How low can AT&T go? Customers retaining “unlimited data plans” that were discontinued in 2010 were throttled to as little as 127 kbps after using just 2 GB a month.

AT&T’s lawyers are asking a judge to accept media coverage exposing the company’s allegedly “secret” speed throttling policy for some of its wireless customers as a valid defense in a 2015 class action case that seeks to compensate some AT&T customers for misrepresenting its “unlimited data plan.”

AT&T last month asked the judge to have the long-running case thrown out, claiming AT&T well publicized its new speed throttling policy it imposed on a legacy unlimited data plan the wireless company stopped selling in 2010, but allowed existing customers to keep. By 2011, some customers still subscribed to the grandfathered unlimited plan started noticing data speeds plummeting to near dial-up if they used a lot of data. At first, AT&T appeared to impose a speed throttle on customers using over 10 GB of data per month, but by 2012, AT&T was accused of speed throttling unlimited customers after they used as little as 2 GB of data during a billing period.

The resulting class action lawsuit, filed in California, alleged that AT&T misrepresented its unlimited data plan as ‘unlimited,’ when in fact in practical terms it was not. The plaintiffs are seeking damages from AT&T to discourage the company from engaging in false advertising in the future, and to compensate customers that paid for an unlimited data plan that eventually became almost useless after customers used just over 2 GB a month.

AT&T’s defense partly relies on the company’s claim it extensively publicized changes to its legacy unlimited data plan as early as 2011, and the plaintiffs should have been aware of it. The Federal Communications Commission was aware of AT&T’s actions and just a month before the class action case was filed, the regulatory agency issued a notice of apparent liability to AT&T proposing a $100 million fine for unwarranted speed throttling.

AT&T’s attorneys have worked hard to stop the lawsuit over the last five years. In addition to claiming customers were notified of their excessive data usage through text messages and billing notices, AT&T last month sought to introduce a dozen media reports covering its speed throttling policy into the court record to convince U.S. District Judge Edward Milton Chen the plaintiffs don’t have a case and to get the lawsuit dismissed.

One of the news articles cited in AT&T’s May 14 filing was written by former DSL Reports’ author Karl Bode, who has been roundly critical of AT&T’s data caps for over a decade. Ironically, AT&T’s defense team is arguing Bode’s report, “AT&T Wages Quiet War on Grandfathered Unlimited Users” offers proof AT&T was not keeping its speed throttling policy “secret,” as at least one plaintiff claimed. Bode suggested AT&T had engineered its speed throttling plan to push grandfathered unlimited data plan customers off the plan in favor of more profitable plans offering a specified data allowance and overlimit fees.

Bode

“In other words, pay $30 for “unlimited” service where you’re actually only getting 2 GB of data before your phone becomes useless, or sign up for a 3 GB tier for the same price so you’re in line to get socked with the usage overages of tomorrow,” Bode wrote at the time.

His views have not changed in 2020.

“For nearly a decade AT&T has tap danced around the fact it misleadingly sold an ‘unlimited’ data plan packed with confusing limits. No amount of legal maneuvering can hide the fact that AT&T lied repeatedly to its customers about the kind of connection they were buying,” Bode told Stop the Cap! “Instead of owning its mistake, learning from it, and moving forward, AT&T’s now trying to point to critical news coverage from the era to falsely suggest consumers should have known better. It’s utterly nonsensical and speaks volumes about the lack of ethical leadership at a company that routinely sees some of the lowest customer satisfaction ratings in American industry.”

AT&T’s lawyers are not prepared to concede, however. Since the lawsuit was filed, AT&T’s legal team attempted to force the case into arbitration in 2016. That effort was successful until a 2017 California Supreme Court decision in another case gave the plaintiffs ammunition to claim that it was against California law to force consumers into arbitration. The Ninth Circuit court agreed, and the case reverted to district court, where AT&T immediately began efforts to have the case dismissed outright.

AT&T is not alone throttling so-called “heavy users” that have either legacy or current unlimited data plans. All major cellular companies enforce fine print policies that allow speed throttling after customers consume as little as 20 GB of wireless data during a billing cycle. The fact companies still advertise such plans as “unlimited” irks Bode.

“An unlimited data connection should come with no limits. If giant wireless carriers can’t respect the dictionary, they should stop using the word entirely,” Bode told us.

