Start Again - Communications Reform Has Inauspicious Start in House

The popular history of 1986 tax reform has it that Congress was headed in its usual direction: a monstrosity of special interest provisions, impenetrable complexity and ersatz social engineering through the tax code. Indeed, that is what public choice theory predicts will happen with legislation--well-organized constituencies will grab spoils from the less-well-organized ones, and the public-at-large's interest will get lost in the deal making.

With 1986 tax reform, something different happened though. All of the complexity got to be too much, too intricate, too impenetrable. So they started over.

In the end, a bipartisan group of legislators enacted sweeping, simple and equitable tax policy that eliminated the special interest rentseeking and tax professionals' love of complexity. It was a signal legislative achievement. Let us hope that the House Commerce Committee staff can have the simplicity and equity of '86 tax reform in mind as it considers communications law reform, because it is off to an inauspicious start.

Offering a staff discussion draft replete with complexity, that picks regulatory favorites based on technology, that misses the central role competition policy should play going-forward, and that extends forward social obligations with an anticipatory regulation model from the progressive-era into the Internet space, the staff draft is such a well-intentioned, but misguided, attempt to solve immediate problems in communications. But the price for solving these problems is extending forward the technology-favoritism and monopoly-based "public interest" obligations of the current system. It would be better to start over with the basic goals of simplicity, equity and consumer welfare in mind.

If the staff draft can start over with these values in mind, we could end up with a concise, general, non-special interest riddled framework that benefits U.S. competitiveness in the world and consumers in particular. The discussion draft will then be remembered as having served the useful purpose of launching a productive dialogue.

The staff draft begins with a laudable end: to create new categories of service free from the legacy regulatory past. The draft eliminates rate, term and service regulation at both the federal and state level for these new service categories. The draft also gives a definitive answer to the threshold existential question for a regulatory framework: Should it be based on technology or competition policy?

The discussion draft, like the '96 and '34 communications acts before it, answers that the regulatory structure should be set up along technological definitions. Like previous communications legislation, the discussion draft then rests on a conceit that Congress can accurately define and adequately predict what technology is and where it is going. Thus, legacy categories of "telecommunications service," "cable service," and the like, are succeeded by new, Internet age technology categories like "broadband Internet transmission service" (BITs), "broadband video service," and "VoIP service."

Today, these categories look accurate in predicting the migration of communications onto broadband, packet-based, Internet-like systems. But predicting technology and its paths is an uncertain business where Congress already has a spotty record. For instance, the '96 Act largely missed the Internet, just as it was getting ready to take-off. Unlike the '96 Act, the discussion draft tries to get ahead in the technology predicting game. But it still is a limited view into the future. As the DACA Regulatory Framework Working Group Report put it:

Thus the first risk the discussion draft takes is by choosing a technology regulation path. While this may be reasonably workable in a static industry like phones in through most of the 20th Century or the electric system, it is quite perilous to make regulatory distinctions based on where you think the technology is going.

What's more, the techno-functional regulatory distinctions of the staff draft become self-fulfilling industrial policy decisions on the direction of technology and the industry structure of the communications industry. Providers will make technology decisions based on the ability to opt-into or evade a given regulatory category. This is not market-driven technology development; but regulation-driven technology decision-making. Finally, as my colleague Randy May has written, drafting statutes based on techo-functional distinctions leads to an enormous amounts of time and energy being directed toward making "metaphysical" distinctions to determine a given-service's regulatory status. While this provides good work for lawyers, it seems nonsensical to technologists.

In contrast to the staff draft's technology-centric framework, the DACA regulatory framework working group counsels the opposite path: competition policy, not technology distinctions, should guide regulation. [2] While there seems little political interest in extending competition policy -- a field of law that successfully governs nearly all aspects of the economy - to communications, it is difficult to understand why. While a technology-centric model depends on correct definitions and discernment of the technology and functions of a given service; competition policy is general and adaptable across all technology platforms. More important, competition policy gives substantive normative guidance and demands legal rigor before regulating a given service. By contrast, the technology-centric model gives no insight into if or how you should regulate these newly created Internet service categories.

The DACA Regulatory Framework working group explained the dilemma from the IP Migration [3] model as follows: "nothing in the IP-Migration model describes the level of regulation to which IP-based networks and services should be subjected." And here is where the problems with the discussion draft really begin. For, while the draft clarifies the regulatory status of currently evolving technology platforms, it also answers the question how they will be regulated. The answer given in the discussion draft is unfortunate. Despite significant and salutary reductions in rate regulation, in too many ways, the staff draft carries forward the legacy problems of current regulation and adds some new ones.

