Monday, July 05, 2010

Regulating broadband: Lessons from history

A very interesting piece on Larry Downes' blog on Albert Gallatin and the first National Broadband Plan - the one he sent to Congress in 1808, on the need to build a national network of public roads and canals - "No other single operation, within the power of the government, can more effectually tend to strengthen and perpetuate that union, which secures external independence, domestic peace, and internal liberty.” Some interesting commentary on Techdirt.

Downes commends Gallatin's insight in recognising the importance of private-public partnerships:
“Private capital, motivated by the hopes of profit, provides the bulk of the investment and absorbs the bulk of the risks. Federal and often local governments provide support in the form of land grants, rights of way, investments in basic research for key technologies, encouraging standards, and financial markets that created liquidity for instruments necessary to finance construction...Financing is essential, for infrastructure is characterized by heavy initial investments relative to on-going operating costs once construction is complete. The risk of failure is very real. If you build it and they come, to paraphrase “Field of Dreams,” the network can provide solid profits for generations. If they don’t come, you lose everything. Timing also matters. Traffic may arrive later than you think, by which time you had sold out your investment at bargain-basement or bankruptcy rates, only to watch others ultimately profit from your vision...In some cases governments provide outright subsidies for populations or geographies for whom the cost of providing service clearly exceeds any hope of profit under a traditional capitalist model. Serving these communities is important not merely for egalitarian reasons, however, but because of the increasing returns to scale that network technologies offer. The more people who are connected to a network, the faster its value to all participants increases. Many of those who cannot afford access, according to network theory, will contribute value beyond the cost of providing it to them for free or at a price below cost.”
I agree with this in principle, but with an important caveat: national and local governments need to provide a counterbalance to private investment. They should act as intelligent commissioners to procure and manage an effective, universally available, standards-based broadband infrastructure, creating a true partnership with the private sector's capacity and ability to innovate. And standards should be mandated, not just encouraged. This goes beyond the role Downes outlines above, which to me perpetuates the mistaken belief that the market will deliver of itself. That's the most important lesson from recent broadband policy in my opinion. Interestingly, the point about the more people that are connected, the greater the value of the network reminded me of the positive network externalities described in the ITIF's 2009 Need for Speed report. 

Downes goes on to argue that the US "lags not in investment or availability but in adoption": 95% of Americans have access to broadband, but only two thirds have signed up for a broadband service, paralleling the situation in the UK. Rather than looking to further regulation, the FCC should focus on providing incentives to adopt broadband, according to Downes:
"Blaming the lower per capita adoption - and using it to justify regulatory intervention - on the supposed failure of ISP competition, the absence of enforceable net neutrality rules, and the lack of meaningful FCC supervision of ISP business practices under Title I of the Communications Act are all red herrings. They ignore the historical lessons of Gallatin’s plan and the infrastructures that have been successfully deployed since then. Worse, they are leading the FCC down a rabbit hole of unnecessary regulation that is likely to do more harm than good."
Regulation is not without merit, but must be of the right sort if we are to learn lessons from the past and avoid the law of unintended consequences:
"U.S. railroads sunk into a morass of regulation and proved unable to compete with new modes and new technologies. Today, the most frequent activity for federal courts reviewing railroad matters is to review applications to abandon routes, which requires government permission...There are important lessons in the story of earlier infrastructures that will determine whether or not the implementation of the NBP succeeds or fails. History also offers important warnings the FCC would do well to heed as it pursues a parallel efforts to appoint itself enforcer of the neutrality principles that have presided without significant government involvement since broadband was invented in the mid-1990’s."
This "morass" was caused by the regulatory body, the Interstate Commerce Commission or ICC, seeking and gaining greater powers of enforcement, with results that Downes describes as "catastrophic". He commends the FCC's National Broadband Plan for its vision and recognition of the fundamental importance of broadband, and also the assumption implicit in the plan that developments will primarily be funded by private investment. This is the key lesson from Gallatin and the past, according to Downes.

