At long last! After nearly 3 years, a panel of the Federal Court of Appeals for the District of Columbia Circuit (or, as we legal folks abbreviate it, the D.C. Circuit) heard oral argument in Verizon’s appeal of the FCC’s “network neutrality” rules. And believe me, after spending the month of August refighting the same annoying retransmission fights of the last 10 years, there is nothing I wanted more than to sit for two hours watching other people re-arguing the same arguments around network neutrality that we have argued for the last ten years — it’s like Groundhog Day for policy wonks.
You can listen to the 2+ oral argument here. I have a shorter and less snarky summary on the PK Blog. But for you fellow telecom wonks who want the gory details with a healthy side order of snark directed at Judges Tatel and Silberman, see below . . . .
I’m never sure which is worse, judges who think they know economics when they don’t, or judges who think they understand technology when they don’t. This is why you are supposed to defer to the expert agency, which actually employs economists and engineers. But I shall save this general rant for another day.
Happily, on the D.C. Circuit, you don’t have to chose. You can get judges like Tatel and Silberman who are ignorant of both technology AND economics (and apparently living in some alternate universe where simplistic economic models replace the horrendously complicated world the FCC actually described in its Open Internet Order).
Remind Me What We’re Arguing About Again?
Trying to sum up the history of network neutrality, how we got here, the arguments pro and con, etc. would take far too long. The FCC has this resource page. My colleague at Public Knowledge, Michael Weinberg, has this refresher on ‘what the heck is network neutrality‘ while Sherwin Siy has this blog post on the actual rules the FCC adopted back in December 2010. PK has also put up a fun and useful interactive timeline that gives the key milestones starting in 2002 when the FCC declared cable modems an “information service.” Bonus points for those who actually want to follow this link to the rules as codified in the Code of Federal Regulations or this link to the actual FCC Open Internet Order.
OK, But Why Did It Take Nearly 3 Years To Get Here?
Generally, a rule must be published in the Federal Register before it can either go into effect or be appealed to a reviewing court (there are some exceptions not relevant here). The FCC published the rule in December 2010. So it should have appeared in Federal Register some time relatively soon after that.
But things hit a snag. The rules have certain reporting requirements for broadband access providers on their network management practices. Before those could go into effect, the Office of Management and Budget (OMB) needed to approve the reporting requirements. Fed Reg refused to publish the rule until OMB gave the OK for it being a final rule. OMB and the FCC had some squabbles. By the time OMB finally signed off on the rule so the FCC could publish it in Fed Reg, we were in September 2011. (Verizon tried to appeal the rules early, got bounced for appealing too early, then filed again in September 2011 once the rules got published).
After that, various procedural games (such as Verizon trying to get the same panel that ruled on the Comcast/BitTorrent case) and sorting out a rather complicated briefing schedule (MetroPCS, an original Petitioner, insisted on having a separate brief from Verizon, we had scheduling issues, then we had to have another round of additional briefing after the D.C. Circuit decided the data roaming case) strung things out further. It helped that Verizon did not file a motion for Stay or Expedited Treatment. So this got put on the backburner of the D.C. Circuit — which remains understaffed due to filibustering nominees and has the usual backlog impacting federal courts.
Taken all together, it has now taken us nearly 3 years to go from the FCC’s decision to the oral argument.
Whatever, We’re Finally Here. What Happened?
Argument got underway at a little after 9:30 a.m. before Judges Rogers and Tatel and Senior Judge Silberman (Senior Judges are retired, but still hearing cases whenever they want because of that whole ‘life tenure’ thing).
FCC Authority — Surprisingly Smooth Sailing For The FCC
The first surprise of the day was that all three judges seemed totally comfortable with the FCC’s assertion of authority under Section 706 (47 U.S.C. 1302). This was a surprise because, for those who remember the actual opinion in the Comcast/BitTorrent case, the court there found the FCC had failed to provide a clear explanation for its authority to order Comcast to stop blocking BitTorrent. In the FCC’s Open Internet Order, the FCC advanced the theory that Section 706 of the Telecommunications Act of 1996, codified at 47 U.S.C. 1302, provided independent authority to the FCC to regulate broadband access providers (independent of any “ancillary jurisdiction”). Section 706(a) requires the FCC to “encourage deployment of advanced telecommunications capability” (which the FCC has long interpreted as “broadband”). Section 706(b) states that if the FCC finds that broadband is not being deployed “in a reasonable and timely manner,” then the FCC “shall take immediate action to accelerate deployment of such capability.”
