Inteliquent, Inc . v. Federal Communications Commission and United States of America

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued November 1, 2021 Decided May 27, 2022
No. 20-1471
INTELIQUENT, INC .,
P ETITIONER
v.
FEDERAL C OMMUNICATIONS C OMMISSION AND U NITED
S TATES OF AMERICA,
R ESPONDENTS
On Petition for Review of an Order
of the Federal Communications Commission
Kevin King argued the cause for petitioner. With him on
the briefs were Thomas Parisi, Nicole Antoine, and Ethan A.
Sachs.
Philip J. Macres was on the brief for amici curiae Intrado
Communications, LLC, et al. in support of petitioner.
Sarah E. Citrin, Counsel, Federal Communications
Commission, argued the cause for respondents. With her on
the brief were Robert B. Nicholson and Patrick M. Kuhlmann,
Attorneys, U.S. Department of Justice, and Jacob M. Lewis,
Associate General Counsel, Federal Communications

-- 1 of 17 --

2
Commission. Richard K. Welch, Deputy Associate General
Counsel, entered an appearance.
Kevin D. Horvitz argued the cause for amicus curiae
USTelecom - The Broadband Association. With him on the
brief was Scott H. Angstreich.
Before: P ILLARD and R AO, Circuit Judges, and GINSBURG,
Senior Circuit Judge.
Opinion for the Court filed by Senior Circuit Judge GINSBURG.
GINSBURG, Senior Circuit Judge: Inteliquent, Inc.
challenges the Federal Communication Commission’s rate cap
on the provision of tandem switch services, which are links in
the routing chain for toll-free telephone calls placed from a
landline. To reduce the incentives for regulatory arbitrage and
to encourage companies to transition to lower-cost Internet
Protocol technologies, the FCC set a transitional tariffed rate
cap of $0.001 per minute for tandem switch services.
Inteliquent argues the Commission: (1) ignored its evidence
supporting a rate cap of $0.0017 per minute, (2) impermissibly
delegated its rate cap decision to USTelecom, a trade
association, and/or (3) set the rate cap below Inteliquent’s or
other providers’ costs. For the reasons explained below, we
deny Inteliquent’s petition for review.
I. Background
Toll-free, or 8YY, calls are a type of call for which the
recipient rather than the caller pays. When a caller dials an
8YY number, his or her carrier typically queries a nationwide
database to determine the owner of that 8YY number. Toll-
free callers using a landline operating within a “time-division
multiplexing” (TDM) network — a legacy method of

-- 2 of 17 --

3
transmitting telephone signals — have their calls routed
through a service switching point, also known as a tandem
switch. Even if the network does not use a TDM, routing a toll-
free call made from a landline is at least a three-step process:
(1) the caller’s local telephone company, or local exchange
carrier, picks up the signal; (2) which it routes to a tandem
switch or intermediate switch provider; (3) which then routes
the call to the interexchange carrier (IXC) providing service to
the owner of the toll-free number. Tariffs for toll-free calls run
in reverse of the signal, as depicted below:
This process applies only to landlines; wireless call routing is
different because the Commission’s rules prohibit wireless
providers from paying providers of tandem switching.
In the order under review (the Order), the FCC set a rate
cap on tandem switching services, a common alternative for
setting a rate in the regulation of telecommunications. Prior to
1990, the largest local exchange carriers were regulated under
a “cost-plus” system of regulations, in which the rates they
could charge were based on their costs plus a return on their
invested capital. That approach changed in 1990 when the
Commission adopted an order that created “an incentive-based
system of regulation” in order to “reward companies that
become more productive and efficient, while ensuring that
productivity and efficiency gains are shared with ratepayers.”