The death of DSL?

If your local phone company does not offer fiber-to-the-home service, it risks seeing its market share as a broadband competitor drop to zero, according to new research from Wall Street analyst firm MoffettNathanson.

As the cable industry prepares to deploy DOCSIS 4.0, capable of much faster upload speeds in the gigabits and downloads as fast as 10 Gbps, the future of telephone companies that have under-invested in their networks for years is dire. The research firm’s “Equilibrium Forecast” sees DSL’s market share in areas where cable broadband is available dropping to zero. Phone companies that have invested in fiber half-measures, including fiber to the neighborhood, IP-DSLAM, and VDSL technology that traditionally delivers internet speed between 25-75 Mbps are not far behind. Only true fiber-to-the-home service stands a chance at protecting phone company broadband market share.

“DSL [and] mid-tier [fiber/copper combinations are both] obsolete,” researchers said in a private note to investors. “Broadband is increasingly a two-horse race between cable and telco fiber-to-the-home service, where it exists.”

The COVID-19 pandemic has only increased problems at the nation’s legacy phone companies, as customer losses accelerate in favor of cable company delivered internet. In the first quarter of 2020, cable company broadband sign-ups increased 122% compared to the same quarter last year, while phone companies said goodbye to at least 65,000 subscribers. Last year during the first quarter, telcos managed to add 20,000 customers.

Leichtman Research Group reports that most customers are looking for stable and reliably fast internet service, and phone company DSL delivers neither. Having a speedy and dependable connection has become crucial as tens of millions of Americans work from home to avoid contracting the illness. Sharing that internet connection with kids staying home from school quickly caused a spike in upgrade orders.

“The increased level of usage was enough to convince many customers that they needed higher speeds to handle the number of simultaneous users in their home,” MoffettNathanon wrote.

Many phone companies lacking fiber were unable to deliver on upgrades, and customers that could went shopping for alternatives. At the same time, large DSL providers like Frontier Communications and Windstream have become mired in bankruptcy and have been losing residential customers for years. MoffettNathanson told its investor subscribers it was time to declare DSL effectively dead as a competing technology, with fiber service variants like U-verse and other flavors of VDSL near-dead.

“As with legacy DSL, it is increasingly clear that this segment is simply not competitive anymore. Equilibrium market share in this cohort, if one looks out far enough, is 100/0.”

MoffettNathanson expects cable operators will achieve an 85% market share for broadband service in markets where their chief competitor is a phone company yet to provide fiber-to-the-home service. If phone companies do not embark on immediate fiber upgrades, the damage to their market share could be permanent, especially after DOCSIS 4 arrives, according to the researchers, because the newest cable broadband platform may be able to erase fiber’s speed advantage.

Frontier Communications is seeking to slow or block rural broadband funding for tens of thousands of rural Americans that live inside Frontier service areas but cannot subscribe to broadband service because the company does not offer it.

Frontier is currently embroiled in a controversy over its regulatory filings with the FCC that sought to block or delay public funding of competing broadband projects in its territories by claiming such funding might be unfair and redundant, since Frontier is already supplying (or will supply) service in those communities.

The FCC’s Rural Digital Opportunity Fund (RDOF) will eventually spend $20.4 billion on rural broadband expansion, but the Commission is bending over backwards to protect incumbent cable, phone, and wireless companies from possible competition that potentially could be funded with public money. The Commission has invited providers to cross check “census blocks” — small geographic areas it has identified as eligible for rural broadband funding and report back if any should be excluded from the first phase of the program.

Incumbent phone and cable companies can protect their service areas from interlopers by claiming broadband service already exists in areas designated for rural broadband funding. Since the FCC continues to depend on voluntary disclosures of service areas by cable and phone companies, there is no immediate consequence if those providers take a more favorable view of what constitutes “service,” even if it ultimately results in long, further delays in rural broadband coverage for tens of thousands of Americans.

In early April, Frontier filed a lengthy submission objecting to the inclusion of 16,987 census blocks where it claims it already provides suitable broadband service. The majority of RDOF funding — $16 billion of the available $20.4 billion will be spent in the first funding phase, and only on census blocks where no provider offers high speed internet. If Frontier gets the FCC to block potential new entrants from qualifying for Phase One funding, it could spare the company from facing competition and leave a lot of homes with no internet service for years.

