The internet is fucked (again) – The Verge

FCC chairman Ajit Pai is fond of saying that “the internet was not broken in 2015” when he argues for repeal of our nation’s net neutrality rules. This is particularly funny to me, because in 2014 I literally wrote an article called “The internet is fucked.”

Why was it fucked? Because the free and open internet was in danger of becoming tightly controlled by giant telecom corporations that were already doing things like blocking apps and services from phones and excusing their own services from data caps. Because the lack of competition in the internet access market let these companies act like predatory monopolies. And because our government lacked the will or clarity to just say what everyone already knows: internet access is a utility.

Most of these things are still true, even after the Obama-era FCC under Chairman Tom Wheeler reclassified internet access as a Title II telecommunications service and imposed strict net neutrality rules on wired and wireless internet providers. And most of these things will get even worse when Pai pushes through his plan to rescind Title II and those rules, despite widespread public outcry.

The lack of competition in the broadband access market is so acute that it doesn’t matter if Comcast is still the most-hated company in America, or that Spectrum (formerly Time Warner Cable) has the worst customer service: you don’t have a choice, so you just have to pay them anyway. Consumers and tech publications can review and argue and debate the merits of products from Apple, Google, and Microsoft, but you just have to take what you get from your ISP.

“I don’t think [reviewing ISPs] matters, because I don’t think anyone has any inkling that an ISP can be good,” my former colleague and current Wired tech reviewer David Pierce tells me. “It’s like reading Yelp reviews of Comcast — the best-case scenario is that you don’t think about it, and it only gets worse.”

And even when you do have a choice, the cost of switching is so high it doesn’t matter. “I could switch from Comcast to AT&T, but I’d have to get a whole new box and set it up. It just seems like too much,” says Pierce. “And it would even save me money! I know this!”

If Ajit Pai and the other critics of net neutrality were out there promising that rolling back Title II would somehow result in every American having 20 choices of ISPs all engaged in vigorous competition, I’d be cheering them on. We’ve already seen what a tiny amount of competition can do in the wireless market: T-Mobile wandered up to the line of violating net neutrality by zero-rating various services, Verizon and AT&T responded in kind, and eventually the arms race ended up in all four major wireless providers offering unlimited data plans — effectively zero-rating everything and offering consumers something that looks an awful lot like net neutrality. It turns out the American people want net neutrality, and when they speak through the market, they get it.

That’s with just four major competitors in the wireless market, and Pai refuses to say whether he thinks that number should remain at four or get smaller by some combination of T-Mobile and Sprint. Of course. Because his north star isn’t healthy consumer outcomes, it’s healthy corporate outcomes. Rolling back Title II is a massive corporate handout that will line the pockets of Comcast and AT&T, while doing nothing for the average American.

But you know what? It’s really simple to just say true things about the state of the broadband market in the United States and ask how Pai would make them better. Just as we did in 2014, we can skip right past the overheated rhetoric of lobbyists and politicians and simply say things that are true. Then we can demand to know why our government seems uninterested in fixing problems — and worse, gleeful that they are ignoring the constituents they are meant to serve.

This isn’t about policy. It’s about whether or not you’re getting what you paid for. And it really isn’t that hard. We just have to start with the truth. Let’s go.


THE AMERICAN BROADBAND MARKET IS NOT COMPETITIVE

This is just a fact: the United States has a stunningly uncompetitive market for wired internet access. Fifty-one percent of Americans only have one choice of broadband provider, according to the FCC’s own 2016 data. Thirty-eight percent of Americans only have two choices. Add it up, and 89 percent of Americans have but one or two options for broadband, and one of them is often much slower than the other. This is not a situation ripe for fierce competition and lower prices.

In fact, the lack of broadband competition means that Americans pay more for slower internet access than in most other developed nations. We are not in the top 10 when it comes to average speeds. We are not in the top 10 when it comes to lowest prices. You can argue that the geography of the United States is such that covering the entire country with fiber is difficult, or that Europeans actually pay more because of VAT. At the end of the day, however, the question is simple: why don’t people in New York, San Francisco, and Chicago have more options for faster, cheaper internet than Seoul or London?

Outside big cities, the situation is dire: The Wall Street Journal just ran a lengthy, excellent piece about the many rural Americans who still only have dial-up internet access, locking them out of the modern economy. “Rural broadband, we need that quite honestly more than we need roads and bridges in many of the counties I represent,” said Austin Scott, a Republican congressman from Georgia.

