufficient financially. See Pub.L. No. 91–375, 84 Stat. 719 (1970). In the Postal Accountability and Enhancement Act of 2006 (“Accountability Act”), Pub.L. No. 109–435, 120 Stat. 3198, Congress created the Postal Regulatory Commission to oversee and administer a pricing regime for the Postal Service. 39 U.S.C. §§ 502(a), 3622(a). In addition, Congress imposed a price cap on Postal Service charges to “create predictability and stability in rates” while “maximiz[ing] incentives to reduce costs and increase efficiency,” 39 U.S.C. § 3622(b)(1) & (2). Postage rates for “market-dominant products”—that is, products over which the Postal Service enjoys either a statutory or practical monopoly (such as first-class mail and periodicals)—may rise only with the rate of inflation. See id. § 3622(d)(1)(A).
But hard times can call for hard measures. So Congress created a safety valve in the new pricing system that allows the Service to raise rates for market-dominant products above the inflation level if the Commission determines that an increase is warranted “due to either extraordinary or exceptional circumstances.” 39 U.S.C. § 3622(d)(1)(E). To permit such an exigent rate change, the Commission must find, after notice and public comment, that “such adjustment is reasonable and equitable and necessary to enable the Postal Service, under best practices of honest, efficient, and economical management, to maintain and continue the development of postal services of the kind and quality adapted to the needs of the United States.” Id.
II
Procedural History
Round One
The Postal Service filed its first request for an exigent rate increase in 2010. The Postal Service claimed that the last recession (which the Commission dubs the “Great Recession”) caused a “dramatic, rapid, and unprecedented decline in mail volume,” and sought to raise prices by more than five percent. See Exigent Request of the U.S. Postal Service, Postal Regulatory Commission Docket No. R2010–4, at 1 (July 6, 2010).
The Commission denied that request. Although it agreed that “the recent recession, and the decline in mail volume experienced during the recession” counted as an “extraordinary or exceptional circumstance,” the Commission concluded that the Postal Service had not shown that it needed a rate increase “due to” the recession. Postal Regulatory Commission, Order Denying Request for Exigent Rate Adjustments, Order No. 547, Docket No. R2010–4, at 3 (Sept. 30, 2010) (quoting 39 U.S.C. § 3622(d)(1)(E)). In particular, the Postal Service had failed “to quantify the impact of the recession on postal finances, address how the requested rate increases relate to the recession's impact on postal volumes, or identify how the requested rates resolve the crisis at hand.” Id. at 4.
The Postal Service petitioned for review. This court affirmed the Commission's determination that “the plain meaning of [39 U.S.C. § 3622(d)(1)(E) ] requires a causal relationship between the exigent circumstances and the proposed rate adjustments.” United States Postal Service v. Postal Regulatory Comm'n, 640 F.3d 1263, 1267 (D.C.Cir.2011). The court also held, however, that the phrase “due to” in the Accountability Act was ambiguous, since it “can mean ‘due in part to’ as well as ‘due only to.’ ” Id. at 1268. The court accordingly remanded to the Commission “to fill the statutory gap by determining how closely the amount of the adjustments must match the amount of the revenue lost as a result of the exigent circumstances.” Id.
Round Two
On remand, the Commission issued Order 864, in which it interpreted “due to” to mean “that exigent rate adjustments are permitted only if, and to the extent that, they compensate for the net adverse financial impact of the exigent circumstances.” Postal Regulatory Commission, Order Resolving Issues on Remand, Order No. 864, Docket No. R2010–4R, at 25 (September 20, 2011). In demonstrating that causal linkage, the Commission elaborated, the Postal Service must “exclude non-exigent impacts, such as on-going electronic diversion of mail volumes.” Id. at 42.
