When the usage based billing controversy grabbed national headlines last month, I posted several long pieces on the issue. My first post attempted to unpack the issue, focusing on the some the misleading claims about the supposed need for UBB at the wholesale level and putting the use of data caps in Canada in context. Later posts discussed concerns with the CRTC approach on UBB, the claims regarding network congestion, and a piece on what should come next, including what the CRTC should do on wholesale UBB and the broader policy measures on foreign ownership restrictions, fostering greater competition, and addressing the retail UBB concerns.
Those posts attracted some attention and soon afterward I was asked by Netflix if I was interested in digging deeper into these issues. The company provided some support so that I was able to quickly assemble a great team of students – Keith Rose, Peter Waldkirch, Tyler Nechiporenko, and Rachel Gold – to delve into issues such as congestion claims, comparative UBB approaches, the cost of transferring a GB of data, and some potential solutions. I completed the paper over the weekend and have posted it here. I’ll be posting on several elements over the next few days, concluding with my proposal for the establishment of a UBB equivalent for Internet Traffic Management Practices (ITMPs), which I’ve dubbed Internet Billing Usage Management Practices or IBUMPs.
This first post features a comparative look at usage based billing in other countries.
The OECD has noted that Canada is one of the only jurisdictions in the world where virtually all providers utilize some form of UBB. For most providers, data caps have been in place for years: Rogers established a monthly cap to some of its plans in 2005 and Bell implemented data caps in 2006.
While much of the focus has been on the widespread use of UBB in Canada, more instructive is the near-uniform implementation of UBB. In Canada, UBB invariably means a cap on monthly data use combined with overage charges for all data used that exceeds the cap. The plans do not distinguish between peak and off-peak use, the source of the data, or contemplate other solutions beyond overage charges.
The near-uniform Canadian approach is notable because other, more competitive countries feature a wider variation in approach, as highlighted in the comparative chart of 18 other countries in Appendix A of my paper (the chart is also available as a standalone document). Countries such as Germany give consumers the ability to select from among unlimited and capped services. Deutsche Telekom, the largest providers, offers unlimited plans, while providers such as 1und1 offer lower cost plans with a 100 GB data cap. The U.S. market similarly features a mix of capped and unlimited plans. Comcast offers a 250 GB cap and AT&T recently announced plans to implement a 150 GB cap. Other leading ISPs such as Cablevision and Verizon offer unlimited plans.
In addition to choice from among unlimited and data capped plans, other countries feature far more variation in approach to UBB. In Australia, data caps are very common, yet exceeding the cap does not result in overage fees. Instead, Australian ISP â€œrate limitâ€ subscribers that exceed their cap by reducing download speeds for balance of the month. In other words, all plans are effectively unlimited plans but with different speeds once a data threshold is met. The use of rate limiting is a relatively common approach in countries with data caps as it is also found in New Zealand, Germany, and India. Moreover, in response to competitive pressures, British Telecom recently announced plans to eliminate data caps but retain rate limiting for heavy Internet users.
The Australian market also features another notable variation on UBB. Many ISPs feature two separate data caps each month – one for peak time usage and a second for off-peak usage. For example, Optus offers several broadband plans which each feature separate data caps for peak and off-peak usage. The peak period runs from noon until midnight, while the off-peak period runs from midnight until noon. If a subscriber exceeds the peak or off-peak cap, Optus reverts to a rate-limited service. IiNet offers a similar peak/off-peak data capped service with consumers able to purchase as much as 1 TB of data per month.
Another alternative approach is to distinguish between domestic and international data. For example, Vodaphone Iceland offers a 10 GB capped plan but â€œall data downloaded domestically is unlimited without any extra cost.â€ The effort to distinguish between the origin or destination of data reflects different costs â€“ regional data transfers may remain within a single ISP’s network or is subject to a peering arrangement that reduces transfer costs. By contrast, international data will often require a transit arrangement that increases data transfer costs.
Yet another approach is to distinguish between upstream and downstream traffic. Data caps are still relatively rare in Japan, but those ISPs with data caps limit their application to upstream traffic. Moreover, the caps are so high – 900 GB of upstream traffic per month is standard – as not constitute caps in comparison with the policies in Canada.
What makes these UBB approaches noteworthy when viewed from a Canadian perspective is that there is sufficient competition such that consumers in other countries can typically choose between unlimited and capped plans. Moreover, the capped plans elsewhere bear a more direct relationship to cost and congestion concerns. In contrast, Canadian UBB bears little relation to actual cost, but is instead largely a function of market dynamics and the lack of competitiveness. I’ll focus on those market dynamics issues in tomorrow’s post.