Every few months it’ll send me a postcard. Once, in the midst of the pandemic, the company even sent salespeople to my door.
AT&T wants to sell me 1,000Mbps fiber-optic internet, and I’m beyond ready to pay for it. In fact, I’ve already signed the contract at least nine times now.
But like tens of millions of my fellow citizens across the United States, I don’t actually have the ability to switch. Across the country, the FCC and internet service providers are pretending there’s competition in an unimaginable number of places where it doesn’t actually exist. It’s a nationwide problem, but one that I’m experiencing even in the Silicon Valley — where you’d think the internet might be better.
Today, I pay $100 a month for Comcast internet, and the cable company’s speeds leave something to be desired. My wife, a video producer, can’t reliably upload her company’s videos to its FTP servers. The family videos we shoot on our smartphones take hours to appear in Google Photos, and there’s no point in me even trying to stream games to YouTube or Twitch. While my download speed is decent at 200Mbps, the uploads are absolutely atrocious at just 5Mbps up, and I’d have to pay Comcast even more to change that. And my bill has already gone up: I paid $70 until the pandemic hit and we started running into Comcast’s completely arbitrary 1.2TB data cap. Now, I have to pay extra for unlimited so I don’t rack up overage fees every single month. (Disclosure: Comcast and NBCUniversal are investors in Vox Media, parent company of The Verge)
AT&T seems like the obvious answer. The company’s fiber network offers 1,000Mbps downloads and 1,000Mbps uploads for $70 a month, no data cap whatsoever. Competition is good, eh? The only problem is that AT&T doesn’t actually serve my address... even though it continues to insist it does. The company’s field technicians have visited my home easily a dozen times now, but their conclusion is always the same: AT&T buried its fiber in the wrong place.
A few houses down from me, there’s a concrete cover embedded in the pavement. If you look underneath, you’ll find a fiber-optic node labeled with my exact street address. It’s supposed to be in front of my home; AT&T’s service maps say it’s in front of my home, but those maps are wrong.
The more grizzled AT&T vets warned me I’d get no satisfaction from AT&T. They told me it was illegal to run wires over my particular street — and that there was little to no chance AT&T would spend $10,000+ to dig up the sidewalk just to serve more houses on my side of the road.
Some friendly techs filed dig tickets anyhow! I was told at least twice it was actually going to happen. (AT&T sales reps told me it had already happened according to their records.) Once, I even got an actual notice from AT&T that my order had been canceled. But every other time, I’d call weeks later to find out my order had simply, quietly vanished.
If the company wasn’t going to give me internet, why couldn’t it remove me from its marketing lists? Why couldn’t it simply update its maps?
That last question, it turns out, is far bigger than I imagined. It may help explain why America’s internet is so broken to begin with, and get us on the path to fixing it.
As FCC Commissioner Jessica Rosenworcel wrote for The Verge last March, as many as one in three US households doesn’t have broadband internet access, currently defined as just 25Mbps down and 3Mbps up — which feels like the bare minimum for a remote learning family these days. Even before the pandemic, that statistic might have been shocking; now, it’s the difference between whether millions of schoolchildren can attend classes and do their homework or not. Nearly 12 million children don’t have a broadband connection at home, the Senate Joint Economic Committee reported in 2017. And the “homework gap” hits harder if you’re poor, of course: only 56 percent of households with incomes under $30,000 had broadband as of last February, according to the Pew Research Center.
But even if they could afford broadband, there’s no guarantee they’d get it — because the FCC has spent the past decade trying to paint a rosy picture of America’s broadband instead of figuring out where it actually exists.
In 2009, the US spent $350 million on a “National Broadband Map” that turned out to be nothing more than a chance for the wolves to guard the hens: it relied on ISPs like Comcast and AT&T to submit their own data, which the FCC does not audit. According to the FCC’s map, I have 11 different broadband providers at my address! But if you break it down:
- Two of them are “fixed wireless” that cater to businesses, not homes, one of which starts at $99 a month for 3Mbps (not a typo)
- Two of them are slow, data-capped satellite internet
- Four of them are duplicates of the same providers (including three different tiers of AT&T DSL, two of which don't meet the FCC's definition of broadband)
- Two of them don’t offer service at my address at all
That leaves me with just Comcast — and you with the reason why the United States and its ISP lobbyists get to pretend they’re doing a decent job, even when there’s no meaningful competition. About 88.3 million Americans only have a single choice of broadband provider, according to the Institute for Local Self-Reliance, because their other choice is DSL that doesn’t even meet the FCC’s own definition of broadband at 25Mbps down / 3Mbps up. That’s effectively a monopoly, and the report claims 22 million of them are dependent on Comcast alone. And those are *low* estimates, by the way, because they’re based on the same faulty, self-reported data as the FCC’s terrible maps.