The FCC also cited other statutes under an “ancillary authority” theory (for a refresher on “what the heck is ancillary authority” see this ancient blog post here). But the most controversial assertion of authority in the 2010 Open Internet Order was the FCC finding explicitly that Section 706 provided separate authority for the FCC to regulate broadband access services, and that adopting the network neutrality rules would encourage broadband deployment and investment in infrastructure.
Most folks expected this to be a tough hurdle for the FCC to overcome since it required the court to agree with the FCC on the statutory interpretation (which would give the FCC straightforward direct authority to regulate broadband access and potentially other IP-based services), and required accepting the FCC’s link between the network neutrality rules and facilitating investment and deployment in infrastructure. Surprisingly, however, the panel seemed quite comfortable with the assertion of authority in Section 706.
Verizon’s counsel, Helgi Walker (who represented Comcast in the Comcast/BitTorrent case) argued strenuously that neither Section 706(a) or Section 706(b) could provide the FCC with authority. Walker argued that it was inconceivable that Congress would invest the FCC with such vast regulatory powers when the FCC had utterly lacked authority to regulate “the Internet” previously. However, as Judge Rogers pointed out, prior to the Telecom Act of 1996, the FCC regulated “the Internet” rather thoroughly under its ancillary authority, so this was not a new. Silberman asked whether, if Verizon had market power, the FCC would be justified in acting to alleviate that market power. (Walker at first seemed to concede the point, but later backed away from this very hard.) Tatel noted that there was plenty of evidence in the record to support either the FCC’s theory that a net neutrality rule would drive deployment and infrastructure investment or Verizon’s argument that a net neutrality rule would discourage deployment and investment, so shouldn’t the court defer to the FCC?
Walker fought back with considerable energy that it was simply inconceivable that Congress would make such a huge grant of authority in Section 706, and that the FCC was simply trying to “bootstrap” its authority because it had no other possible source of authority. (I am not sure why an agency saying “hey, we think this is important and a court told us we need to reexamine our authority so lets see what we got” is a legal argument, but I know that those opposed to regulation always regard this as a cardinal sin whose evil intent voids any other doctrine of administrative law.)
On the whole, however, it appeared that all 3 judges felt the FCC had authority under Section 706 to somehow regulate Verizon under some set of circumstances. The most sympathetic judge for Verizon on this point was Judge Silberman, who thought the FCC fell down on the job by not making an explicit finding of market power (although market power is nowhere mentioned in Section 706, and it is unclear on what authority a finding of market power is necessary for exercise of regulatory authority. But again, this is all the rage among the Federalist Society set. The second circuit went on an endless blather about this last week in the context of the cable program carriage rules. But I shall save my rant about “judges who think they are economists” for another time. For now, I’ll just say that Silberman’s comments about “market power” and how he conceptualizes it (despite also recognizing that Verizon is a “bottleneck facility” between the subscriber and the edge provider without any showing of market power) demonstrate either a complete ignorance or complete indifference for the last 30 years of economic literature.)
I’ll jump out of order to say that when Sean Lev for the FCC got up for his innings, he faced relatively few questions on the Section 706 argument. So, while it is always dangerous to generalize from oral argument, it would seem that the panel will affirm that Section 706 provides the FCC with some authority to do something wit regard to regulating broadband access.
This generally positive result for the FCC has been consistently underweighted in descriptions of the oral argument because of the enormous length of time discussing the “common carrier” rule and the likelihood that the critical component of the net neutrality rules — no blocking and no discrimination — gets either tossed out or modified.
The “Common Carrier” Limitation Argument.