-- 3 of 17 --

4
See In the Matter of Policy and Rules Concerning Rates for
Dominant Carriers, 5 FCC Rcd. 6786, 6787, ¶ 1 (1990). The
1990 order modified the tariff review process for incumbent
Local Exchange Carriers — those that were once local
monopolies — by capping their rates and dropping the cap each
year to encourage them to improve productivity in order to
continue to profit. Id. at ¶ 2. Since then, the FCC has used rate
caps in other contexts, including the Order Inteliquent
challenges here. See, e.g., In re: Core Communications, Inc.,
455 F.3d 267, 273 (D.C. Cir. 2006) (discussing rate caps on
carrier charges for delivering a call to an internet service
provider); In the Matter of Rules for Interstate Inmate Calling
Services, Third Report and Order, Order On Reconsideration,
and Fifth Further Notice Of Proposed Rulemaking, WC
Docket No. 12-375, FCC 21-60, at *2 (May 24, 2021) (setting
rate cap for interstate and international calls placed from
prisons).
As part of its push to improve telecommunications
practices, the FCC in 2011 reformed intercarrier compensation
for toll calls. See In the Matter of Connect America Fund, 26
FCC Rcd. 17663 (2011). In that order, the Commission
adopted a presumption that pricing should move to a “bill-and-
keep” model, in which a carrier bills only its own retail
customers instead of billing other carriers in the path of a phone
signal in order to cover its costs. That order did not apply to
toll-free calls, in part because the companies providing tandem
switching services for toll-free calls must receive payment
from either the interexchange carrier or the local exchange
carrier. In the Order under review here, however, the FCC
transitioned “end office charges” for toll-free calls — that is,
charges local carriers bill to IXCs for connecting their users to
the IXC’s users — to a bill-and-keep model over three years,
beginning in July 2021. Because tandem switch providers have

-- 4 of 17 --

5
no retail customers to bill, the Commission did not transition
tandem switching tariffs to bill-and-keep.
After the FCC transitioned the toll market to the bill-and-
keep model, the toll-free calling industry was beset by arbitrage
schemes. These schemes take several forms, but we need
explicate only one of them, “traffic pumping,” to illustrate the
flaws in Inteliquent’s petition. In that scheme, a tandem switch,
local telephone operator, or bulk termination service provider
— any firm that purchases and routes 8YY traffic — has a high
enough price/cost margin to make it profitable for it to pay
others to place robocalls to toll-free numbers and route those
calls over its facilities. The design of the scheme is depicted
below:
The arbitrageur — the LEC in the figure above — profits from
the tariff fee for switching, minus the cost of providing the
switching service and the cost of paying a robocaller to place
the calls. This is possible, in part, because the competitive
intermediate switch carriers’ tariffs are not based upon their
costs; instead, they are capped at the rates charged by the
incumbent LEC, see 47 C.F.R. § 51.911(c). The perversity of
the arbitrage is compounded because the caller, even if it is a
robocaller, does not pay for the toll-free call and therefore has
no incentive to select the provider with the lowest rates.

-- 5 of 17 --

6
In combination, these conditions create a market distortion
because LECs originating toll-free calls not only lack the
incentive to minimize intercarrier compensation and tandem
switching charges, they also have an incentive to inflate those
charges fraudulently through robocalling. As a result, arbitrage
and this sort of fraud was widespread and increasing; indeed,
AT&T submitted a study during the notice and comment period
for the Order showing that 83% of all originating toll-free
traffic in 2019 was part of such a scheme. These schemes cause
system disruptions, congest incoming lines, and thereby impair
carriers’ ability to complete legitimate calls. They also burden
the owners of the toll-free number with sorting real calls from
the many robocalls.
To combat these schemes, the FCC in 2020 adopted the
Order, aimed at making arbitrage unprofitable by bringing the
price of switching toll-free calls to a point closer to the
providers’ costs. Because arbitrageurs rely upon a high per-
call profit margin in order to compensate robocallers, a low
enough rate cap would minimize the incentive to arbitrage.
Although setting the rate cap low enough to discourage
arbitrage without putting providers out of business seems
simple in principle, the Commission’s job was made more
difficult because none of the commentors on its proposed order
submitted data on its costs. Consequently, the FCC was left to
discern as best it could, without information about a key
component of the equation — providers’ costs — the rate cap
that would discourage LECs or tandem switchers from
encouraging, or engaging in, arbitrage. Lacking that
information, the Commission decided to adopt the lower of two
rate caps proposed by commenters; USTelecom, a nationwide
trade association, proposed a cap of $0.001 per minute based
upon a midpoint rate calculated from data provided by its
largest members. Inteliquent proposed a cap almost twice as
high, to wit $0.0017 per minute. Unlike Inteliquent, which is a

