Two Open Broadband Internet Proposals vs the Pay-per Internet Model

Two new research papers released in the past week add insight into the Usage-Based Billing (UBB) debate in Canada, or what I have been calling the evolution of the pay-per Internet model. The papers are by Michael Geist, the University of Ottawa law professor, and by Bill St. Arnaud, the Chief Research Officer for CANARIE for 15 years (until 2010) and a telecoms engineer. Geist’s paper can be found here, while St. Arnaud’s paper is here.

Both papers were commissioned by Netflix, in light of the fact that developments in Canada are sucking it and others such as Google, Apple, and so forth deeper and deeper into digital media policy issues. All are becoming fixtures in CRTC proceedings. Both papers bear one significant subtle influence of this sponsorship (as I will discuss briefly below), but other than that provide extremely valuable help wading through the technological, economic and regulatory issues surrounding the UBB debates.

Geist and St. Arnaud are both convinced that the CRTC’s plan to revisit it’s January 25th UBB decision that ignited the firestorm over the pay-per Internet model in Canada is far from sufficient. As Geist indicates, a whole series of decisions over the past few years will have to be revisited and the regulator and policy-makers are going to have to deal head-on with the fact that underlying these problems is a heavily concentrated market for Internet access in Canada. I feel similarly, and have laid out the ‘long march’ to the pay-per Internet model in an earlier post.

Playing on earlier decisions regarding the incumbent telecom and cable companies use of technical measures to ‘throttle’ different types of Internet uses that they argue put excessive strain on their networks — the so-called Internet Traffic Management Practices — Geist’s first proposal is for a series of what he brands IBUMPs (Internet Billing Usage Management Practices). The basic gist of which is to make the incumbents’ billing practices for Internet services easy to understand and reasonable when it comes to so-called excess usage charges.

His second set of proposals aim to promote greater competition in the Internet access market. This includes allowing more foreign competitors to enter Canada.

It also involves allowing smaller ISPS and Content Distribution Networks (see below) more scope to interconnect with the incumbents’ networks much deeper in the network and closer to subscribers’ homes (especially the cable companies, who have dragged their heels on this matter for more than a decade). Finally, it means cultivating a greater role for alternative Internet access providers, from city-owned networks, to cooperatively run ISPS, as well as expanding the role of provincial and federal broadband development programs.

As an interesting aside, the Liberal Party’s platform announced on April 2nd as part of the current federal election campaign effectively doubled the commitment that the Liberals would put into expanding broadband networks in remote and rural areas compared to the modest $225 million announced by the Conservatives in 2009. The Quebec Government went even further in the 2011-12 budget passed in March, where it announced that it will invest around $900 million in bringing very high speed Internet access to all Quebecois (see here at pages E.93-96; also see St. Arnaud).

It did not specify the exact capacities of the network, but its references to similar plans in Australia, France, Finland and the US suggest that the bar is high, probably around 100 MBps. Neither the Liberal Party’s election platform nor even the much more ambitious Quebec Government’s scheme are equivalent to or the same as Australia’s National Broadband Company initiative, and nor should they be.

However, they do underscore (1) the under-development of broadband Internet in Canada, (2) the lack of competition offered by the current market, and (c) a willingness to rely on a variety of providers, from the traditional incumbents, to municipalities and provincial governments to improve on the situation at hand. They also suggest that Geist’s proposals, far from pie-in-the-sky, are grounded and with some real, even if tentative support in some crucial quarters.

Bill St. Arnaud’s paper also offers much food for thought and complements Geist’s paper very well. He makes three key points.

First, the massive growth of video online is not necessarily causing congestion. Huh?  How could this be, with clear evidence that the growth of video traffic has been stupendous, ranging from 50 to 100 percent per year and with continued high rates of growth expected in the next few years ahead?

This is because sources responsible for this massive increase are increasingly turning to Content Distribution Networks that, basically, bypass the public Internet and deliver their content as close to their subscribers as possible. These so-called Content Distribution Networks are not only being deployed by outfits such as Netflix, but other large Internet content and service providers, from Amazon, to Google and Facebook. The basic point is that they take traffic off the network for much of the distance a message has to travel.

Second, to the extent that congestion is a problem, this is an outcome of decisions made by the incumbent telecom and cable companies about how to apportion the capacity of their network. As Geist quips in his paper, the ‘chicken roasting channel’ recently introduced by Rogers, for instance, is just so much bandwidth allocated to that ‘service’ rather than to the Internet.

Third, and this is where I think St. Arnaud has an amazingly powerful and clear point, the incumbent telecom and cable companies — the ‘big six’, as I have called them: Bell, Rogers, Shaw, Telus, Videotron and Cogeco — appear to have no problem with congestion when it comes to launching their own video content services delivered over the internet, e.g. CTV.ca, globaltv.ca, TVA.ca, etc. Congestion is only caused by other providers’ video services.

Lurking in the background of all this is that we’ve seen this all before. A few years ago, P2P/file-sharing and music downloading sites were the culprit; now the target is online video services. The cable companies have been especially remiss in dragging their feet for a dozen years or more on allowing independent ISPs to access their distribution infrastructure. Despite being required to do so before the turn-of-the-century, the cable cos have thrown one obstacle after another in the path of ISPs requiring last mile access through cable facilities to gain access to subscribers.

All said and done, Geist and St. Arnaud’s paper respectively do a great service. They are timely interventions that help us understand the issues at hand and, if successful, they may help to frame the debates that take place at the hearings that the CRTC has scheduled to revisit the UBB decision in June.

