AT&T’s residential market shortcomings

 

As AT&T would have it, the telecommunications giant is enthusiastic about serving the residential market and connecting homes to fiber. AT&T Communications CEO Jeff McElfresh told a Bank of America Merrill Lynch TMT Conference in June 2020 the company will increase its investment in fiber connections. "We are laser-like focused on finding the most efficient path to expanding the footprint of our fiber offerings," McElfresh said. "It's a great business. It's got great margins. It's got great returns. There's nothing not to like about it, and we're going to lean into it."

McElfresh’s comments represent a turnabout from a year earlier, when AT&T downplayed its residential fiber ambitions and dismissed hundreds of field technicians after completing a limited build out to meet regulatory obligations attached to its acquisition of DirecTV. "That's behind us now," McElfresh’s predecessor John Donovan told FierceTelecom. "We'll continue to invest in fiber, but we'll do it based on the incremental, economic case. We're not running to any household target."

For single family home neighborhoods, AT&T makes residential fiber available only to discrete pockets, reports industry observer Doug Dawson. AT&T also markets residential fiber to multifamily dwellings that require less capital investment and produce comparatively faster returns. One analyst suggests AT&T is weak at executing fiber build outs, unable or unwilling to focus on the necessary details of neighborhood telecommunications infrastructure deployment. (Jim Patterson, Curing AT&T’s Sickness, 10/12/20) Other analysts point to high debt on AT&T’s balance sheet that constrains its ability to finance a broad move into residential fiber.

In less dense exurban and rural neighborhoods, AT&T is phasing out its legacy ADSL service, halting new connections as of October 1, 2020. In these areas, AT&T offers fixed wireless residential service over its 4G LTE mobile infrastructure but with throughput limited to a small fraction of what a fiber connection could handle. Dawson notes the company has not actively marketed the service (most likely to preserve limited radio spectrum at the same time the company encourages high bandwidth video streaming). Moreover, the company was notably absent among bidders for the FCC’s recently closed Rural Digital Opportunities Fund (RDOF) subsidy reverse auction.

Where AT&T is building fiber to serve enterprise consumers (via dedicated Ethernet) it is not generally investing in premise drops and field distribution equipment to serve adjacent single family home residential neighborhoods. According to an October 2020 report by the Communications Workers of America, the labor union representing AT&T line technicians, and the National Digital Inclusion Alliance, 63 percent of 1,500 line technicians surveyed report that AT&T is not installing splitting equipment to enable home connections even where a fiber backbone exists.

With little focus on residential fiber, AT&T is instead looking to gain revenues in the consumer segment from streaming video and mobile wireless offerings as it experiences a steady decline in linear TV subscribers, legacy and wireline delivered services, according to Zacks Investment Research.

AT&T seeks state sanction to exit residential premise service, transition customers to mobile wireless

Fellow blogger Steve Blum of Tellus Venture Associates calls bullshit on AT&T for sponsoring California legislation that would relieve it from its premise landline universal service obligations under Title II of the federal Communications Act. Blum has the same problem with the bill as I do. It's dressed up as enabling AT&T to transition from copper POTS service to Internet protocol-based service. As Blum points out, AT&T can do that without the need for enabling legislation. It has chosen not to make an orderly transition over the past two decades. That's a business issue, not one of regulatory policy.

The bill is essentially seeking state sanction to transition AT&T residential landline customers to its mobile wireless service. The thing is, that's not premise service under Title II's universal service obligation. However, with the U.S. Federal Communications Commission not enforcing its 2015 Open Internet rulemaking bringing IP-based services under Title II's universal service requirement, AT&T faces no regulatory consequence for "mobilizing the world" of its residential customers with service not engineered or priced for residential premise service.

Strong parallels between individual health insurance and Internet service markets

Telecommunications providers, consumers and policymakers should be aware of the strong parallels between wireline residential Internet service and the individual health insurance marketplace as it existed prior to the market reforms of the Patient Protection and Affordable Care Act.

Both residential Internet providers – and formerly individual health insurers -- shared a business model whose success is ironically predicated on not selling to all potential customers in their market areas. The underlying principle is risk aversion: vendors believe they cannot adequately manage the risk of loss associated with an expanded market. A smaller, more certainly profitable customer base is better than a larger one notwithstanding the potential for greater revenues.

Before the Affordable Care Act outlawed the practice starting this month, individual health insurers employed underwriters charged with selecting relatively healthy individuals less likely to incur high medical costs, refusing to offer coverage those who didn’t meet specified underwriting standards. Similarly, wireline residential Internet service providers offer service to one address while declining to serve another nearby – even as close as quarter of a mile away or less. As individual health insurers did, these providers reject the latter residences (as well as some small business sites) as more costly to serve and thus less potentially profitable. Their infrastructures are engineered and built to accommodate preferred addresses and redline the rest.

The problem with this business model for individual health insurers is that medical underwriting limited the size of the pool of individuals and families who could pay premiums to cover claims costs. Consequently, the pool and the number of healthy people staying in it shrank to the point it was on the verge of collapse when the Affordable Care Act was enacted in 2010. The federal law intervened to head off market failure by requiring health plan issuers to sell to anyone applying for coverage regardless of medical history or health condition.