Immediate concerns were raised with the FCC regarding Frontier’s filing, including independent research that suggested Frontier was not being entirely honest about providing broadband at speeds at or exceeding 25 Mbps.

NTCA-The Rural Broadband Association:

Simply put, as the Wireless Internet Service Providers Association and the National Rural Electric Cooperative Association noted, “it is difficult to believe that Frontier was able to provide voice and 25/3 Mbps service in each of these 16,000 census blocks in just eight months.” Such incredulity is compounded by the fact that Frontier operated under the specter of a looming bankruptcy during this period, making it difficult to envision deployment to such a large number of locations within just several months’ time after years with little meaningful progress. Indeed, as WISPA and NRECA correctly point out, Frontier just four months ago alerted the Commission to the likelihood that it would be unable to meet its interim deployment milestones to which it was beholden pursuant to broadband commitments made in 2015. Moreover, Frontier’s financial disclosures, again as WIPSA and NRECA reference, showed an operator losing subscribers and working with a financial structure that would appear to have severely limited its ability to invest capital in broadband deployment.

The Institute for Local Self-Reliance also shared its concerns with the FCC, reminding the agency of Frontier’s lengthy track record of misrepresenting its service performance:

Frontier’s record in recent years offers numerous warning flags that the Commission should consider before accepting its nearly 17,000 challenges. The company has been the subject of numerous official complaints and investigations in the states in which it operates and has settled investigations in several states after extremely lengthy records were compiled showing its inability to regularly provide basic services. Consider this nonexhaustive list in just recent years:

California
CPUC investigating Frontier outages after transfer from Verizon in 2016 (2020)
Connecticut
AG and Dept. of Consumer Protection investigating Frontier for bad quality and billing (2019)
Florida
AG sent letter to Frontier after hearing complaints after transfer from Verizon (2016) and collected complaints (2016)
Ohio
PUC filed complaint that Frontier didn’t maintain service quality (2019)
Minnesota
PUC organized public hearings (2018) and settled with Commerce Department (2019)
Commerce launched a second investigation into billing and customer service (2019)
New York
PSC requested review after complaints of poor quality and outages (2019)
Nevada
Cited by AG’s Bureau of Consumer protection for misrepresenting speeds and service quality (2019)
North Carolina
AG issued civil investigative demand (2019)
Pennsylvania
AG Bureau of Consumer Protection settled with Frontier after investigation into poor quality and speeds (2020)
Utah
PSC investigated telephone outages (2019)
West Virginia
Settlement with AG for misrepresenting speeds (2015)
PSC ordered independent audit after complaints of poor quality and outages (2018)

The Commission faces a crisis of credibility on matters of broadband and telecommunications data collection, with two significant scandals in just the past 6 months.

Frontier’s claimed DSL speeds compared with actual average speeds (Courtesy: Smith Bagley, Inc.)

One company, Smith Bagley, Inc., went even further, building a spreadsheet of several disputed census blocks in an independent investigation. The company called Frontier repeatedly, posing as potential new broadband customers to test Frontier’s claims it supplied 25/3 Mbps service in several rural census blocks in New Mexico and Arizona. It found no instance where Frontier was ready to sell 25 Mbps service to any of the locations requested.

“Frontier either does not offer broadband service, or offers service at below 25/3 Mbps, in every one of the 1,300 census blocks it challenged in Arizona and New Mexico,” Smith Bagely noted in its letter to the FCC, citing another third party broadband availability database.

In a haughty response to the FCC dated May 26, Frontier waved off the criticism, claiming it was based on “a scattershot challenge to one-off census blocks, ad hominem attacks, and irrelevant sources.”

But the company also made a crucial admission about the broadband speeds it claims to offer that is worthy of a closer look:

“Frontier does not claim it serves every location in each census block at 25/3 Mbps. Under the Commission’s rules, carriers report the fastest speed available for sale in that census block, even if it is only available in one or a handful of locations.”

In other words, if Frontier found in its own internal testing, unverified by an independent third party, that it managed to provide 25 Mbps to even one out of hundreds of households, it can ignore the rest of the area’s much slower DSL speeds and petition the FCC to exclude funding for a new, more capable service provider. In fact, it need not disclose the abysmal speeds other homes might be enduring from Frontier and declare that census block to be adequately served by broadband and unworthy of additional funding, at least during Phase One.

