CRTC’s UBB decision: pretty good economics, but a bad explanation

In its usage-based billing (UBB) decision (telecom decision 2011-44) with respect to the pricing of Internet access to independent Internet service providers (ISPs) announced on Jan. 25, 2011, the CRTC has seriously blotted its copybook—but not for the reasons implied by Industry Minister Tony Clement a few days later. The CRTC did a pretty good job on the basic economics of the issue—although we need far more detail to be sure. The regulator messed up because it provided a very poor exposition of issues it should have known would be controversial. Economics of UBB The economics of the pricing of Internet access go in the right direction. In the economy as a whole, UBB is the rule rather than the exception. However, the decision (even augmented by the chair of the CRTC's speech to the House industry committee on Feb. 3, 2011) raises more questions than it answers. UBB pricing makes sense—it improves allocative efficiency—in almost any situation. The exception is in the short run when an existing network has excess capacity at all times of the day. Then, "all-you-can-use," for a set price, makes sense. However, in an industry with very rapid rates of growth in demand (like in Internet access), such conditions are not likely to prevail for long. The fundamental point is that the owners of a network must—from all of its users—cover long-run average total costs, which includes the capital costs of increased capacity. Congestion? The decision did not address the congestion issue. Yet it appears to be the fundamental justification for permitting (but not mandating) the owners of networks (the telecoms and cablecos) to lease space to independent ISPs based on UBB. Chairman Konrad von Finckenstein, in his speech to the House industry committee on Feb. 3, said, "This demand causes congestion on networks, which can push the available bandwidth to its limit." He also said that the CRTC had "developed a comprehensive regulatory approach for Internet traffic management ... in 2009." Unbelievably, neither the decision or the chair's speech provided any data on the alleged problem of congestion. Further, no data were provided on the growth in demand and the growth in capacity. Hugh Thompson—a well-informed commentator—argued in a Feb. 10 piece in The Globe and Mail that, "In simple terms, the bandwidth explosion is real, but it's been more than offset by more powerful and more energy-efficient machines. So, we can reject the notion that increased usage is the a significant rationale for huge Internet price increases and usage-based billing.” The decision provided no information on the daily timing of the congestion claimed by the networks. This is a serious omission. I appreciate that Canadian networks are connected to those of other countries, and local use constitutes only part of the traffic—particularly for the Internet. But we still need to know about the timing of congestion to see if off-peak pricing is possible. Metering There was no discussion of the metering problem at the level of the retail customer—the focus of the decision. Usage-based pricing can only achieve its potential in terms of allocative efficiency (and fairness) if users are aware of usage (hence costs incurred) "on the fly." However, there is no way for households to do this! There is not even a device for households to measure usage every few days—like a gas gauge. They only know that they must have used less than their monthly quota because they did not get a bill for any "overage." Prof. Tim Wu argued in a column in The Globe on Feb. 9 that, "Any rule that asks the consumer to guess at usage, and punishes you if you're wrong, is abusive.” Multi-use networks The CRTC failed to explain that there is no structural separation. The networks used to supply Internet access are also used to deliver a variety of other services, and that the incumbent network operators are in two businesses: (i) the "pipeline" (or wholesale) business of supplying capacity to other firms (like independent ISPs) who use the network to supply services to others; and (ii) the sale of services over the same network to their own customers (the retail business). Thus the CRTC needed to get from the network owners data on the composition of the amount of throughput used by various types of services delivered over each major network. Pricing overage The decision contained no discussion of the incumbent network operators' pricing of additional gigabytes of usage above the monthly quota (overage). Thompson contacted several industry insiders who told him that the operational and fixed costs of transporting a gigabyte of data had dropped to around six cents two years ago and is about three cents today (The Globe and Mail, Feb.10, 2011). Thompson stated that Bell's "true cost of data transfer is well below the 10.5 cents per gigabyte they are charging wholesalers." He noted that various network operators had posted prices from $2 to $5 per gigabyte for usage above the monthly quota of bandwidth. He concludes that, "Assuming an inflated cost of 10 cents per gigabyte, it means that Bell, Shaw, and Rogers are charging consumers between 10 and 50 times what it costs them to deliver data" at the margin. Only in the chair's speech is the crucial point about the market structure mentioned: "The market is dominated by what many have called a duopoly: large telephone companies such as Bell Canada and Telus, on the one hand, and cable companies such as Rogers, Shaw, and Videotron on the other." In the decision, however, the CRTC notes