-- 6 of 17 --

7
single, independent tandem switch provider, USTelecom
represents a large number and a wide range of market
participants — including LECs, IXCs, and tandem switching
providers. Because USTelecom’s footprint is so broad, and its
members overwhelmingly supported its proposed cap, the FCC
concluded USTelecom’s suggested cap of $0.001 was
reasonable and would not set prices below providers’ costs.
Inteliquent and other tandem switch providers, including
amici Peerless Network, Inc. and Intrado Communications,
LLC objected. They argued that a rate cap of $0.0017 per
minute, based upon a weighted national average of
Inteliquent’s own rates, would better ensure rates were not
capped below providers’ costs.
II. Standard of Review
The parties devote a lot of attention to the degree of
deference we should accord the Commission’s Order but their
dispute is not material to the outcome here.1 The
Administrative Procedure Act “requires that agency action be
reasonable and reasonably explained.” FCC v. Prometheus
Radio Project, 141 S. Ct. 1150, 1158 (2021). “Judicial review
under that standard is deferential, and a court may not
1 The FCC claims we owe the Order greater deference because
it is a transitional or interim order. See AT&T, Inc. v. FCC, 886 F.3d
1236, 1249-53 (D.C. Cir. 2018) (discussing deference to a rate that
did not reach the Commission’s ultimate goal of universal broadband
access); see also Competitive Telecommunications Association
(CTA) v. FCC, 87 F.3d 522, 531 (D.C. Cir. 1996). Inteliquent
counters that the Commission’s interim rate does not change the
degree of deference we afford the agency because, as we also said in
CTA, “[E]ven an interim rule expected to be in place for only a brief
time is subject to review, or agencies would be free to act
unreasonably for that time.” Id.

-- 7 of 17 --

8
substitute its own policy judgment for that of the agency.” Id.
“A court simply ensures that the agency has acted within a zone
of reasonableness and, in particular, has reasonably considered
the relevant issues and reasonably explained the decision.” Id.
Because the Order is “reasonable and reasonably explained”
we need not determine whether additional deference to the
Commission is warranted. See id.
The Administrative Procedure Act requires that we “hold
unlawful and set aside agency action, findings, and conclusions
found to be . . . arbitrary, capricious, an abuse of discretion, or
otherwise not in accordance with law.” 5 U.S.C. §§ 706(2) &
(2)(A). Prior to adopting any regulation, the FCC must
“examine the relevant data and articulate a satisfactory
explanation for its action including a rational connection
between the facts found and the choice made.” Motor Vehicle
Manufacturers Association of U.S., Inc. v. State Farm Mutual
Automobile Insurance Co., 463 U.S. 29, 43 (1983) (cleaned
up). Here this means the Commission “cannot ignore evidence
that undercuts its judgment; and it may not minimize such
evidence without adequate explanation.” Genuine Parts Co. v.
EPA, 890 F.3d 304, 312 (D.C. Cir. 2018).
III. Merits
The FCC adopted the rate cap on tandem switching
services to discourage arbitrage in toll-free calling and
encourage adoption of lower-cost technologies. Although
Inteliquent does not dispute these rationales, it argues that the
Commission acted arbitrarily and capriciously in setting the
rate cap for tandem switching at $0.001 per-minute for three
reasons: (1) the FCC ignored the data Inteliquent submitted
showing its weighted national average rates; (2) the
Commission should not have relied upon USTelecom’s
proposal as much as it did; and (3) the FCC needed to show the

-- 8 of 17 --

9
cap was above providers’ costs. We first discuss the
Commission’s justifications for the rate cap, then turn to
Inteliquent’s arguments against it.
A. Arbitrage and Lower-cost Technology
Inteliquent concedes that the rate cap reduces arbitrage
and encourages adoption of lower-cost Internet Protocol
technologies, the very points the FCC says justify the rate cap.
Inteliquent, however, claims the Commission’s “decision is []
unreasonable because it fails to take [providers’] cost into
account.”
The agency should consider any reasonable cost
information in the record but may then balance policy
considerations beyond costs in setting a rate. See CTA, 87 F.3d
at 529 (citing National Association of Regulatory Utility
Commissioners v. FCC, 737 F.2d 1095, 1137 (D.C. Cir. 1984))
(“The FCC is not required to establish purely cost-based
rates.”). Here the Commission considered but declined to rely
upon the outdated and unrepresentative cost information
Inteliquent submitted in favor of using a reasonable proxy for
costs, namely, the proposal made by USTelecom. The
Commission’s decision to set the rate cap at $0.001 per minute,
then, was reasonable if the record supports its determination
that a lower rate cap would better deter arbitrage or encourage
a shift to lower-cost technologies and was reasonably
calculated to avoid setting the cap below providers’ costs. See
National Rural Telecommunications Ass’n v. FCC, 988 F.2d
174, 178 (D.C. Cir. 1993). Inteliquent says the Commission’s
explanation of its choice merely states the obvious, namely,
that a lower rate “seemed better calibrated to deter arbitrage.”
Inteliquent is correct in saying the Commission’s rationale
is simple but the company ignores that, in its effort to deter