Interconnection, 'Net Neutrality,' Broadcast Regulation and Extending Regulation to the Internet

The real problems with the discussion draft come from answering the question of how these new IP-migrated technology service categories will be regulated. The new categories of "BITs" "BVS," "BVSP" and "VoIP" all begin promisingly with a recitation of "thou shalt not regulate..." This includes rates, terms and condition, and unifies all regulation under a federal framework. But the general admonition not to regulate a given-service class declines quickly with the inevitable "except as provided in this act" clause. And this staff draft provides too much.

To begin with, for all these new Internet technology categories, the staff draft imposes certain prophylactic regulatory requirements: registration, interconnection, net neutrality, public safety. Requirements like registration seem benign, but even this statutory grant remains broad enough so that real entry barriers could be raised to new providers.

The other regulatory burdens on the new service categories are carried forward as an ersatz mix from Title II common carriage and Title VI cable regulation. In other cases, the draft would impose regulation where none had existed before. For instance, Internet interconnection, which is now done successfully through private contracting, would be subject to regulation and state or FCC mediation and arbitration. And, despite claims that rate regulation is obviated, interconnection arbitration involves the FCC or the state in little else. By mandating interconnection and arbitration, the FCC (or state) becomes the decisionmaker about how carriers compensate one another and at what rate.

For video services, in exchange for laudable local franchise relief, the draft carries forward the host of Title VI cable social obligation regulations, and it also threatens build out and anti-redlining regulation. The tax purpose of the franchise is kept intact, even though franchising lost its purpose as anything other than a taxing authority under the Cable Act of 1992. [4]

While the statute attempts to temporize and moderate the effect of some of these mandates so they do not become overly burdensome, the core presumption of the statute is that certain types of regulation are the norm, and providers have to "litigate out" from under regulation to attain market freedom. So, with net neutrality, a provider can under the statute reasonably make (presumably price) distinctions based on bandwidth usage for premium services. Nevertheless, the default rule is the Procrustean end-to-end requirement, where departure from it requires proof from the regulated party that it deserves market freedom.

For VoIP, more regulation is probably inevitable, but prophylactic mandates threaten to keep these new services bottled up with big providers with substantial regulatory wherewithal. Not only is universal service taxation extended to this application, it also comes under a fairly stringent public safety E911 requirement and a two-way number portability requirement. Nevermind how this gives incentives for different session initiation protocols (SIPs) to avoid the numbering system altogether, but all of these mandates raise barrier in the VoIP market and risk shutting out the market to smaller, less regulatorily sophisticated new entrants. VoIP must also enter into regulator-supervised agreements to exchange and pay for traffic.

The fundamental problem with carrying forward these old regulations and applying these mandates to the unregulated Internet world is that the necessary prior question is never asked. What are these mandates there for? What is the consumer harm without the mandate and why does not a competitive market yield a superior result?

Is interconnection mandated because there is a nontrivial chance of consumer harm from non-interconnected networks--a plausible argument--or because we have always regulated interconnection fees and it is there that regulators' can accomplish their cross-subsidies and other meddling? Is net neutrality there because markets will surely lead to closed networks that harm consumers, or because of a rentseeking play by content and application providers to cut out network owners from earning a negotiated producer surplus? What are the potential costs of such an across the board mandate, particularly if the facts on which a mandate is justified are wrong or changing? Before imposing these sweeping regulatory mandates into the Internet space, Congress needs to be sure of its facts, aware of the error costs and mindful of the special interest pleading that hides behind many a benign mandate.

Of course, there is the competition policy path that contemplates all these possible mandates and evaluates them on a case-by-case basis based on proven facts. Not only would this lower the error costs and entry barriers from regulation, it would be adaptable to situations where unforeseen consumer harm arises.

Legislation can only accomplish so much, and in many ways is an extremely crude tool. One axiom that can be roughly correlated to the complexity and intricacy of legislation, however, is the opportunity for the law of unintended consequences to do its damage.

I am uncertain where that law would kick in with this framework, but I have no doubt it would. Will the plenary network reliability power be used to bootstrap regulations of all sort? Perhaps not, but see what FERC seems poised to do right now with a similar provision relating to electricity and realize that all sorts of regulation can be justified in the name of "reliability." In the '96 Act, reciprocal compensation seemed to be an innocuous provision. But that provision ended up resulting in hundreds of millions of dollars of arbitrage-driven transfer payments from local exchange carriers to terminators of Internet-bound dial-up traffic.

