United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued September 9, 2014 Decided June 5, 2015
No. 14-1009
ALLIANCE OF NONPROFIT MAILERS, ET AL.,
PETITIONERS
v.
POSTAL REGULATORY COMMISSION,
RESPONDENT
AMERICAN FOREST & PAPER ASSOCIATION, ET AL.,
INTERVENORS
Consolidated with 14-1010
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.
2
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.”1 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.
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
1
United States Postal Service, Unofficial Motto,
https://about.usps.com/who-we-are/postal-history/mission-
motto.pdf (Oct. 1999).
3
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.
4
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
5
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.
6
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
7
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.
8
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.
9
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.
10
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). See 39 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).
11
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
12
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 (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
13
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, 131 S. Ct.
2630, 2642 (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.2
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.
2
See Palsgraf v. Long Island R. Co., 248 N.Y. 339, 340 (N.Y.
1928).
14
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.
15
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
16
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
17
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.3
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
3
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.
18
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
19
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. Unites 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”
20
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.