The restrictive marketing practices of Internet service providers bring about a similar problem. Just as insurance pools are more viable with more people in them, telecommunications networks are more valuable when more people are on them or able to get on – both for providers and subscribers. This principle is known as Metcalfe’s Law. Those who argue for broader deployment of fiber to the premise (FTTP) Internet infrastructure carry over the Metcalfe principle to economic activity and education. Greater numbers of premises with modern Internet access can lead to more online commerce and business formation. Increased access to information and educational curricula, similarly, lead to a more informed and better educated society.

As time goes on and these broader benefits – and conversely costs of not having affordable premises Internet access – become more evident, it could lead to large scale market reforms such as are now reshaping the individual health insurance market.

First indication of AT&T withdrawal from residential wireline market

Sensing AT&T's lukewarm commitment to its residential wireline business segment, in 2008 I predicted that AT&T would abandon the segment in the first half of 2010. The telco is still in the residential wireline business as 2013 draws to a close. But a slow withdrawal could now be underway, one state at a time starting with Connecticut.

Bloomberg reports today that AT&T will spin off its Connecticut residential landline unit, including Internet and TV services to Frontier Communications for $2 billion.

AT&T relies on copper cable plant to deliver premises Internet service, scotching plans dating back to the late 1980s developed by regional bell operating companies AT&T absorbed in the 1990s to replace the last mile copper network with fiber optic cable. That reliance has technologically limited the reach of AT&T's Internet-based service offerings since copper was designed to carry analog voice service and not digital Internet signals that can be reliably delivered over only short distances using copper.

AT&T's relationship with Connecticut hasn't been a copacetic one. In 2007, then-Attorney General Richard Blumenthal pressured the telco to make its U-Verse product offering available to all residences in the state. Blumenthal, now a U.S. senator, said this week the deal should be reviewed to ensure it is in the interest of consumers.

Copper vaporware as AT&T chief declares DSL obsolete

Every couple of years or so, an article like this one by Tara Seals of V2M appears arguing legacy copper telecommunications infrastructure designed for a pre-Internet analog era is far from obsolete. Technical innovations can extend its lifespan, even as bandwidth demand is increasingly challenging its carrying capacity, particularly from Over The Top (OTT) video content:

Telcos are seeking cost-effective solutions to maximize their legacy infrastructure. Reducing crosstalk across copper bonded pairs using the ITU-T G.vector standard (G.993.5), introducing software solutions to maximize network logistics and using caching in the network are all solutions that are occurring right now, as telcos position themselves to meet the rapidly growing consumer OTT demand.

If that's the case, then why did AT&T CEO Randall Stephenson declare this week that the workhorse technology that has transported Internet protocol content over AT&T's copper network for the past decade and a half -- Digital Subscriber Line (DSL) -- is obsolete?

What about that innovation to stave off copper obsolescence? If it were for real instead of vaporware hype, it would truly provide AT&T tremendous opportunity to offer more wireline Internet services to a lot more customers over its legacy copper plant. Clearly for AT&T, that's not the case as the telco shifts away from residential wireline and is instead concentrating capital expenditures on personal wireless services.

Motorola CEO, FCC agree: Wireline is for premises Internet connectivity, wireless for mobile access

With telco wireless providers rolling out 4G service, some have raised hopes that it has the bandwidth to rival facilities-based customer premises wireline infrastructure to deliver the full spectrum of Internet service.

Motorola CEO Sanjay Jha dismisses that notion as unrealistic. Wireline will continue to be the optimum method of providing Internet connectivity and high bandwidth video content for “a long time to come,” Jha was quoted as saying this week by The Wall Street Journal.

Jha added that wireless spectrum in the U.S. remains “very limited,” and wireless providers will continue to be the primary supplier of Internet access to consumers outside their home.

Jha's perspective on the respective roles of wireline and wireless is shared by the U.S. Federal Communications Commission. In a May 20, 2011 report to Congress on the availability of advanced telecommunications capability as mandated by the Telecommunications Act of 1996, the Commission rejected arguments by some mobile wireless providers claiming they can provide sufficient bandwidth to substitute for wireline service.

AT&T exec suggests wireless will save its residential market segment

AT&T may be the nation's largest telecommunications company. But its size hasn't helped it meet the challenge of upgrading its cable plant to transport Internet protocol-based services. AT&T provided wireline Internet connectivity first through dial up and ISDN connections in the early 1990s, and then DSL as the 1990s turned into the 2000s. Starting in 2006, AT&T brought fiber closer to customer premises -- but not to them -- with its FTTN (Fiber to the Node) U-Verse service utilizing VDSL. Some new, dense greenfield developments received U-Verse service via direct fiber to the premises connections.

New home construction cratered shortly after U-Verse rolled out, leaving only more challenging FTTN brownfield opportunities. They are more challenging because the old cooper cable plant designed for POTS (Plain Old Telephone Service) is used to carry high compressed VDSL signals that quickly degrade with distance, limiting the size of the potential U-Verse customer base.