-- 9 of 17 --

10
arbitrage, the FCC wanted to adopt not just the lowest rate cap
but the lowest rate cap above providers’ costs. The
Commission’s explanation for choosing the lower of the two
proposed rate caps to deter arbitrage need not be more
complicated than that because the incentive structure for
carriers to engage in arbitrage is not more complicated. Put
simply, a lower profit margin allows less room to redirect
profits to robocallers in exchange for generating phony toll-free
calls. The FCC’s Order made clear the connection between
arbitrage and its decision to select a lower rate cap, quoting a
submission from Verizon: “[A]s long as 8YY tandem-switched
transport rates remain high . . . there will be strong incentives
for carriers to engage in such arbitrage schemes.” In the Matter
of 8YY Access Charge Reform (“8YY Reform”), WC Docket
No. 18-156, 2020 WL 6055137, at *17 (Oct. 9, 2020).
Therefore, the record supports the Commission’s anti-arbitrage
rationale for selecting USTelecom’s lower rate cap submission
over Inteliquent’s nearly 70% higher submission. Further,
Inteliquent does not challenge the Commission’s finding that a
higher rate cap “could retard the transition to [Internet-
Protocol-based] networks.” Id. at *20. It follows that the
record supports the Commission’s decision to set a rate cap on
tandem switching services as low as possible — though
preferably above providers’ costs — in order to deter arbitrage
and encourage transition to lower-cost technologies.
B. Inteliquent’s Data
We now turn to Inteliquent’s arguments against the
reasonableness of the rate cap. First, Inteliquent claims the
FCC ignored the company’s data and failed to explain its
reason for doing so. In fact, however, the Commission
acknowledged Inteliquent’s submission and pointed out flaws
that undermined the utility of the study Inteliquent submitted.
First, the study was based upon rates Inteliquent charged but

-- 10 of 17 --

11
did not include rates charged by any other carrier; therefore, it
did not demonstrate that Inteliquent’s rates were representative
of tandem switch service providers generally. Second, the
study produced a weighted national average of rates rather than
a weighted national average of costs, which provided the FCC
little if any assistance in setting a rate cap above providers’
costs. Although Inteliquent argues those rates reflected costs
at one time, the Commission responds, correctly, that the
incumbent providers’ rates Inteliquent relies upon have not
been exclusively cost-based since 1990: “those cost studies are
almost three decades old and, given the generally declining
costs of providing telecommunications service . . . almost
certainly overstate carriers’ current costs.” In short, the FCC
did not ignore Inteliquent’s submission; the Commission
considered Inteliquent’s data and reasonably explained its
decision not to rely upon it.
Citing Radio-Television News Directors Association v.
FCC, 184 F.3d 872, 887 n.20 (D.C. Cir. 1999), Inteliquent next
argues that even if its data were old and possibly flawed, the
FCC could not disregard them because it had no more recent or
more credible data to support its rate decision. It is true the
Commission may not disregard relevant evidence, but here
Inteliquent’s evidence was not relevant to the agency’s goals
of reducing arbitrage and encouraging the adoption of lower
cost technology. Indeed, Radio-Television supports this
conclusion because the study we there faulted the FCC for
ignoring directly laid out the potential costs of the challenged
rule. See id. Here, by contrast, Inteliquent’s study was not
nearly so relevant; instead, as the Commission pointed out,
being old and outdated, it was only weakly related to the
current costs of providing tandem switch services.
Most important, Inteliquent never suggested how its rate
data could be used to calculate costs, even old costs; instead, it