Pretty much everyone knew that if the FCC got past the initial authority hurdle, the next hurdle in the form of the “common carrier” rule that Tatel made up last year in the Data Roaming Case. This spanking new doctrine is so significant (and raises so many collateral issues, especially for the transition of the phone system to IP) that it deserves a blog post and rant all its own. For now, suffice it to say that in the provision Communications Act defining “telecommunications service provider” (47 U.S.C. 153(51)), it says that a telecommunications service provider (e.g., Verizon when providing phone service) “A telecommunications carrier shall be treated as a common carrier under this chapter only to the extent that it is engaged in providing telecommunications services.” A similarly worded provision in Section 332(c)(2) governing mobile services says that anything not classified as “commercial mobile radio service”(CMRS) is “private mobile radio service” (PMRS) and “shall not, insofar as such person is so engaged, be treated as a common carrier for any purpose under this chapter.”
In the Data Roaming Case, Tatel decided that this language creates an absolute statutory bar on regulated broadband services in any way that resembled “core common carriage” obligations. This did not, however, preclude the FCC from imposing (assuming appropriate authority) other regulations that also apply to common carriers. For example, the FCC has a specific statute mandating privacy rules for common carriers, but that does not mean that all privacy rules are “common carrier regulation” and therefore prohibited under the statute.
My PK colleague John Bergmayer explained the potential implications of this decision for the net neutrality case soon after the Data Roaming Case came out in 2012. The court held an extra round of briefing on the implications of the case. Especially when Tatel (who authored the Data Roaming Case) got appointed to the panel for the net neutrality argument, it seemed likely that the Common Carrier Prohibition (as it is now getting called) would get major attention if the FCC made it past the Section 706 argument.
Verizon argues that when an end user subscriber requests information (for example, by entering www.google.com in a browser), then Verizon must go and make the request of Google’s servers for the information and transport the information back to customer. By requiring Verizon to carry Google’s information back to the customer, and prohibiting Verizon from negotiating separately with Google for premium treatment, the FCC requires Verizon to be a “common carrier” for Google with a fixed rate of “zero” for carriage. In plain language, this is a restatement of the “net neutrality gives Google (and other edge providers) a free ride.
The FCC argues no, that totally misconstrues the relationship. Verizon broadband is in the business of getting stuff the customer (with whom it has a contract) requests and transmitting stuff generated by the customer to where the customer wants to send it. Verizon has no relationship with Google or any other edge provider unless Verizon wants to negotiate something, in which case it is free to negotiate all manner of agreements for things like interconnection, content delivery networks, and so forth. As Cogent would certainly argue if it could, Verizon has no specific interconnection obligation and has no problem refusing to interconnect or refusing to exchange traffic when it feels like it. Instead, the FCC argues, the no blocking and no prioritization rules simply means that, if you have decided to enter the specific business defined by the rules as “broadband access service,” once a customer asks Verizon (or any other broadband provider) to do the thing Verizon contracted with the end user to do, i.e., get stuff or send stuff to someone else, Verizon cannot charge the edge provider a second time to actually deliver the packets to the end user, or mess with the packets in any way.
It became clear very early in the game that while Rogers apparently agreed with the FCC (she didn’t say much, but what she said did support the FCC’s conceptualization of the market), Silberman and Tatel agreed with Verizon.
But For These Rules . . .
Judge Rogers pushed back on Verizon’s characterization by arguing that Verizon does not, in fact, have any such agreements with edge providers and there is no evidence that any edge providers want such agreements. Verizon responded that “But for these rules” Verizon would, in fact, be “exploring those commercial arrangements.”
My PK colleague Michael Weinberg has this blog post on why this statement potentially raises concerns. My own comment here for the moment is that Verizon actually is free to “[explore] those commercial arrangements” on the wireless side. I am not saying I would be happy about that, mind you. But, from a rules prospective, the restrictions on blocking and discrimination with regard to wireless only apply to competing voice and and video calling/conferencing aps. (This was why many other public interest org refused to support the rules — because the wireless provisions are so much weaker than those for wireline.) Despite this, in the last 3 years, the closest anyone has come to cutting a deal with Verizon Wireless for preferential treatment has been tentative discussion with ESPN to pay people’s data cap overage costs.
So Rogers point that there does not exactly appear to be much demand for this market appears well taken. This is relevant to the argument that the FCC (tried) to make during its time that the agency is allowed to take the market as it exists now and make rules based on its predictive judgment how changes to that market would negatively impact things like investment and overall demand. Verizon stressed repeatedly their “right” to convert broadband access from its existing “one sided” market (where it charges only the subscriber for the service) to a “two-sided market” (by which I take it Verizon means the right to charge both the broadband subscriber and any party sending information to the subscriber). Mind you, this is not exactly what “two-sided market” means in the technical economic sense. But let that pass for now.