Under that standard, the Postal Service must “[q]uantify the net adverse financial impact of the exigent circumstances” to justify an exigent rate increase. Order 864, at 25. The Postal Service need not achieve “absolute precision” in its calculations, but larger proposed increases require “more rigorous estimation techniques” and a “persuasive showing that the sums sought are the result of the exigent circumstances.” Id. at 44. The Commission concluded that its standard was workable because the Postal Service, in the past, had “demonstrated an ability to develop and refine methodologies for measuring and projecting costs in a variety of Commission proceedings” by using “a volume forecasting methodology that enables it to distinguish and account for the impact of multiple factors that have affected First–Class Mail volumes.” Id. at 50.
No party challenged Order 864 when it came out, and all parties agree that it provides the framework for this case. See Brief for Petitioner United States Postal Service at 8, United States Postal Service v. Postal Regulatory Commission (No. 14–1010); Brief for Petitioners Alliance of Nonprofit Mailers et al. at 37, Alliance of Nonprofit Mailers et al. v. United States Postal Regulatory Commission (No. 14–1009); Brief for Respondent United States Postal Regulatory Commission at 13, United States Postal Service et al. v. United States Postal Regulatory Commission (Nos. 14–1009 & 14–1010).
Round Three
Two years later, the Postal Service renewed its request for a rate increase, seeking an open-ended 4.3% increase in rates. Renewed Exigent Request of the U.S. Postal Service, Postal Regulatory Commission Docket No. R2010–4R, at 2 (Sept. 26, 2013). To demonstrate that the increased rate was “due to” the recession, the Postal Service provided an econometric analysis prepared by its designated expert, Thomas Thress. Thress used a methodology based on “a set of calculations which underlie all of the Postal Service's demand equation analysis and volume forecasts.” Further Statement of Thomas E. Thress on Behalf of the United States Postal Service, Postal Regulatory Commission Docket No. R2010–4R, at 5 (Sept. 26, 2013).
After receiving comments from interested parties, including industry groups and major mailers, the Commission granted the Postal Service's requested rate increase in part. Postal Regulatory Commission, Order Granting Exigent Price Increase, Order No. 1926, Docket No. R2013–11 (December 24, 2013) (Order 1926). The Commission reaffirmed that the recession qualified as an extraordinary or exceptional circumstance warranting a rate increase. Id. at 44. But the Commission disagreed with the Postal Service about the amount of lost volume that the recession itself caused on a forward-looking basis. Id. As a result, the Commission allowed the full 4.3% increase, but ordered that it could only last as long as necessary to yield $2.8 billion of additional profit—a period of less than two years. Id. at 181.
As relevant here, the Commission's decision rested on two distinct determinations. First, the Commission rejected the Postal Service's view that the loss of mail volume could be attributed to the recession unless and until the volume returned to levels consistent with pre-recession trends. Order 1926, at 85. Instead, the Commission decided that mail-volume loss could no longer be considered “due to” the exigencies of the recession once a “new normal” in operational levels was achieved. Id. at 83–94. In the Commission's view, that “new normal” was established once “all or most of” four conditions were met: “(1) the disruption to a sufficient number of relevant macroeconomic indicators demonstrate[d] a return to near historic positive trends; (2) application of the macroeconomic variables accurately project[ed] change, and the rate of change on Postal Service mail volume is positive; (3) the Postal Service regain[ed] its ability to predict or project mail volumes following an extraordinary or exceptional event; and (4) the Postal Service demonstrate[d] an ability to adjust operations to lower volumes.” Id. at 86. The new normal, the Commission added, arrived at different times for different classes of mail, id., ranging between the start of fiscal year 2010 and the start of fiscal year 2012, id. at 94.
Secondly, despite having just found that the Postal Service did not regain its ability to predict or project mail volumes or to adjust operations to lower volume levels until the “new normal” was achieved, Order 1926, at 86, the Commission announced that it would only count decreased mail volume one time, and that would be in the first year in which it was lost. Order 1926, at 96. The Commission said that it would disregard any enduring loss of mail volume after that one-year cycle because, in all “subsequent years, the Postal Service is aware of that loss” and should “adjust[ ] its expectations to continue without that mail piece.” Id.