The law of unintended consequences thrives on the complexity, the differential regulatory treatment and the unknown arbitrage opportunities that a labyrinthine statute like this offers. What's more, a statute that starts out this complex will only get more complex in the legislative process. Given "no regulation except as stated here" the structure of the statute, the political economy challenge of removing "excepts" from your technology category and adding them to you rivals' is too good to pass up. In the end, the "excepts" will add up - perhaps, if we are lucky like in the '96 Act different "except" clauses will contradict each other - and the law of unintended consequences will work its magic.

As it stands, the staff draft is something that only Washington communications lawyers will love. [5] The framework and technology-predictions implicit in the bill virtually guarantee revisiting this Act ten years down the road, just as we are now doing with the Telecommunications Act of 1996. The staff draft purports to solve the immediate issues in broadband: franchising, preemption of state regulation, and a clear framework. It does that, to a degree, but at a high cost of extending the regulatory presumptions of the progressive age into the Internet age.

Tax Reform in 1986 happened because of the courage and leadership of a few officials who saw an opportunity, for once, to avoid "business as usual." [6] The imperative for the U.S.'s broadband leadership should be another opportunity to set aside business as usual, and grasp for a great legacy of communications reform. The reform need not be 'radical' in the sense of eliminating all specialized regulation of communications, but it can begin to treat communications like the rest of our economy governed in the first instance by property and private contract rights, and regulated only where market power abuse and public safety are concerned. With that, a draft should:

Start with competition policy, not technology prediction, at the core of the statute. Competition policy encompasses issues like 'net neutrality' and interconnection, but does not regulate them prophylactically in advance without firm evidence of the costs and benefits of such regulation.
Set an arbitrary (and seemingly low) page limit for the legislation and stick to it, say 20 pages. Complexity is loved only by lawyers and regulated companies; it is new companies and consumers' enemy. Furthermore, a concise, general statute cannot be so easily corrupted in the legislative process, and the law of unintended consequences has fewer nooks and crannies from which to later spring and do its damage.
There is no "third way" between competition policy regulation and social regulation. The former is based on legally rigorous, yet broadly adaptable, standards; the latter allows no predictability and invited special interest seeking for regulatory favor.
Migrate from "anticipatory regulation" as the FCC now does to "rule of law" regulation as is now practiced at the FTC, FERC and the SEC. No administrative agency is immune from political and legislative manipulation, but the FCC's current anticipatory regulation model invites gobs of special interest lobbying.
Technology will always seek to escape the techo-functional regulatory categories you try to create for it. Call it "BITs," "BVS," or VoIP, if you want, but technology will follow paths to avoid regulation and its attendant taxes. The industrial policy predictions implicit in creating these techo-functional categories virtually ensure having to revisit the framework frequently.

The staff discussion draft does the service of continuing the conversation about communications reform. That said, the model presented there is so cumbersome, so intricate and so inviting of more prosaic expansion by special interests that Congress should just go back and Start Again.

* Raymond L. Gifford is President and Senior Fellow at The Progress & Freedom Foundation. The views expressed here are his own and do not necessarily reflect those of The Progress & Freedom Foundation, its officers or Board Members.

See, Dixon, Kyle D., James L. Gattuso, Raymond L. Gifford, Randolph J. May, Howard A. Shelanski, Douglas C. Sicker, James B. Speta, and Dennis Weisman. Digital Age Communications Act - Proposal of the Regulatory Framework Working Group, Release 1.0. The Progress & Freedom Foundation, June 2005, available online here.
Id.
The "IP Migration" model discussed in the DACA draft is a precursor the the technology-centric statutory categories offered in the staff draft. Like the staff draft, the idea behind "IP Migration," roughly put, is that once you are on an IP platform, you are freed from regulation. As the staff draft makes clear, however, once you are on an IP-based platform, the questions of what regulations apply are wholly a matter of political will and compromise - and are in no way constrained by a competition policy standard.
For more detail here, see Adam Thierer's forthcoming PFF PS on this topic.
I should add--as an impartial statement against interest--that think tanks studying communications policy also stand to thrive under a staff draft bill. The rulemakings to dissect, the litigation to comment on, the instant suggestions that the Act needs to be overahauled again are pure gold in the ideas industry.
In the regulatory context, similar leadership was displayed in the 1970's when notable Democratic politicians like Ted Kennedy, with able staff work from then-counsel Stephen Breyer, took a hard look at airline regulation. They concluded - and their conclusions have been borne out - that airline regulation disserved the public. Similar insights were then applied to natural gas and interstate trucking building on the scholarshop of both conservative and liberal scholars. In each case, the broad, largely categorical deregulation of the industries lead to enormous increases in consumer welfare.