TMCnews Featured Article

The Court of Appeals for the DC Circuit recently sent the FCC a big lump of coal. Comcast (News - Alert) won its appeal of an FCC Order that addressed its network management practices.  A few years ago, Comcast was found to be monkey-wrenching some BitTorrent (News - Alert) uploads on its cable modem service.  Then-FCC Chairman Kevin Martin had a cow and suddenly got the "network neutrality" religion.  Citing its "ancillary" jurisdiction over communications, and its policy statements, though not actual rules or enabling laws, the Commission ordered Comcast to change policies.

The Court's order comes as no surprise to observers who have been following the case.  It was not decided on the direct merits of whether or not the FCC had any power to ensure some kind of neutral access to Internet content and applications.  It was decided on a narrower, but critical, technical ground.  Ancillary jurisdiction has to be ancillary to something else, some actual, real, legal power that the FCC has under the law.  Perhaps such a law exists; perhaps it doesn't, but the FCC didn't cite one, so it lost the case.  The Comcast order thus lacked sufficient connection to any such law. While the FCC has jurisdictional authority over "communications", it still can only act within the scope of the law.

In other words, the Court was upholding a general principle called "rule of law".  This has been in short supply recently, so the decision is a welcome one, even if one disagrees with the result.

Giving the net a bypass

The FCC's principles for regulating telecommunications evolved out of a strange history.  There was a time, long ago, when almost any communications beyond a company's own property was the sole business of The Phone Company, a regulated monopoly.  This was justified on all sorts of specious grounds, but by the 1970s a little bit of competition was allowed, and the 1984 breakup of AT&T (News - Alert) was designed to make long distance fully competitive while local service remained a monopoly.

But in 1985, some of the newly-formed Bell companies proposed rates for their new "high capacity" (as "T1" was also known) services that were much higher than the FCC's rules should have allowed.  The FCC decided to accept the high rates, but in exchange, it authorized companies to compete with them for local leased-line (tariffed under the name Special Access) services.  These newcomers were called Competitive Access Providers (CAPs), and they later became the first CLECs.  They were subject to lighter regulation than the incumbent carriers that they were competing with, since they didn't have the same market power.

The CAPs played an important role in making high-speed Internet access available to businesses during its initial growth spurt of the mid-1990s.  Contemporary FCC policies also allowed other ISPs, who weren't telephone companies, to provision their own networks without being subjected to common carrier regulation.  This included the cable companies, whose cable modems were unregulated, and wireless ISPs, who took advantage of unlicensed radio spectrum.  Cable was the big kahuna for high-speed consumer access, though.  The term "broadband" had essentially referred to cable technology before it was co-opted to refer to the high-speed Internet access which cable had pioneered.

The incumbent telephone companies, with benefit of their monopoly-built networks, poles, conduits, subsidies, and customer base, were still subject to common carrier regulation, and were the "carriers of last resort".  When, in the late 1990s, they finally decided that if they ignored the Internet it would not go away, they decided to fight back with their most powerful weapons, lawyers and lobbyists.

We used to take competition for granted

Before the phrase "network neutrality" was coined in 2005, there was simply no real issue over how Internet Service Providers should be regulated as to what kind of services they had to carry.  They weren't, even if they were self-provisioned ("facilities-based"). You picked your ISP and got the service they delivered.

Telephone companies were regulated, though, when they provided access to ISPs, including their own.  The key rule - indeed, the basis of public access to the Internet in general - was a 1980 decision by the FCC known as Computer II.  Among its holdings was a distinction between "basic" and "enhanced" services.  Basic services were what telephone companies traditionally did, usually as common carriage.  Enhanced services were done within the content of the basic services.  Things like computer time-sharing and voice response systems were enhanced.  Under Computer II, the Bell telephone companies could only provide enhanced services on a "fully separate subsidiary" basis.  A few years later, Computer III softened that rule, and allowed them to offer enhanced services if they maintained a separate set of books, and if any other enhanced service provider could get the same basic services used by the affiliate's enhanced services, at the same wholesale price.