In rejecting the Postal Service's request, the Commission also begged to differ with certain aspects of Thress's econometric analysis. Thress's econometric technique, known as “backcasting,” “start[ed] with a specific outcome”—the decreased mail volume—“and then work[ed] backwards in an attempt to determine the causes of that outcome.” Order at 61–62 n. 53. In an econometric forecasting model like Thress's, “postal mail volume is set as a function of multiple independent variables.” Id. at 62.
The bottom-line problem for the Postal Service, the Commission explained, was that some of Thress's variables made more sense than others. In particular, Thress's assumption that the recession alone could be blamed for lost mail volume year after year over-strained the “due to” test. Thress reached his result by using both “linear” and “non-linear” “intervention variables” in his models. The intervention variables “refer to any change in the level or trend of mail volumes which starts at a particular time.” Order 1926, at 48 n. 32. Although the Commission acknowledged that intervention variables can be an appropriate part of an econometric model, “the usual practice is to include them only when it is clear what they represent.” Id. at 71. Thress, however, used variables that were “often ambiguous in the sense that there is no definitive way to identify the causes for the effects that these variables or trends capture.” Id. Worse still, the Commission explained, was that while “the Great Recession is a cyclical event,” the “linear intervention trends that Thress attributes to the Great Recession continue in a negative direction forever.” Id. at 82 & n. 71. Whatever changes in mail volumes those variables may have picked up, the Commission concluded, “[t]here is nothing [in the record] that suggests changes in long run trends or changes in the rate of electronic diversion are due to the Great Recession.” Id. at 81.
By contrast, the Commission determined that most of Thress's non-linear intervention variables passed muster because they “exhibit[ed] characteristics of cyclical variables and shift[ed] to a positive impact on mail volumes that coincides with the point in time that the macroeconomic variables used by the Postal Service in their corresponding demand equations begin to improve.” Order 1926, at 82–83. Unlike the linear intervention variables, in other words, the non-linear variables did not imply that the recession would continue costing the Postal Service volume for the rest of time. Instead, they fit the down-and-then-back-up pattern of the recession itself.
Having backed out those aspects of the Postal Service's analysis that it considered deficient, the Commission calculated that roughly $2.8 billion in losses could be attributed to the recession. Order 1926, at 106. The Commission then ruled that recouping that amount through a rate increase was “reasonable and equitable and necessary,” id. at 107, 147, because the “the Postal Service's liquidity levels are so low that they pose an unreasonable risk to the Postal Service's continued operation,” id. at 122.
III
Analysis
Both the Postal Service and an array of groups representing major mailers sought this court's review of the Commission's order. We review those challenges under the familiar rubric of the Administrative Procedure Act, 5 U.S.C. § 706(2). See39 U.S.C. § 3663 (applying APA standards to review of Commission actions). Accordingly, we must uphold the Commission's decision unless it is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A).
Applying that deferential standard of review, we uphold most of Order 1926 as neither arbitrary nor capricious, and as supported by substantial evidence. We reverse, however, the Commission's determination that lost mail volume can only be counted for one year, as the rationale that the Postal Service should have been able to identify and adjust to that downturn immediately is at war with the Commission's “new normal” holding, which openly endorsed a longer period of time for such adjustments.
The “New Normal” Causation Determination
The Commission concluded that, while the “Great Recession” constituted an exigent circumstance warranting some rate relief, the effects of the recession on the Postal Service would only be caused by, or “due to,” those “extraordinary or exceptional circumstances” within the specific meaning of 39 U.S.C. § 3622(d)(1)(E) until the Postal Service had an opportunity to adjust to the “new normal” in the mail economy. Order at 83–94. That “new normal” arrived and cut off the causal “due to” linkage between the exigency and its economic impact, the Commission ruled, once (1) macroeconomic indicators “demonstrate[d] a return to near historic positive trends”; (2) macroeconomic variables “accurately project[ed] change, and the rate of change on Postal Service mail volume [became] positive”; (3) the Postal Service “regain[ed] its ability to predict or project mail volumes”; and (4) the Postal Service “demonstrate[d] an ability to adjust operations to lower volumes.” Id. at 86.