Faced with these challenges to reach customer premises and seeing strong growth on the wireless side of its business, AT&T not surprisingly sees its future in the wireless space. "The future is wireless broadband and we must keep that in front of us at all times," Tim Ray, executive director for AT&T External Affairs in Northern California, said at a recent roundtable discussion hosted by Sacramento-based Valley Vision.

In 2010, Valley Vision formed the Connected Capital Area Broadband Consortium (CCABC), a coalition "which seeks to identify and coordinate strategic broadband investments in the six-county Sacramento region aimed at improving broadband infrastructure, access and adoption." Ray, who sits on Valley Vision's board of directors, appeared to suggest wireless Internet connectivity will be able to substitute for wireline connectivity, noting "27 percent of homes no longer have wire line and this trend will continue to grow."

Ray's wrong and engaged in wishful thinking. There's currently nothing indicating wireless Internet service -- which is aimed at mobile devices with a low bandwidth allocation per customer  -- can provide sufficient capacity to handle burgeoning bandwidth consumption and be able to reliably deliver to customer premises high definition video content and applications like video conferencing and telemedicine. Indeed, AT&T's wireless infrastructure is already choked with far lower bandwidth traffic from devices such as the iPhone.

AT&T is in conflict with its own business model. It's in the telecommunications business which by its nature requires lots of CAPEX and OPEX. But it expects to get a full ROI within 5 years on its CAPEX. That's not going happen in most places except perhaps in new dense greenfield developments, which as previously mentioned also aren't happening.

U-Verse won't bail AT&T out of its residential wireline woes

Here's a notable report by Todd Spangler of Multichannel News on AT&T's revenues from its hybrid fiber/copper VDSL triple play U-Verse service that suggests while posting increases in customers and revenue, they may be too little and too late to offset a dramatic decline in AT&T's residential wireline market segment. In a Dec. 21, 2009 filing with the U.S. Federal Communications Commission, AT&T in unusually blunt language called the downward trend a "death spiral."

Spangler reports that while AT&T's U-verse revenue nearly tripled over 2009 (despite a sharp economic downturn) and is approaching an annualized rate of $3 billion, it nevertheless represents less than five percent of total wireline segment revenue. Spangler notes even that strong growth isn't sufficient to offset flagging wireline segment revenues, which fell six percent in 2009 to $65.7 billion.

Meanwhile, AT&T disclosed this week it would spend $2 billion on its wireless infrastructure -- money that won't be going into wireline CAPEX to build out the U-Verse footprint. Doing that is a costly proposition given U-Verse involves expensive field distribution equipment that can deliver service only 3,000 feet over existing copper cable plant -- plant that often requires even more money to bring it up to technical standards to reliably carry VDSL signals. That's not an issue in new neighborhoods, where U-Verse is delivered over fiber to the premises. But few such locales are being developed with new home construction at its lowest level in decades.

In sum, U-Verse isn't likely going to bail AT&T out of its troubles in residential wireline and may ultimately lead to the big telco pulling out of the market segment to concentrate on wireless in the retail market as I predicted in September 2008.

Why telcos drag their feet on residential broadband

In the fall of 2007, Ralph de la Vega, AT&T's group president for regional telecommunications and entertainment made a pronouncement with profound implications that were largely overlooked in the mainsream media.

de la Vega told Investor's Business Daily that AT&T would ultimately shut down its existing voice network and replace it with a VOIP (Voice Over Internet Protocol) system in metro areas where U-Verse is being deployed.

Since U-Verse deployment has been delayed and scaled back, it calls into question the future of AT&T's wireline residential market segment. Essentially de la Vega pronounced the beginning of the end of the Publicly Switched Telephone Network (PSTN) and its replacement by the Internet with Next Generation Telephony.

That also means telcos' proprietary central office switches are on a fast track to obsolesence, destined to be replaced with Internet servers and field-based fiber optic distribution equipment. Industry observers like Bob Frankston are right to accuse telcos of foot dragging by creating artificial bandwidth scarcity and restricting broadband access in order to live in the copper-bound PSTN world for as long as possible. This is the unspoken subtext to the larger Strum und Drang on this blog and elsewhere over the pathetically poor state of broadband availability in much of the United States. It's typically explained as a simple return on investment problem, but there's more to it than that.

As the Internet wreaks massive disruption in mass media, it also threatens an end to the days of Plain Old Telephone Service (POTS) delivered over twisted copper. Just as people are canceling their newspaper subscriptions, they are also ditching their residential land lines. And who can blame them when all they can get over them is POTS and perhaps DSL (an acronym that should mean Doesn't Serve Lots)?

It also explains why first tier telcos like AT&T are redefining the residential wireline segment as "personal wireless" services since this segment can remain proprietary if residential wireline moves out of the old proprietary, closed system scheme and into one where last and some middle mile infrastructure is owned and operated by small local providers, local governmental entities and telecom cooperatives.