-- 11 of 17 --

12
argues in its reply brief for the first time that the Commission
could have used some unspecified means to calculate
providers’ costs from the rate data it submitted and then used
that information to adjust USTelecom’s proposed rate cap. The
FCC was not required to make a complicated, perhaps
impossible, adjustment to USTelecom’s proposal. Instead, the
Commission was required only to explain why Inteliquent’s
submission did not bear upon the problems the Order sought to
address, which is what it did.
Inteliquent next argues the FCC may not dismiss as dated
a study aggregating and weighting the “current rates for
tandem services” because they are still the lawful rates and
presumably, therefore, still just and reasonable. But the FCC
did not approve the prior benchmark rates because they were
the uniquely reasonable rates; the Commission approved them
because it found those rates were within the zone of reasonable
rates. See WorldCom, Inc. v. FCC, 238 F.3d 449, 462 (D.C.
Cir. 2001) (“The relevant question is ‘whether the agency’s
numbers are within a “zone of reasonableness,” not whether its
numbers are precisely right.’” (quoting Hercules Inc. v. EPA,
598 F.2d 91, 107-08 (D.C. Cir. 1978))). That does not mean a
lower rate could not also be within the zone of reasonableness.
Put another way, Inteliquent’s argument would be a reason to
reject the Commission’s rate cap only if the FCC had
previously determined each rate in Inteliquent’s study was the
lowest just and reasonable rate. Because Inteliquent has made
no such showing, its arguments about its data submission do
not show the Commission’s rate cap is arbitrary and capricious.
C. USTelecom’s Proposal
Inteliquent next argues the Commission arbitrarily
adopted USTelecom’s proposed rate cap despite its lack of cost
justification. Inteliquent’s argument assumes the FCC must

-- 12 of 17 --

13
adopt cost-based rates, which we have already seen is not
correct. See CTA, 87 F.3d at 532, supra p. 9. The same is true
of the rates Inteliquent cited in its own data as “cost-based”;
those were benchmark rates, drawn from the tariffs of
incumbent local service providers, and did not reflect the costs
of those incumbent carriers. As noted previously, the
Commission abandoned cost-based ratemaking for its current
rate cap approach so that firms would have the incentive to
lower their costs in order to increase their profits, which was
also one of its reasons for capping rates for tandem switching.
Finally, even if the FCC needed some evidence the rate cap was
above providers’ costs, it reasonably relied upon evidence that
many carriers already provided service at or below $0.001 per
minute, and upon the overwhelming support of the numerous
and diverse members of USTelecom.
Inteliquent next argues the FCC, in accepting
USTelecom’s proffered rate, failed to exercise its independent
judgment. It is true the Commission may neither rely upon a
commenter’s proposal without analyzing the support for that
proposal, see City of New Orleans v. SEC, 969 F.2d 1163, 1167
(D.C. Cir. 1992), nor delegate its decision-making authority to
a private party, see Texas Office of Public Utility Counsel v.
FCC, 265 F.3d 313, 328 (5th Cir. 2001). Here, however, the
FCC did neither. In City of New Orleans v. SEC we vacated an
agency’s finding because it relied upon a submission
containing “no explanation or underlying support” and did not
“ascertain[] the accuracy of the data contained in the study,”
which taken together suggested the agency did not engage in
reasoned decision-making. See 969 F.2d at 1167. Here, in
contrast, the FCC reached its decision by analyzing the various
arbitrage studies and comparing the submissions of
USTelecom, Inteliquent, and others. The accuracy of the data
upon which the FCC relied was not questioned; indeed,
Inteliquent never argues USTelecom’s members were not