Because the scope of the Accountability Act's “due to” causation standard is ambiguous, Congress left it to the Commission to “determin[e] how closely the amount of the adjustments must match the amount of the revenue lost as a result of the exigent circumstances.” United States Postal Service v. Postal Regulatory Comm'n, 640 F.3d 1263, 1268 (D.C.Cir.2011). The only question before us is whether the Commission's use of the “new normal” to measure causal effect falls within the permissible bounds of reason. See Chevron, U.S.A. Inc. v. Natural Resources Defense Council, 467 U.S. 837, 843, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984); United States Postal Service, 640 F.3d at 1266. We hold that it does. Given the Accountability Act's central focus on tightly restricting Postal Service rate increases and increasing efficiency, the Commission sensibly concluded that the statutory exception allowing higher rates when needed to respond to extraordinary financial circumstances should only continue as long as those circumstances, in fact, remained extra-ordinary.The Commission's “new normal” test is designed to capture precisely the time when the exigent character of a circumstance dissipates—when its effects lose their exceptional character—even though the effects in some literal, but-for causal sense linger. In other words, the Commission permissibly reasoned that just because some of the effects of exigent circumstances may continue for the foreseeable future, that does not mean that those circumstances remain “extraordinary” or “exceptional” for just as long.
The Postal Service advances a number of objections to the “new normal” rule, none of which pass muster.
First, the Postal Service protests that the “new normal” analysis would be better housed not in the Accountability Act's requirement that the increased rate be “due to” exigent circumstances, 39 U.S.C. § 3622(d)(1)(E), but in the Act's separate requirement that any rate imposed be “reasonable and equitable and necessary,” id. The very language of the “new normal” test, the Postal Service argues, shows that the recession continues to cause losses because, if there had been no recession, mail volumes would still be in the old normal.
The “due to” provision in the statute is not as woodenly literal as the Postal Service suggests. To be sure, “due to” looks at causation, and, in at least some sense, “the consequences of an act go forward to eternity, and the causes of an event go back to the dawn of human events, and beyond.” CSX Transportation, Inc. v. McBride, ––– U.S. ––––, 131 S.Ct. 2630, 2642, 180 L.Ed.2d 637 (2011) (quoting W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts § 41, p. 264 (5th ed.1984)). Thus perhaps in some Palsgrafian sense, the effects of the recession may well continue to ripple for as long as the Postal Service's proposed unending rate increase.
See Palsgraf v. Long Island R. Co., 248 N.Y. 339, 340, 162 N.E. 99 (1928).
But the Commission acted well within its discretion in concluding that the “due to” test is concerned with determining the extent of the impact of an extraordinary or exceptional past event. The “reasonable and equitable and necessary” test, by contrast, applies only after exigent causation for a loss has been established and turns on the Postal Service's current need to get back on its feet in the wake of the now-defined exigency. More specifically, the “reasonable and equitable and necessary” test looks to present conditions to determine what the Postal Service requires “to maintain and continue the development of postal services of the kind and quality adapted to the needs of the United States,” 39 U.S.C. § 3662(d)(1)(E), given the realities of the post-exigency marketplace. And that inquiry focuses not on causation, but recovery. The Commission thus appropriately addressed those separate requirements in separate parts of its Order.
Second, the Postal Service contends that the Commission provided insufficient notice of its “new normal” test. That argument fails too. The Postal Service itself was the first to introduce the concept before the Commission, see Postal Regulatory Commission, Responses of U.S. Postal Service to Questions 1–9 of Presiding Officer's Information Request No. 1, Docket No. R2013–11 (Oct. 23, 2013), Question 6, and its own expert even suggested a rough start date of 2010 for the “new normal,” see Transcript of Proceedings Before the Postal Regulatory Commission, Docket No. R2013–11 (Nov. 19, 2013) (Hearing Transcript), at 119 (Thress: “And so to some extent I think it's fair to call maybe 2010 through 2013 the new normal for standard mail.”).