Frontier also suddenly announced on May 23 it was now open to RDOF Phase One funding in its contested census blocks, which appeared to be a significant concession. But the company also noted the FCC’s established rules are the rules, regardless of what Frontier thinks:

“But to the extent the Commission decides to maintain its decision to include partially served census blocks in RDOF Phase II, SBI, Frontier, and any other company will be able to bid on those locations after mapping is complete and Phase II is implemented.”

Rosenworcel

The end impact of that could be a concession without any meaningful change.

Frontier also asked the FCC to dismiss concerned public interest groups and consumer complaints about its service because they are anecdotal. Besides, Frontier argued, companies seeking to enter Frontier-served areas can always apply for Phase Two funding, which will only be a small fraction of the funding available in Phase One. Because Phase Two is designed to help providers pay for improving existing service, sizeable portions of that funding will likely be awarded to companies like Frontier.

That the FCC plans to spend billions on broadband improvements based on flawed broadband availability data and imprecise census block criteria has infuriated Democratic FCC Commissioner Jessica Rosenworcel.

“Time and again this agency has acknowledged the grave limitations of the data we collect to assess broadband deployment. If a service provider claims that they serve a single customer in a census block, our existing data practices assume that there is service throughout the census block. This is not right. It means the claim in this report that there are only 21 million people in the United States without broadband is fundamentally flawed. Consider that another recent analysis concluded that as many as 162 million people across the country do not use internet service at broadband speeds,” Rosenworcel said in 2019. “Adding insult to injury, the same flawed data we rely on here is used to populate FCC broadband maps. For those keeping track, one cabinet official has described those maps as ‘fake news’ and one Senator has suggested they be shredded and thrown into a lake.”

This year, her Democratic colleague Commissioner Geoffrey Starks added his own concerns.

“I have zero tolerance for continuing to spend precious universal service funds based on bad data,” Starks said. “There is bipartisan—and nearly universal—agreement that our existing broadband deployment data contains fundamental flaws. And yet today’s order presses ahead with funding decisions based on mapping data that doesn’t reflect reality.”

Even with the threat of COVID-19 and a virtual nationwide work-from-home initiative, the new owners of Frontier Communications’ network in Washington, Oregon, Montana and Idaho are moving rapidly to repair persistent network issues, create a backup network, and lay the foundation to bring fiber to the home service to 85% of its customers over the next three years.

Ziply Fiber of Kirkland, Wash., formerly known as Northwest Fiber, acquired the Frontier Communications service areas in the Pacific Northwest just as Frontier itself was on the verge of declaring bankruptcy. It will waste little time upgrading Frontier’s copper wire network to get fiber service to customers fast.

“After Frontier bought Verizon’s landlines and FiOS networks in Washington and Oregon in 2010, it felt like the last decade was a phone company driving in neutral,” said Dale Prescott, a FiOS customer in Washington State. “You could feel Frontier never wanted to spend any money out here. It was like they were a caretaker of Verizon’s network, and while we got some service improvements here and there, Frontier also took away a lot too.”

Service reliability suffered, especially in areas that remained served by copper over the last decade. Customers reported lengthy outages and waiting times for repairs, and DSL speeds were actually reduced in some areas because deteriorating network infrastructure could no longer support earlier, faster speeds. In a decade of service, Frontier only managed to provide fiber connections to about 33% of its customers, the vast majority of it acquired from Verizon.

“Frontier never invested much in its network, and what it did invest seemed mostly to keep the lines from falling off the poles,” Prescott said. “Businesses got slightly better service when Frontier boosted its fiber capacity, primarily to serve commercial customers. But if you lived in the sticks, your service got worse over time, not better.”

Ziply Fiber plans to change that experience with a promise to regulators to spend about $500 million overhauling Frontier’s network in the region. Most of that spending will be devoted to upgrading customers to fiber optics. Just a few weeks after closing on its acquisition of Frontier landlines, Ziply told residents in 13 communities to expect fiber upgrades that began this spring. The majority long suffered with Frontier DSL, often at speeds as low as 3 Mbps.