The Telecommunications Act of 1996 introduced new terms, "telecommunications service" and "information service", which almost corresponded to the earlier basic and enhanced services.  The Computer III rules remained in effect; TA96 essentially built upon them.  But in 1996, the Bells had almost no DSL in service.  The telephone companies could own ISPs, but any other ISP could offer service over the same wires.  So ISPs could always compete for DSL customers.  As the phone companies built their DSL networks, they still had to make wholesale raw DSL (used by ISP services) available to third-party ISPs.  Not that it cost them much market share:  Independent ISPs were more of a theoretical threat than a real one, but the rule meant that users did have access to a broader choice of DSL ISPs than one.  Consider it a brake on their market power.

The FCC broke it; now it has to figure out how to fix it

But the Bells whined loudly that it was unfair that they were not subject to the same rules as the cable companies, who, thanks to their head start and major upgrades to their networks, actually had a larger market share of "broadband" ISP customers.  In 2002, the FCC affirmed that cable modems were not subject to the same rules as DSL.  This was not a change in policy, though it has often been represented as one.  The FCC also exempted new fiber-to-the-home networks from most regulation, even the Bells'.  So only DSL was still readily available to ISPs.

In 2005, the FCC took that away.  They essentially repealed Computer II, and allowed the ILECs to treat the underlying DSL service as part of their unregulated Internet service.  If other ISPs wanted to use the wire, they'd need a commercial contract, which the telephone company was under no requirement to offer.  In practice, many existing DSL-leasing ISPs did receive three-year contracts, mostly for 2001-speed DSL.  But there is no guarantee for the future, or that any new ISPs could come along and join in.  So now independent, non-facilities-based ISPs were left with approximately zero options.  Hence the choice of ISP was limited to the number of wires going to the home, typically but not always two.

So is it any surprise that almost as soon as the new rules had been announced, there rang out a clamor to regulate the Internet itself, under the feel-good name "network neutrality"?  The FCC went along with the charade, first by asking a small ILEC, Madison River, to stop blocking VoIP, and then the later Comcast Order (the one that was just overturned).

Consider these policies to be the result of the FCC's War on ISPs.  This was really the Commission's key telecom policy initiative of the last decade.  It was a three-pronged attack.  First, when dial-up was still dominant, it attacked the CLECs who supported it.  The key attack there was to take away the reciprocal compensation revenue that supported the CLECs and held down dial-up rates.  The second prong was on the ability of CLECs to offer DSL to ISPs.  The key rule change was to take away line-sharing, by which the CLEC could ride atop the ILECs' voice lines for the same price that the ILEC-captive ISPs paid (usually nothing).  Along with that were new rules to make it almost impossible for CLECs to provide DSL to subscribers who were not near the ILEC central office - closing in on half, by now. Prong three was taking away ISPs' access to the ILECs' own DSL services, without CLEC involvement.  That final attack trapped the ISPs, who have been dropping like flies.

The DC Circuit drops some heavy clues

In its recent Comcast ruling, the DC Circuit basis its decision on the scope of "ancillary" jurisdiction, the FCC's justification for being able to regulate ISPs.  Judge Tatel offers page after page of examples of cases that raised the ancillary flag.  Those that were upheld were, he held, more tightly coupled to specific laws than those that were overturned, and the Comcast Order lacked that coupling.  Why did this happen?  The FCC's Comcast Order went out of its way to avoid raising Title II (common carrier) issues.  It was very intentionally based solely on vague ancillary powers.  This was the Commission's way of dancing around its withdrawal of the Title II common carrier protections of Computer II in the first place. This approach turns out to be legally untenable.

And Judge Tatel gives some really obvious clues here about what he thinks the Commission actually can do.  The ruling spends about two pages citing one particular case where ancillary jurisdiction was approved.  Which case?   The same Circuit's 1982 ruling in the CCIA (News - Alert) case, in which Computer II itself had been challenged.  Tatel cites these gems: "with respect to the AT&T component of the order…  '[r]egulation of enhanced services was . . . necessary to prevent AT&T from burdening its basic transmission service customers with part of the cost of providing competitive enhanced services.'" And then, "'Given [the] potentially symbiotic relationship between competitive and monopoly services,' we concluded that 'the agency charged with ensuring that monopoly rates are just and reasonable can legitimately exercise jurisdiction over the provision of competitive services.'"  In other words, the local incumbent phone company's behavior even in competitive markets could be subject to stricter regulation than a small competitor's.