-- 13 of 17 --

14
willing to accept $0.001 per minute as the rate cap — and
therefore that the information USTelecom’s members provided
through their revealed preference in accepting the proposal was
inaccurate. It argues instead that the FCC should not have
inferred the members’ support for the $0.001 rate cap implies
their costs are less than $0.001. Considering, however, that
USTelecom’s members are for-profit enterprises, we can
hardly imagine an inference more reasonable. To have
required that the association’s proposal be supplemented with
cost data from its member firms would have complicated the
proceeding to no purpose.
Inteliquent next argues the FCC could not adopt
USTelecom’s findings because they are not representative of
the industry. Specifically, Inteliquent points out the midpoint
rate USTelecom proposed was based upon the experience of
large carriers only. Inteliquent is correct, but USTelecom
members of all sizes nevertheless agreed to the rate, which
implies it was above their costs as well. Indeed, the FCC said
as much at the outset of the Order, recognizing that “not all
carriers have endorsed the USTelecom proposal,” but it has
support of “carriers whose size and business models vary
significantly.” 8YY Reform, 2020 WL 6055137, at *8 n.73. In
addition, USTelecom adopted its suggested rate of $0.001 per
minute with the support of the same incumbent carriers upon
which Inteliquent’s rate study was based; in other words,
Inteliquent’s rate data are no more representative of the
industry as a whole. Id.
Inteliquent next criticizes the Commission’s reliance upon
the support of Bandwidth Technologies, another commenter,
for the $0.001 rate cap on the ground that Bandwidth does not
provide independent tandem switch services. As an initial
matter, the petitioner overstates the Commission’s reliance
upon Bandwidth’s support; the FCC pointed to Bandwidth’s

-- 14 of 17 --

15
assent to the rate cap merely to point out that not only
USTelecom’s members supported its proposal. More
fundamental, Inteliquent’s assertion that Bandwidth does not
provide the same kind of services as Inteliquent is not correct;
it does, but it uses a lower-cost, internet-based technology. In
any event, Inteliquent’s argument that its own legacy
technology deserves special cost consideration conflicts with
the Commission’s stated purpose of encouraging adoption of
lower cost technologies; we cannot substitute Inteliquent’s
preferred policy of maintaining old technology for the
Commission’s contrary policy.
In sum, the FCC did not delegate its decision-making to
USTelecom. Nor did the FCC improperly rely upon
USTelecom’s proposal despite the lack of explicit cost data
supporting the proposal.
D. Cost-justification of the Rate Cap
Inteliquent and the amici argue broadly that the rate cap
the FCC has adopted is below some tandem switch providers’
costs and therefore unreasonable. It is true, as Inteliquent
notes, that the FCC may not ignore “costs that the Commission
acknowledges to be legitimate,” Global Tel*Link v. FCC, 866
F.3d 397, 413 (D.C. Cir. 2017), but as discussed above, other
considerations may justify its departure from cost-based rates,
see CTA, 87 F.3d at 532. Next, as the FCC and USTelecom
point out, Inteliquent never provided evidence that the rate cap
is below its costs, nor does it claim that now. Instead, it argues
the rate cap is below its current weighted national average rate,
but it produced no authority suggesting it is entitled to that rate.
Therefore, it was reasonable for the Commission to conclude
that Inteliquent’s data did not prove it needed to continue
charging its current rates to remain profitable.

-- 15 of 17 --

16
Inteliquent then argues that the agreement of USTelecom’s
membership to a $0.001 per minute rate cap does not
demonstrate that the rate is above-cost because its members
may cross-subsidize by shifting costs to “tandem connection
charges.” Although that is possible, there is no record evidence
that USTelecom’s members will cross-subsidize tandem
switch services capped at $0.001. In addition, the FCC points
out that the independent tandem switch providers among
USTelecom’s members, which could not cross-subsidize,
nonetheless agreed to the rate cap.
Finally, Inteliquent argues the rate cap is below-cost for
some rural carriers, pointing to a letter in the record that so
asserts. As the Commission points out, however, the letter does
not contain any supporting data and Inteliquent made no effort
to reconcile the letter with the support of USTelecom’s rural
members. Finally, even if Inteliquent is correct that the rate is
below cost for a small number of providers, Inteliquent cites no
precedent, and we are aware of none, requiring the FCC to set
a rate cap above the costs of the highest cost provider.
IV. Conclusion
Inteliquent’s petition rests upon weak data and an outdated
approach to price regulation. Incentive-based regulation need
not accommodate the high-cost practices of every regulated
firm, particularly when exigent circumstances, in this instance
widespread arbitrage, provide the impetus for the agency’s
order. Further, Inteliquent’s submission did not show the
Commission’s rate cap was below cost for itself or for any
other provider.
The FCC Order setting the rate cap for tandem switching
services at $0.001 per minute was not arbitrary and capricious;
therefore, the petition for review is

-- 16 of 17 --

17
Denied.

-- 17 of 17 --