Among the first towns to get fiber service are Kellogg, Moscow, and Coeur d’Alene — all in Idaho. Work has already commenced and is expected to be finished by fall. Ziply wants to keep construction costs as low as possible, so it intends to do aerial deployment of fiber by wrapping the optical cable around existing copper wire telephone cables already on the pole. This process, known as “overlashing” will simplify installation by not requiring additional space to place fiber cables next to existing telephone wiring or going to the effort of removing the existing copper wiring, which raises costs.

Overlashing has met with some controversy, however. Telephone companies are strongly in favor of allowing the process for optical fiber installation because they rarely need permission or costly permits from utility pole owners, often electric utilities. Opposition comes primarily from some electric companies, which claim overlashing can make existing installations “unsafe” by placing too much weight on existing wiring, which may have been installed decades earlier. Those electric utilities also stand to make money from forcing companies to seek new permits for placing fiber on poles, and that permission does not come free of charge.

Fiber customers will be able to select internet plans up to 1,000 Mbps. Enhanced DSL service in some areas is available at speeds up to 115 Mbps, but most of these service areas will probably be served by fiber to the home service, eventually.

Ziply Fiber Upgrade Projects (May, 2020)

• Washington—Anacortes, Kennewick, Pullman, Richland and Snohomish
• Oregon—Coquille, Coos Bay, La Grande, North Bend
• Idaho—Coeur d’Alene, Kellogg, Moscow
• Montana—Libby

To further speed fiber upgrades, Ziply acquired Wholesail Networks, already contracted to manage fiber network design for Ziply. Company officials quickly identified multiple weak spots in Frontier’s network, particularly relating to its resiliency when fiber cables were cut or copper wiring was stolen. Ziply is building in network redundancy, with each portion of its network served by at least two sets of fiber cabling and identical equipment in each of more than 130 central switching offices. In many markets, Ziply will maintain at least three redundant fiber connections to make certain if one (or two) networks go down, customers can still be served by a third with no interruption in service.

Ziply is also avoiding the usual nightmares customers experience when switching between one company’s systems to another. Frontier’s customers suffered significantly from a cutover from Verizon’s operations and billing systems, which often left them disconnected or mis-billed. To prevent that from happening again, Ziply literally cloned Frontier’s existing back office systems, so customers won’t experience any “cutover” problems.

Ziply executives have been candid about the network they are acquiring. They told regulators the network was in reasonably good condition in some places, but not all. Ziply promised to fix the network weak spots, resolve customer repair orders at least two-thirds faster than Frontier did, and make comparatively broader investments in network operations. Analysts predict Ziply has a better chance of success than Frontier did, primarily because Frontier’s operations were mired in debt, making new investment in network upkeep and upgrades difficult.

Spectrum internet customers in parts of Central Florida and South Texas are getting twice the download speed they used to receive thanks to a series of quiet service upgrades still in progress.

Customers in parts of suburban Orlando, including Seminole County, first noticed the speed upgrade in April in towns like Lake Mary. Parts of Kissimmee saw a service upgrade earlier this month. Some neighborhoods in Orlando also began reporting speed upgrades as of mid-May. Some parts of Pasco County, north of Tampa, also received a 200 Mbps upgrade, particularly in planned communities.

Charter Communications is gradually upgrading capacity in the area, formerly served by Bright House Networks. Spectrum traditionally does not announce speed upgrades until an entire service area is complete, which will likely happen in parts of Florida and Texas by early this summer.

In South Texas, San Benito is one of the communities between Brownsville and McAllen seeing Spectrum’s usual download speed doubled from 100 to 200 Mbps.

The speed upgrades come without any additional charges and usually appear automatically. Spectrum has been slowly upgrading its national service footprint to offer the new, higher-speed 200 Mbps Standard service tier. For more than two years, customers in many AT&T landline areas in the midwest and south have had 200/10 Mbps service, designed to help keep the cable company competitive with AT&T’s fiber offering. But service remains stubbornly fixed at 100/10 Mbps in just under half of Spectrum’s service area, particularly in the Northeast, mid-Atlantic, and West regions.

Stop the Cap! expects Spectrum to upgrade all of its service areas to provide 200/10 Mbps service. It remains uncertain exactly when that will happen, however.