This call-out to the impact of enhanced services on monopoly rates is timely today, as it was thirty years ago.  Telco-captive Internet services themselves have never been particularly profitable, if they break even at all.  Behavior in the "enhanced" or "information" business does seem to impact basic service rates.  Verizon, in particular, has been raising rates on its copper-based dial tone services to finance its heretofore-unregulated and costly FiOS (News - Alert) build out.

Remember, this all began when the FCC overturned Computer II, claiming that it was obsolete in light of the Supreme Court's Brand X ruling.  But wait… what did Brand X really say?  Brand X allowed, but did not require, the Commission to continue to treat cable Internet services and the underlying transport as a single, integrated service, unlike the way it had been treating DSL.  Tatel noted, "In particular, the Court suggested that the Commission could likely 'require cable companies to allow independent ISPs access to their facilities' pursuant to its ancillary authority, rather than using Title II as Brand X urged."  So in one sentence, Tatel notes that Title II does not apply to cable companies, but that the FCC could have chosen anyway to apply open-access requirements to them.  (It didn't.)  The FCC's claim that Brand X required cable and DSL to be treated the same way was, simply, a lie, and Tatel didn't buy it.

So this DC Circuit ruling essentially reminds the FCC that the Computer II framework of regulating ILECs (the successors to the original AT&T) to ensure access to their basic services is a valid option, while treating cable, and by inference other non-ILEC ISPs, differently is also a valid option.

Hence the FCC has a clear path forward, even if it turns out to look a bit like the Firesign Theater's old line, "forward, into the past".  The FCC can simply require ILECs to offer DSL to ISPs again.  It can probably expand that mandate to cover FTTH, such as FiOS and U-Verse, to the extent that there is a basic service offered by a carrier with monopoly power.  It could even choose to impose open-access requirements on some cable providers too, within the scope of the Brand X ruling.  Other non-dominant facilities-based ISPs would be harder to regulate at all.

One can argue that Computer II is obsolete, as it drew a line based on the technology of its day. .  If Computer II is obsolete, though, we may actually be moving back towards the 1970 Computer I ruling.  That was an early attempt to deal with the "convergence" of computers and the telephone network.  It created three categories.  The pure communications and pure data processing categories applied when they were obvious; the latter was unregulated.  A large telephone company who offered both had to do so via maximum separation.  Hybrid service was the big grey area in between.  The proper regulatory treatment of such offerings would be decided case-by-case until a new, less cumbersome, rule could be arrived at.  This took them a decade... The FCC today is no faster than it was then.

But regulation of ISP content and practices, which makes "network neutrality" different from "open" network requirements such as those of Computer II, will be harder to justify: "[T]he Commission cannot justify regulating the network management practices of cable Internet providers simply by citing Brand X's recognition that it may have ancillary authority to require such providers to unbundle the components of their services."

This ruling is not a bad thing at all.  It is an excellent result, actually.  It suggests to the Commission that it is still the "basic services", the "telecommunications services", that need to be regulated, and that the result of such regulation should be more competition amongst ISPs.  Only the Internet Access portion, the last mile bottleneck going to the ISP, belongs in play now, not the vertical (retail and hosting-service) portion of the ISP business, or the backbone ISPs.  It is that access to competition which made the Internet work in the first place, not litigation at the FCC over peering practices, spam blocking, and other network management issues.  Functional neutrality comes from open competition amongst ISPs, and this ruling supports that.  Thus by rejecting the FCC's hurried and ill-conceived first attempt at forcing "network neutrality" and encouraging an open access network instead, the DC Circuit's ruling may turn out to be the most pro-neutrality ruling possible.