that these "carriers' retail UBB rates are market-based and are not subject to prior commission approval—that is, they are forborne from regulation" (see paragraph nine). I would emphasize that "market-based" does not necessarily mean that such prices are as low as would prevail in a highly competitive market. Policy direction Having indicated that the incumbent networks owned by the telcos and cablecos have considerable market power, the CRTC decided not to regulate the prices of gigabytes above the monthly usage quota because of the policy direction issued by the federal cabinet in December 2006. Thus, in the UBB decision, the CRTC mandated a new price structure, but ignored the elephant in the room. It ignored the evident ability of the incumbent network operators to set their prices for marginal usage far, far above plausible estimates of long-run average total costs as Thompson and others suggest. The commission failed to address this point. It could have held a mirror up to the Harper government—the same one that issued the direction—and said that it exercised forbearance in order to comply with its direction. But it should have advised the government that the market power of the incumbent network owners was sufficient to permit supra-competitive pricing. Another unmentioned elephant The CRTC should have also referred to that other elephant in the room, namely the government-imposed barriers to foreign competition in the supply of broadband network services (and all telecom and broadcasting services). The Harper government has waved it arms periodically about reconsidering the foreign ownership rules (and it overturned the CRTC in the Globalive decision), but it has not moved to increase competition at the network level by eliminating the suffocating constraints on foreign ownership. Cross subsidization The decision said nothing of the unfairness of cross subsidization. However, in his speech on Feb. 3, the chair stated: "The vast majority of Internet users should not be asked to subsidize a small minority of heavy users. For us, it is a question of fundamental fairness." The chair cited data from Bell Canada indicating that less than 14 per cent of users are responsible for more than 83 per cent of Internet traffic. But no detail was provided, so all we know is that Pareto's Law applies here as expected. More importantly, the CRTC completely failed to address the positive argument that cross-subsidization results in allocative inefficiency. Fifteen per cent discount While Bell argued there was no basis on which to determine an appropriate level of discount (see paragraph six of the decision), the independent ISPs said it should be a minimum of 50 per cent (paragraph eight). The CRTC provided no justification—in the sense of a reasoned explanation—for the 15 per cent discount for independent ISPs. It is hard to avoid the conclusion that it was simply arbitrary. That is unacceptable. Discussion The adverse reaction to the UBB decision was reasonably predictable—despite the application of UBB in some form since at least 2006. Hence the need for the CRTC to set out fully the facts, inferences, larger context and reasoning for the decision—which it conspicuously failed to do. This is not to say that the participants in the Facebook eruption (over 400,000 signed in)—prompted by some of the independent ISPs most affected by the decision—objecting to UBB will ever be placated by the most thorough examination and explanation of the logic for UBB pricing. Many seem to believe that the Internet has repealed the basic laws of supply and demand. The crux of the problem appears to be the growth in the use of the Internet to deliver services that use a great deal of network capacity. "According to Rogers Communications, downloading about one-and-a-half standard-definition movies, or about two-thirds of a high-definition movie, represents one gigabyte of usage, as do each of the following: downloading about 200 songs; playing online games for 240 hours; viewing 26,000 web pages. Sending 105,000 e-mails," wrote Rita Trichur in a Globe and Mail story published Feb. 6. One heavy user states, in another Globe and Mail report on Feb. 9: "I used to buy CDs and physical books and DVDs like everyone else … Nowadays, I download it all." However, the CRTC failed to discuss this issue in the decision. If deference is to be paid to quasi-judicial and quasi-independent regulatory agencies—by the government, by the courts, and by ordinary folks subject to the decisions—then they must constantly demonstrate their expertise. The CRTC has failed to do this in the UBB decision. On March 11, the CRTC stated that in its review of the UBB decision, it will focus on usage-based charges, peak traffic periods for Internet use, and the issue of network capacity and congestion. However, it "will not review, as was also requested by several parties, the overall regulatory framework for wholesale high-speed Internet access services." The sad fact is that even if the CRTC does provide a well-reasoned analysis to support UBB in the review now underway, the government has indicated that it will not accept UBB for Internet access for households. W.T. Stanbury is a professor emeritus of the University of British Columbia. This article originally appeared in The Hill Times on March 21. news@thewirereport.ca

CRTC’s UBB decision: pretty good economics, but a bad explanation

News | March 23, 2011

Unlock all the Canadian telecom, broadcasting and digital media news you need.

FREE TRIAL

SUBSCRIBE

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us