However, we should not hold our breadth on that, and in that regard these papers do a real good job at holding the regulator’s feet to the fire (see my earlier post on this point). The CRTC has a very broad remit to regulate in these matters, as the Telecommunications Act (1993) (sec. 27 (5)) makes clear, but has chosen to draw the proverbial camel through the eye of the needle. With the magnitude of the issues at stake, this is unacceptable.

However, I also think that both papers need to go even further in at least four ways. First, both papers make claims about the highly concentrated state of the telecom, cable and Internet access markets in Canada, but offer little to no data to illustrate and support these claims. Good quality data is now available on these points and they should use it.

Second, both papers focus on the UBB issue, or what in regulatory parlance is now called an economic measure for managing congestion on the Internet. However, the CRTC’s Internet Traffic Management Practices decision (2008) sets out a hierarchy of preferences for dealing with such problems when they can be shown to exist: (1) network investment, (2) economic measures such as UBB, and (3) technological measures, aka throttling.

Neither paper says much, if anything, about the top priority: network investment. Why? At between 15-18 percent of revenues, current levels of investment in their networks by the big six is low by historical and global comparative standards (although in line with similarly low levels in the U.S.). And this despite the fact that the Internet represents a massive new source of revenue ($6.5 billion in 2010).

Third, neither paper pushes as hard as they might on how the use of UBB and the allocation of network capabilities by the incumbents to their own services may constitute a form of “unjust discrimination”. The issue is not totally ignored by any means, but I think it could be pushed further and that doing so is important not just to the question of whether or not we’re going to be stuck with a highly concentrated Internet market and the pay-per Internet model in Canada, but concentration across the whole sweep of the network media ecology, from traditional media to the Internet.  Let me explain. I’ll conclude by returning to my fourth point.

Insofar that these papers deal with ‘unjust discrimination’ they seem to have in mind section 27 of the Telecommunications Act that specifically outlaws such practices. It is a good victory to be had, if it can be had. And the CRTC has, as I stated above, much discretion in how it goes about making such determinations. To the extent that it has chosen to blinker itself is a problem of the first order.

However, it may be possible to go even further and look to the next clause of the Telecommunications Act, section 28, that specifically makes the issue of discriminating between video services, or broadcasting as such things were known when the act was written nearly 20 years ago, a matter of potential concern. Indeed, the CRTC has enormous authority under this section to deal with the issue of discrimination while meeting other objectives of the Broadcasting Act.

Herein, however, may lay the rub, given that both of the papers being discussed here were funded by Netflix, and the last thing that it and other services like it (read: Google, Apple, etc.) want is to be defined as broadcasting services, which could happen if we were to assign the ‘online video distributor’ label on them like the FCC and Dept of Justice did recently in the US in relation to the Comcast/NBC merger.

I, too, am very leery about slapping the label of broadcaster on such entities because of all that would mean with respect to CanCon rules and the like. The CRTC has always indicated that it believes that it has the authority to regulate online video distributors under the Broadcasting Act (see its seminal 1999 new media decision here), but up to now has not seen Internet television services as being significant enough and too experimental to actually do so.

The question of whether ISPs could also be brought under the purview of broadcasting regulations so that, just like cable and Direct-to-Home satellite providers, they too could be required to contribute to funding and displaying CanCon has also been hotly contested. That route seemed to be foreclosed by a Federal Court of Appeal decision in 2010, but that too has now been appealed to the Supreme Court of Canada.

Now, the incumbents en masse are pushing hard to have OVDs like Netflix, Apple and Google regulated as broadcasters just like their own broadcasting-related services. The irony here is that for Netflix to push its case on UBB as hard as possible, adding some water to its wine by accepting some such designation could go a very long way to putting a stop to the discriminatory practices that are now hobbling its access to Canadian subscribers.

While this is far beyond the scope of what I can say here, perhaps a new designation along the OVD line devised in the US might be imported into Canada for just such purposes. That would mean distinct treatment from broadcast television in general, but also some obligations to open up their services to Canadian media creators.

It might also allow a much more forceful push against the anti-discrimination rules of not just one section of the Telecommunications Act, but both sections 27 and 28. Done right, this need not ‘trap’ new players like Netflix in the maw of outmoded aspects of the Broadcasting Act. Instead, it could potentially help to usher in an entirely new media model where all of the bits and pieces that make up the traditional media model are disassembled and reassembled anew in light of the realities of the digital network media industries in the 21st century.

And finally to return to my fourth critique of the Geist and St. Arnaud papers. Both papers target the upcoming UBB decision. This is great, but I think it might be helpful to try and kill two birds with one stone by putting another potentially even more important upcoming regulatory review in their sights: namely, the CRTC’s hearings scheduled for June 2011 on vertical integration.

The ‘vertical integration’ hearings were scheduled late last year but given added impetus when the CRTC approved Bell’s acquisition of CTV last month. The idea of holding such hearings reflects the fact that Canada now also has the dubious honour of standing alone in the extent to which fully-integrated media conglomerates have become the norm. In the U.S., the fully integrated media conglomerate has become the exception (e.g. Comcast/NBC-Universal) after the disastrous AOL Time Warner merger and is pretty much in retreat in almost every other developed capitalist democracy.

There is indeed every reason to be very skeptical about these hearings given that they are a classic case of “bolting the barn door after the horse has already left the stable”. However, given that the use of UBB is completely tangled up with the crucial question of whether or not the “big six” media conglomerates in Canada — Bell, Shaw, Videotron, Rogers, Shaw, Telus (the latter to a lesser extent) — are using the pay-per Internet model to disadvantage competitors and to protect their own traditional television services, as well as their recently-minted internet video services, we must keep our eyes on the full range of big issues before us.