Analyzing the spectrum “crisis”: traffic vs revenue

Posted in Regulatory, Spectrum at 4:53 pm by timfarrar

After looking at the consequences of what appears to be a mathematical error in the FCC’s Mobile Broadband Spectrum Forecast last week, I thought it would be useful to look at some of the underlying traffic growth forecasts themselves.

The FCC forecast gives some detailed projections of traffic growth per device and device penetration in 2014 in a couple of its exhibits, based apparently on Cisco data, split between “basic and feature phones”, “mobile broadband handsets” and portable devices (“computing devices (netbooks and notebooks), tablets, handheld gaming consoles, e-readers, digital cameras and camcorders, digital photo frames, and in-car entertainment systems”) either as substitutes or complements to fixed broadband. The forecast assumes rapid growth of take-up and traffic for data intensive devices, but what is remarkable is that more than two-thirds of all the traffic in the Cisco model comes from portable devices. However, under any reasonable assumptions about ARPUs from each type of device (unless you assume “voice will be free” as Cisco used to assert) – say $40 per month for “basic and feature phones”, $69 for mobile broadband, $35 for fixed broadband and $25 for portable devices, then less than 10% of 2014 revenue comes from this 68% of traffic, as shown in the chart below. (Note that these ARPUs were chosen to match with annual revenue growth of 7.5% p.a., which is in line with the growth rate from 2005 to 2010 according to the CTIA).

In this case, the solution to the spectrum “crisis” is pretty clear (especially for AT&T and Verizon) – sell people DSL/FiOS/U-verse/cable broadband with WiFi, and limit their use of complementary portable broadband devices, unless they are either high end (laptop) users, or don’t use much bandwidth (Kindles). In any case, certainly don’t allow them unlimited usage. If you’re a second tier player then you have to decide if you can afford to meet the needs of these portable device customers, or if targeting the other 90% of revenues is enough for now. If you’re a new entrant trying to go after these portable device customers, then you’d better hope your spectrum really is very cheap. Perhaps you can hope that by serving people’s portable devices you can capture all of their mobile spending (i.e. provide them with a phone as well)? Maybe that could be the case some of the time (though somehow I doubt that buying a Kindle or a camera or a photoframe is going to make you switch cellphone provider), but isn’t it just as likely that the big players will offer “family plan” type bundled rates for various other devices and just keep a tight lid on their usage? How much more do you really want to spend each month on all these devices in addition to your current cellphone bill (indeed is $25 per month even remotely achievable as an average ARPU for these sorts of devices)? After all, a lot of people (myself included) have apparently decided not to bother with 3G iPad connectivity at $25 per month.

Perhaps it could be argued that in fact voice (VOIP) might well be free eventually, so operators have to get used to that and go after these portable device customers instead. However, that outcome is even worse for spectrum valuations, because if 70% of current wireless revenues are destined to disappear then investment dollars for 4G buildouts and spectrum will be pretty hard to come by.

UPDATE: We understand that there is an inconsistency in the FCC report between the data for usage growth per device from 2009-13 (from the Cisco 2009 report) and the penetration in 2014 (from the Cisco 2010 report) due to a major change in the definition of “portable complementary” devices. As a result the 2009 Cisco report predicts there will be about 11M portable complementary devices in 2013 but the 2010 report predicts there will be 51.8M of these devices in 2014. Because the underlying data in the Cisco report is not published (and no definitions appear in the paper), it is hard to see whether the 2009-13 data usage numbers are still valid for 2014 or not. However, our estimates are quite close to the total data usage predicted for North America in 2014 and Cisco certainly assumes that the vast majority of data usage in 2014 comes from “portables, netbooks and tablets”.

Leave a Comment

You must be logged in to post a comment.