In addition, the concept was discussed extensively before the Commission. See generally Public Representative Comments; Postal Service Reply Comments, Docket No. R2013–11. Thus, although no commenter suggested the precise details of the test that emerged in Order 1926, the Postal Service had ample notice that such a test could emerge logically from the Commission's proceedings.
Third, contrary to the Postal Service's argument, the Commission acted well within its discretion in starting the date of the new normal separately for each class of mail, rather than smaller sub-classes. In his testimony, the Postal Service's expert, Thress, himself suggested that the new normal arrived at different times for different classes of mail, and offered only a fleeting reference to further extending that calculus to different sub-classes. See Hearing Transcript at 117–119. In any event, the “new normal” test turns largely on factors that cut across different classes of mail, such as the state of the macro-economy and the accuracy of the Postal Service's forecasts.
Finally, the Postal Service complains that, in calculating the amount of net losses that the Service incurred even under the “new normal” rule, the Commission was wrong to discount the linear intervention variables in Thress's models. Those variables, the Postal Service insists, captured volume losses caused by the recession. But to prove that claim, the Postal Service can point only to timing. That is, because intervention variables do not model specific changes in the real world, the best the Service can argue is that its variable started at the right time to pick up recession-based losses.
In a different procedural posture, that argument might gain traction. But not here. The Postal Service bore the burden of showing its net losses from the recession. And substantial evidence supported the Commission's determination that the Postal Service had not proved that its linear intervention variables reliably captured only the effects of the recession. Most glaringly, Thress's models had no separate variable to account for loss of mail volume to the Internet. So if people shifted to email at a faster pace during the recession than before, that effect would have been swept up wholesale in the linear intervention variables as attributable to the recession, rather than as, perhaps, the simple progress of inevitable change.
The “Count Once” Rule
In enforcing a “count once” limitation for lost mail, the Commission refused to recognize the cost to the Postal Service of lost mail volume beyond the year in which it first disappeared. Order 1926, at 96. For example, a worker laid off during the recession might cancel her cable subscription, and no longer pay her monthly bill by mail. The Commission would count that change as a loss of no more than twelve pieces of mail; the Postal Service would count it as lost volume for as long as the recession stands between that worker and her cable subscription. If it takes her four years to find a new job and resubscribe, the Postal Service would count forty-eight lost pieces of mail.
Order 1926 offered two rationales for its “count once” rule: First, the Commission worried that counting mail as lost any year beyond the first “makes it impossible for the Commission to fulfill its statutory mandate to calculate the total amount lost due to the exigent circumstance.” Order 1926, at 95. Second, “once a piece of mail is lost in a given year due to the Great Recession, in subsequent years, the Postal Service is aware of that loss and adjusts it expectations to continue without that mail piece.” Id. at 96.
Neither of those rationales makes sense juxtaposed against the Commission's immediately preceding explanation that the “new normal”—not the arbitrariness of turning a calendar—defines when the Postal Service “regain [ed] its ability to predict or project mail volumes” or to “adjust to the lower volumes.” Order 1926, at 86.
The “new normal” rule also demonstrates that it is entirely possible—not “impossible” at all, Order 1926, at 95—to identify a stopping point for the recession's exigent impact on lost mail volume. The Commission, in fact, did just that in adopting its “new normal” rule. Specifically, to pinpoint when the Postal Service regained the ability to accurately predict mail volumes, the Commission credited Thress's testimony that “when we made a forecast in 2008 and 2009, there were terrible, terrible forecasts. * * * Now, 2011, '12, 13, we're back to a world similar to where we were before in terms of we have a better handle on our forecast.” Order 1926, at 93. To determine the Service's ability to adjust to lost volume, the Commission then considered macroeconomic variables: “[a] good measure of the Postal Service's ability to adjust to changing circumstances is Total Factor Productivity,” a variable that suggested that the Service regained its ability to adjust in 2010—more than one year after the start of the recession. Id. at 94. There is no reason that the same considerations, rather than a mechanical tally of the time passed since the recession, could not guide the Commission's determination of when to stop counting lost mail volume.