When Comcast knows you have no other alternative, it can get away with things you’d never see anywhere else. Exhibit A: the 1.2TB home internet data caps it’s about to erect in a dozen additional states next year.
Forget for a moment that Comcast itself was caught red-handed explaining that data caps have nothing to do with network congestion. Forget that the CEOs of several smaller ISPs have admitted that internet capacity is anything but scarce. Forget that Comcast disabled its own congestion management system because it found it was unnecessary. Forget even that Comcast is a wildly profitable company whose cable division spends only a tenth of its yearly revenues on keeping that network strong. The proof that data caps are a swindle is something you probably witnessed yourself earlier this year: Comcast, AT&T, and T-Mobile all suspended their data caps when the pandemic hit, and the internet kept on working without a hitch.
But Comcast’s generosity didn’t last for long. Now, with cases and deaths repeatedly hitting record highs in the United States, tens of millions of Americans out of work, and many of them thousands of dollars behind on rent, the ISP is so unconcerned about angry customers that it’s bringing those data caps to even more states. AT&T, to its credit, extended its initial data cap waiver through the end of the year — but that wasn’t enough to pressure Comcast to do the same. Comcast’s waiver expired on July 1st, alongside most other ISPs, and now it’s back to extracting money hand over fist.
Exhibit B: Though the United States once led development of the world’s internet, we now pay more than the rest of the developed world on average.
We pay twice as much as Europe for high-speed internet — before you even count caps and fees
We consistently pay more than Europe regardless of speed, according to a fascinating, approachable study you should read from the New America think tank. In fact, we pay roughly double that of Europe at the 100Mbps and 1,000Mbps tiers, and eight to 17 times more to rent a modem on average than Asia and Europe do, respectively. Only one US city cracked the top ten in affordability but only because it had an ace up its sleeve: a municipal fiber-optic network erected by the city itself, where ISPs provide their services across fiber that the residents themselves own. Those sorts of municipal networks create competition that simply doesn’t exist in many places in the US because it wasn’t designed to exist. In places that do erect municipal networks, New America shows that both speed and affordability far outpace the rest of the US.
That’s why it’s a real shame many states (and telecom lobbyists) have erected roadblocks to keep those municipal networks from spreading. In one particularly egregious example, Comcast sued the city of Chattanooga, Tennessee to try to block its municipal network in 2015, lobbied for a law to prevent its award-winning network from expanding to rural areas, then finally convinced the state to spend $45 million of taxpayer money to subsidize the big ISPs instead.
We need competition. We need accurate maps to clearly see just how little competition there actually is, and we need to change the laws to let citizens fed up with being unserved and underserved build their own networks instead. We need those maps to show how much people actually pay. We need to stop pouring taxpayer money into hugely profitable telecoms that claim they’ll build out internet access, since they’ve found they can often just straight-up lie or wait to be sued instead of fulfilling their obligations. (If you want to be infuriated, follow that last link and keep on clicking through the examples.)
At least in 2020, the US is finally making some progress. This year, Congress finally passed a bill that requires the FCC to actually audit its broadband map data and create a “user-friendly challenge process” so people like me can point out that no, AT&T doesn’t actually offer fiber (or even fast DSL) at my address. Here's hoping it actually works, unlike the millions upon millions of net neutrality comments the FCC simply ignored. Earlier this month, the FCC also already awarded $9.2 billion to 180 different companies to bring broadband to 5.2 million rural Americans over the next ten years — though big companies like Charter, CenturyLink, Frontier, and Windstream still account for at least $2.4 billion of it.
But another way of looking at things is that we’re already too late: the FCC just awarded a decade’s worth of money to bring broadband to only five million additional addresses, without knowing whether it was awarding those contracts to communities that will now see competition — or just erecting more pockets of monopoly on the fringe. What is the rest of America supposed to do, wait another decade?
We need to go much, much further than fixing our maps. If the best we can do during the pandemic is ask ISPs to “pretty please make the internet work better for 60 days,” to borrow a phrase, it’s time for more regulation.
The internet is more important than the telephone ever was; let’s regulate it that way
The COVID-19 pandemic has shown, perhaps for the first time, how essential the internet is. The internet was already a powerful tool for learning, but now it’s a critical one for kids to attend school. Many millions are jobless but could maybe find a new job that lets them work from home. Many are already working from home because — with no other choice — companies are learning to adapt.
This all means there might finally be enough political will to declare the internet should be delivered to homes like a utility, the way water and electricity are delivered today, the way landline telephones were under the Communications Act of 1934 that originally created the FCC. The last time the law was truly overhauled was 1996 — back when the internet was delivered one digital screech at a time over standard phone lines, after all.
That’s how we’ve always dealt with necessary monopolies, the ones that own the pipes and wires that feed your house. We put them on a tight leash and make them serve everyone, not just the homes they deem most profitable.