But he closes by describing the FCC's intentions to reclassify broadband as "regulatory schizophrenia" and "high-minded but wrong-headed efforts to protect consumers from a harm that doesn’t exist with little consideration of the potential cost of doing so":
"On the one hand, we have the vision and inclusiveness of the NBP. The plan sets important stretch goals for access providers and as well as application developers. It establishes important roles for the Commission in supporting industry efforts and stepping in where needed to encourage adoption by unwired communities. On the other hand, we have the heavy-handed efforts to push the net neutrality rulemaking along with last week’s announcement that to solve the minor problem of a complete lack of Congressional authorization, the agency will undertake the legally-dubious process of giving itself the power to oversee broadband under Title II common carrier regulations. Those rules increasingly make no sense for the technology they were written to deal with - wireline phone service or POTS. Indeed, the lesson of the years since the 1996 Communications Act should be that Title II destroys value and diminishes service for nearly everyone, much as the ICC’s “just and reasonable” ratemaking did to the railroad industry - once the pride of American technology."
Downes is rather dismissive of the FCC/Comcast issue:
"The handful of isolated incidents where network management activities have inconvenienced some broadband users (some if not many of whom, in the Comcast case, were engaged in illegally downloading copyright-protected content using the BitTorrent protocol) hardly rises to the level of discrimination that justified the Hepburn and Mann-Elkins Acts. A few self-appointed public advocates, with broader agendas for nationalizing the media and communications industries, do not a Progressive or Grainger movement make."
However I would argue that there is an important underlying issue not addressed here - the lack of transparency Comcast demonstrated in its chosen approach to traffic management. This is simply not an acceptable way for private investors charged with building an essential new infrastructure to behave. It's not that they shouldn't manage traffic - this is only sensible in a bandwidth constrained environment - but that any traffic management practices should be made clear to customers. I would argue that Comcast's footprint in US broadband provision meant their actions constituted rather more than a "handful of isolated incidents", and Downes also disregards the many entirely legitimate uses of BitTorrent that Comcast's blunt instrument approach disabled at the same time: for "managing", read "blocking" in this context (see this previous post).

I agree with this remark though: "the poor fit between fast-changing network infrastructures and the slow pace of regulatory process." If anything's likely to bring the law of unintended consequences into play, it's surely this. Returning to my earlier point, that we can't rely on the market to deliver of itself, do some additional answers lie in empowering the customer (where the customer is local and national government, procuring and developing broadband infrastructure for its citizens)? Empower the customer to commission exactly what they want from industry, rather than regulate industry to provide what you think people need?

In essence, grasping the nettle of broadband provision. A lower risk strategy than relying solely on regulation perhaps?

1 comment:

  1. Thanks for your thoughtful critique of my piece. You are quite right that the lack of transparency in the Comcast throttling was the most worrisome aspect of that example (I've written about that elsewhere).

    I'm all for a rule of transparent network management principles. I doubt most consumers would really care or be able to understand disclosure, but there are enough interested parties who would review and promote news of the practices that it would serve to reduce the most ham-fisted and anti-competitive behaviors. (The latter of which is already illegal under antitrust law.)

    It's also true that BitTorrent has non-illegal uses, and more of them now than at the time of the Comcast incident, during which nearly all BitTorrent uses were illegal according to academic studies I've written about elsewhere.

    Even so, Comcast's decidedly bad behavior in throttling and hiding the throttling did not and does not amount to the kind of breakdown that justifies the wholesale regulation of the broadband industry the FCC is now proposing. It's like starting your house on fire to light a cigarette--which you shouldn't be smoking in the first place.

    That would be the case even in a more mature industry, where the risk of unintended consequences was lower. The level of industry malfeasance leading to regulatory intervention has historically and sensibly been much higher (e.g., the railroad example I give in the piece) than the relatively minor breakdowns the FCC is using here.

    There is at least the appearance that the Comcast case and the loss in court are being used to bootstrap what could be very significant changes to a rapidly-changing market. The FCC's motives do not appear to be fully disclosed, to say the least.