In sum, the “new normal” rule was well reasoned and grounded in the evidence before the Commission. It comfortably passes deferential APA review; the “count once” rule's controversion of the new normal rule's premises does not and must be vacated.
At oral argument, counsel for the Postal Service argued that the “new normal” analysis in the Order is also inconsistent with the Commission's analysis of whether the rate increase was “necessary.” See Oral Argument Transcript 19. That argument was not raised in the Postal Service's briefs, and is not properly before this court. The Commission, of course, is free to consider that argument on remand.
The Mailers' Petition For Review
Twenty organizations that engage in extensive mailing as part of their business model (“Mailers”) separately sought review of the Commission's decision, arguing that no rate increase at all should have been allowed. We disagree.
The Mailers open their attack by arguing that the Commission inverted the burden of proof when it accepted some of Thress's conclusions after rejecting his model. But that is not what the Commission did. The Commission did not throw out the entire Thress model as invalid and unreliable across the board, but then cherry-pick portions to rehabilitate. The Commission instead examined each distinct part of the model on its own merits, accepting some parts and rejecting others. Nothing in the law or the record foreclosed the Commission from determining that Thress got it partly right.
Next, the Mailers launch a barrage of highly technical objections to the Commission's econometric methodology. Armed with a bevy of home-baked charts and graphs derived from the Commission's work-papers, the Mailers argue that the Commission confused correlation with causation. Not so. The Commission found causation—it found that the recession did actually cause some exigent loss in mail volume, albeit less than the Postal Service had counted. See Order 1926, at 106. That the Mailers view the same evidence differently is beside the point. The only question before us is whether the Commission's view of the data as evidencing causation was supported by substantial evidence, keeping in mind that we are “particularly reluctant to interfere with [an] agency's reasoned judgments” about technical questions within its area of expertise. NRG Power Marketing, LLC v. FERC, 718 F.3d 947, 953 (D.C.Cir.2013) (internal citation and quotation marks omitted).
The Mailers' objection to the Commission's reliance on Thress's treatment of some variables' “trend” and “cyclical” components suffers the same fate. The Commission reasonably explained that the “trend” and “cyclical” labels were “somewhat misleading since both components typically respond to the business cycle when using macroeconomic variables.” Order 1926, at 71. So the Commission considered the trend components of some variables, and the cyclical components of others, id. at 73, and in doing so credited Thress's analysis of his employment variable, over the Mailers' objection that the “trend” component accounted for the majority of that variable's effect on mail volumes. That judgment falls solidly within the Commission's wheelhouse.
The Mailers' rely on Tex Tin Corp. v. EPA, 992 F.2d 353 (D.C.Cir.1993), to argue that the Commission confused correlation with causation. That reliance is misplaced. In Tex Tin, the agency error was to find causation in the face of an obvious and substantial alternative cause of the phenomenon at issue. Id. at 356. The Mailers point to nothing similar in this record, and given the complex causal determination to be made in this case, we see no fatal error in the Commission's analysis of the close fit between macroeconomic measures and the non-linear intervention variables.
In light of the Mailers' arguments, it bears emphasizing that this court is not a rubber stamp for agency actions, but neither are we a peer review board for an academic journal of econometrics. See City of Los Angeles v. United States Dep't of Transportation, 165 F.3d 972, 977 (D.C.Cir.1999). We are, instead, “a panel of generalist judges obliged to defer to a reasonable judgment by an agency acting pursuant to congressionally delegated authority.” Id. The Commission's analysis of the econometric evidence was reasonable; we need decide no more.
Last and certainly least, the Mailers complain that the Commission did not properly respond to their rather sensational claim that no recovery for pre–2012 losses was necessary because the Postal Service managed to muddle through without discontinuing operations. But nothing in the statute forecloses the Commission's eminently sensible determination that the “extraordinary or exceptional” circumstances provision can apply to exigencies that fall short of a death knell.
IV
Conclusion
We grant the Postal Service's petition for review in part, vacate the “count once” portion of the Commission's order, and otherwise deny the petition. We also deny the Mailers' petition for review. The case is remanded for proceedings consistent with this opinion.
So ordered.