Ignorance and Arrogance On Parade
Silberman and Tatel made it clear in the first few minutes that they totally agree with Verizon’s characterization of the no blocking rules. Silberman started by noting that it was Verizon’s “strongest argument.” There followed a biref interlude where Silberman and Tatel did their imitation of my cats grooming each other over how wonderful this “common carrier rule” they made up last year for the data roaming case is. (To paraphrase: Silberman: Everyone should look at the most excellent and wonderful and sagacious explanation of the common carrier rule written by Judge Tatel (lick lick prrrrrr). Tatel: Aww shucks, ’tweren’t nothin’ (lick lick prrr).)
Unfortunately, while Silberman and Tatel were the ones mutually grooming each other, I am the one getting the hairball sitting through an hour or so of Judges Silberman and Tatel demonstrating their poor grasp of both underlying technology and economic theory, coupled with a willful ignorance of how any segment of the broadband market actually works in reality. Critics like to sneer at the FCC for supposed failure to understand the nuances of network management or competition policy — but even these critics must admit that at least the FCC employs engineers and economists who do understand these things. Not so generalist judges, which is why judges are supposed to defer to the agency on these highly technical and fact dependent matters.
To take but one appalling example, at one point counsel for intervenors was attempting to illustrate something and said: “Assume Verizon wants to put [whatever] on their FIOS service.” Judge Tatel contemptuously responded: “FIOS? That’s cable! Why are you bringing up cable? This is a broadband case.” Of course, FIOS is actually the brand name of Verizon’s entire fiber-to-the-home service — most especially including the broadband portion. One would think that if you presume to have enough knowledge of the broadband market to casually dismiss the expert agency’s findings that you might actually know that.
In fairness to Tatel and Silberman, I should point out they and their colleagues on the D.C. Circuit are not alone in the delusion that judicial appointment and life tenure somehow magically confers an understanding of economics that makes their ignorant opinion a matter of law that the FCC must follow. A quick read through the cable program carriage opinion the Second Circuit released recently shows a similar cavalier disregard for both economic theory and the actual facts of the specific market at issue. But I digress.
So What Is “Core Common Carriage”?
Waiving aside any effort to actually understand the market or technology, the bulk of the FCC’s time focused on the back and forth between Silberman, Tatel, and FCC General Counsel Sean Lev. In particular, Tatel noted that he found the FCC’s data roaming rule was not “common carriage” because it gave Verizon Wireless flexibility to negotiate terms. How could one square a definitive “no blocking “rule and a definitive “no discrimination” rule with the idea of “flexibility?”
Of some relevance for possible outcomes, however, Rogers noted (and Lev echoed) that this was a prima facie challenge to the rules and that the FCC was using an adjudicative approach and that the rules and accompanying Order text had few absolute definitive rules other than no blocking. Even with regard to paid prioritization, the FCC did not say categorically “no,” but only that it had “grave concerns.” To win on a prima facie challenge, Verizon should need to show that there are no cases where the rule would be valid. If, for example, the FCC could order “no blocking” in the presence of market power (as even Silberman seemed to think), then the court should sustain the rule. Objections about specific applications of the rule, and whether the rule provides for something sufficiently “flexible,” should wait for an “as applied” challenge.
Mind you, this is exactly why supporters of network neutrality (including myself) hate about the FCC’s net neutrality rule. Even the core of net neutrality — no blocking and no prioritization — appears totally up for grabs. In defense of FCC Counsel, Lev was trying to save the rule. He caveated his words repeatedly with things like “I don’t want to mislead the court or appear to concede too much” and “we disagree, but accepting this as a hypothetical,” and he repeatedly attempted to return to his core arguments that this wasn’t the right way to look at the market and the FCC made an expert judgment entitled to deference.
The Difference Between “No Blocking” and “No Prioritization.”
As so often happens, the mutual grooming between Tatel and Silberman turned into a full blown catfight during the course of their playing with the poor FCC mousie. Pressed to distinguish between “no blocking” and “no discrimination,” Lev explained as follows. No blocking guarantees a basic level of service. If an end user subscriber requests some content or service from a third party, the broadband provider has to make at least some effort to comply — and not degrade the content or service to a level that effectively counts as blocking. On the other hand, a “no discrimination” rule means you can’t offer a fast lane/premium service (or presumably, other special treatment such as exemption from a data cap).
