Last week I posted on a recommendation by the Technical Advisory Council (TAC) to the Federal Communication Commission (FCC) that the FCC set a date certain for the sunset of the Public Switched Telephone Network (PSTN). The discussion has been great: over 300 comments on slashdot; many more on gigaom, dslreports, fractals of change, and other sites. And lots of good issues raised; some of which I'll discuss in this and future posts.
Note: Although I am a member of the Critical Transitions workgroup of the TAC which made the recommendation, I speak for myself and no one else on this blog.
A common thread in many comments is the claim that the PSTN is better than any existing alternatives for voice communications including wireless and VoIP. Reader Scott wrote:
"Before even considering abandonment of the copper-based PSTN infrastructure, the reliability and quality issue with the alternatives needs to be addressed. Both cellular and VOIP systems require electrical power at both the transmitting and receiving end. When this fails, the service also fails. By contrast, the PSTN is designed so that the system is powered by CO equipment, which tends to be massively redundant. If the electrical grid fails, the PSTN should stay up for weeks, without requiring anything on the customer side. The same cannot be said about today's alternatives.
"Also, voice quality is a huge issue. Both cellular and VOIP systems introduce latency and compression, neither of which is desirable. It is just said that my Western Electric 302 desk set from 1938 has better sound quality than a modern cellular telephone or VOIP telephone. Also, home-type VOIP services tend to not support QoS [nb "quality of service"] routing, and performance can degrade based upon network traffic…
"Until these issues are addressed, discussion of the demise of the PSTN is premature…"
Scott's points aren't wrong. They are selling points for the PSTN. But the public isn't buying these selling points; customers are abandoning the PSTN and cancelling their landline POTS (Plain Old Telephone Service) at an incredible rate despite the fact that Scott and others think that the PSTN is superior to the alternatives. Below from a twice annual huge telephone survey by the Centers for Disease Control:
"Preliminary results from the July–December 2010 National Health Interview Survey (NHIS) indicate that the number of American homes with only wireless telephones continues to grow. Three of every ten American homes (29.7%) had only wireless telephones (also known as cellular telephones, cell phones, or mobile phones) during the last half of 2010—an increase of 3.1 percentage points since the first half of 2010. In addition, nearly one of every six American homes (15.7%) received all or almost all calls on wireless telephones despite having a landline. This report presents the most up-to-date estimates available from the federal government concerning the size and characteristics of these populations."
[BTW, in last week's post I ran a graph from this study which predicted that, by 2018, only 6% of residential landlines would be PSTN. Readers questioned the methodology of the study and I had carelessly omitted a link. Details on methodology and results are at http://www.cdc.gov/nchs/data/nhis/earlyrelease/wireless201106.htm.]
Note that the NHIS survey is only about families abandoning landlines completely for a wireless alternative. They count VoIP lines as landlines, which they are although they are not part of the PSTN. The best estimate I can find of residential VoIP line penetration, including those VoIP services provided by cable companies, is from the FCC. They estimate that, at the end of 2008 (latest data they have), 20% of landlines were VoIP rather than PSTN (they don't count services like Skype but do count Vonage). The percentage has certainly gone up with the success of the cable companies in selling this service; one reputable source estimates that the cable companies alone serve 20 million subscribers with VoIP; Vonage advertises having 2.4 million subscribers (some not in the US). On the other hand, we know that some homes (mine, for example, for just a little while longer) have both VoIP and PSTN landlines. However, if 20% of the 70% of households with landlines actually have only VoIP lines, then PSTN penetration is already down to 56% - and falling. Note that a household with both a PSTN line and a VoIP line (mine for example, again) may be using the VoIP line for all calling and only have the PSTN line so that a home security system can continue to function; this means no toll revenue for the legacy phone company. Home security systems, however, now communicate both over IP networks and the cellular network, so even this need for PSTN is fading.
Back to Scott's points about quality: voice quality on the PSTN has been better than both VoIP and cellular. People have chosen price and mobility over quality; but we won't have to make that tradeoff much longer. New cellphones will support high definition voice – much better than what is available on the PSTN because they are not stuck with the bandwidth limitations which the PSTN coped with so well for so long. VoIP also can support high def voice when it doesn't have to interconnect with the low def PSTN. Have you noticed how good the audio is on a Skype-to-Skype call? That's only the beginning. I can't think of anything that would be better for E911 service than higher definition voice – less confusion between "got into the shed" and "shot in the head"!
The cost of running the copper-based PSTN is largely a function of route miles upon which trees can fall and across which strong winds blow. The revenue available to support the PSTN is dependent on the number of subscribers and the amount of calling they do. In rural areas the PSTN has long been subsidized by both the Universal Service Fund (USF) and a series of indirect subsidies. As people abandon the PSTN and email and text more and talk less, the subsidy required per subscriber goes up. Even if some users like Scott would prefer to keep their PSTN service, we all need to ask how long we will subsidize this once-great but now antiquated service and how much subsidy per user we will pay. People who have chosen not to use the PSTN anymore – regardless of its merits – shouldn't have to pay to subsidize those who do want to use it.
The decision of whether or not the PSTN should live on has already been made by the marketplace responding to better features, mobility needs, and the lower costs of VoIP networks. We need to recognize PSTN's demise and plan accordingly to make sure everyone has an alternative and to understand all the consequences of the end of this venerable network. I believe the best way – perhaps the only way – to have this discussion is set a date certain for the end of government support for the PSTN and government mandates that PSTN services be provided and work backwards from there.