Tatel appeared to accept this distinction, and seemed willing to allow a ‘minimum level of service’ as long as Verizon could ‘customize’ it. Silberman rejected this. If someone refuses to pay for access, Silberman reasoned, then Verizon needs the freedom to block access to make the recalcitrant pay up.
To illustrate the difference. According to Tatel, it is fine to say that if an end user subscriber tries to use Bing as a search engine, Verizon would have to honor that request somehow. (Although Tatel did not mention this, it would be reassuring to believe that Verizon is actually bound by its promise to the end user to provide “Internet access” as commonly understood, and that the FCC can enforce this.) But under the “common carrier rule,” Verizon needs to be free to offer Google an exclusive on “fast lane delivery” or “more secure connection.”
Silberman, by contrast, thinks that Verizon needs to be free to allow Google to pay to block Bing, because otherwise how could it collect from Google for prioritization. Or, alternatively, if Google paid for uber fast speed, but then didn’t pay up, it wouldn’t be enough for Verizon to dump them back to “slow normal.” Unless Verizon could totally block Google, Google would have no incentive to actually pay Verizon for “fast lane” treatment.
Did I mention that I find Silberman’s grasp of economics and incentives a trifle . . . um . . . disassociated from reality? But hey, when you wear the black robe and have life tenure, you don’t need any understanding of how the real world works to make your alternate world view law.
Other Silberman Stuff.
As Stuart Benjamin noted on Volkh Conspiracy, Verizon’s First Amendment Argument turned out to be “the dog that didn’t bark.” Verizon argued that the net neutrality order violated their First Amendment right because they, as speakers, have a right to block other people’s speech (or prioritize it). My PK colleague John Bergmayer generally addresses that argument here. While no one really expected this to get full on “heightened scrutiny” as if Verizon’s broadband service were a newspaper, many people thought the court would bite on giving broadband access service “intermediate scrutiny.”
No one really bit on this except Silberman. Lev made short work of this by noting the relevant case law says that in commercial ‘forced speech’ cases, the relevant question is whether anyone reasonably believes the speech is attributable to you. For example, if Comcast carriers Al Jazeera America, people recognize that Comcast is not “the speaker,” but that Comcast does have a choice as to whether to carry the channel at all. Comcast therefore has an intermediate speech interest (according to Turner Broadcasting) in what channels it selects to carry. By contrast, no one who enters www.wetmachine.com thinks Verizon is the speaker for this blog, even if they are using Verizon’s FIOS (which is a broadband service) to download it.
Silberman asked didn’t Verizon have a First Amendment interest in blocking speech they found totally antithetical to their interests (like, say this blog *wave*)? Lev stood his ground with case citations, which left Silberman mumbling “well, maybe we won’t have to reach the First Amendment question.”
Also, as noted previously, Silberman wagged a finger at Lev for the FCC’s failure to make an actual finding of market power as opposed to simply noting that lots of people don’t have a choice of provider, or have only one alternative, and that switching is a major hassle (that’s transaction cost and switching costs for all you folks mourning Ronald Coase this month). Lev responded that the agency made findings and is entitled to deference. Lev tactfully did not mention that nothing in the statute says diddly squat about market power being a pre-requisite for finding anything — although I probably would have pointed that out.
This is already waaaay too long so I’ll save my impressions and possible outcomes for another time. For now, let me say that counsel for Intervenors (which includes my employer Public Knowledge), Pantelis Michalopoulos, did an excellent job attempting to clarify things for Tatel and Silberman. Unfortunately, their stubborn refusal to understand how any of the things they are talking about actually work in the real world proved unassailable.
Verizon came back for rebuttal, where Walker spent another half-hour trying to persuade the court that Section 706 could not possibly grant the FCC independent authority to do anything. But judicial opinion seemed pretty frozen on this point as well. After over two hours of oral argument, Judge Rogers dismissed the case.
It will be several months at least before the court decides. So we will no doubt have endless things to chew over until that happens.
Stay tuned . . . .