The situation with wireless is similar. American wireless deployment is far ahead of other countries. Our LTE rollout was much faster than in Europe, for example, but we still pay more than other countries, find ourselves locked into more onerous contracts, and lack the ability to easily switch carriers. The furious pace of competition in mobile over the past few years has been almost entirely because T-Mobile refuses to act like a traditional American carrier, and its “Uncarrier” strategy looks like a lot like the everyday practices of European mobile companies.

While we’re spending more and more time on our mobile devices, it’s simply not yet the case that mobile broadband is competitive with wired internet access. Mobile is vastly more expensive, continues to be less reliable, and comes loaded with all sorts of strange restrictions based on opaque backroom deals. You paid DirecTV $300 to stream NFL Sunday Ticket? It’ll work over Wi-Fi, but Verizon pays the NFL to block that service on every mobile network but its own. You signed up for a T-Mobile unlimited plan? Well, you only get 28GB of data before T-Mo starts throttling you. (And that’s better than AT&T or Verizon, which choke your data after 22GB.) And on and on.

If you want a strong, fast, consistent connection to the internet, you’re not looking at mobile broadband. You’re buying a wired connection, and you probably have only one or two choices. Mesh Wi-Fi routers aren’t a hot new category because people are switching to mobile; it’s because we’re loading our homes with more and more devices that need clean, fast, safe network connections. The smart home of the future doesn’t run off a MiFi.

And while there’s a lot of hype about 5G networks competing directly with wired broadband, those networks aren’t anywhere close to reality — and they will depend on fiber in the ground just like anything else. That means a lot of deals need to be made, and as ISPs increasingly become content companies as well, the nature of those deals will turn on factors far outside the basics of cheaper, faster, more reliable consumer internet service. We shouldn’t treat wired and wireless connections differently, especially as the next generation of wireless services come online. We should insist on having rules that protect the values of the open internet from the jump.


BROADBAND INVESTMENT IS DOMINATED BY A TINY HANDFUL OF CORPORATIONS

The strongest argument Pai and other net neutrality critics have offered for reversing the Title II classification is that, by some measures, overall broadband network investment has gone down since 2015. Pai needs to make this argument — and needs it to stick — because government agencies aren’t allowed to simply change rules on a whim. Government agencies can’t just change policy arbitrarily, so Pai has to offer “substantial evidence” that market conditions have changed so much that a flip back to Title I is appropriate, and network investment (or “capital expenditure”) is the evidence he’s banking on.

But Pai’s choice of metric is incredibly revealing. He’s effectively saying that US broadband policy should be dictated by AT&T. Here’s a chart from USA Today, based on this survey from economist Hal Singer:


Comcast’s $1.2 billion increase in spending is wiped out because AT&T decided to buy DirecTV and invest in Mexico instead of its network — after a years-long period of intense network investment called Project VIP. It’s now spending slightly less than it did before. AT&T also decided to buy Time Warner in the past year, another monster investment that will pull money away from network spending (if the Trump administration doesn’t block the deal to punish CNN). You can argue about the background of every number forever — Charter bought Time Warner Cable! Sprint changed major accounting practices! — but it ultimately doesn’t matter what anyone else did. AT&T is so big that when you add up the investments of 12 major ISPs, all but $200 million of the decline is due to AT&T.

Making policy against this metric alone creates obvious incentives for huge internet companies — which are already de facto monopolies for many Americans — to withhold network investment until they get the regulatory changes they want. Put more simply: the FCC is saying we should be held ransom by AT&T. That is a dangerous position for an agency that ostensibly protects the public interest to find itself in — yet it’s exactly where Pai seems headed. Even though the ISPs themselves consistently tell their shareholders Title II hasn’t really changed their investment plans.

There are other obvious metrics Pai could use to measure the state of broadband in America: the number of ISPs providing equivalent services to various American cities, wired and wireless. You could measure the average cost of broadband services (which are flat to up) or the average cost per megabit of data (which is going down). You could measure broadband adoption. Hell, you could just measure the average speeds and data caps available to citizens, and their relative cost. If your only choices are expensive, capped mobile data and a slow wired connection, do you really have a choice at all?

Pai isn’t bringing those measures up, because all of them indicate the same thing: our broadband providers are barely competing, and instead milking more and more money out of their existing customers. This has been true for years, before and after net neutrality.

By reversing net neutrality and giving internet providers the ability to sell fast lanes, Pai will allow those providers to make more money from their existing networks — money he’s presumably hoping they will then spend on infrastructure investment. It makes a certain kind of sense, but absent any real competition, it’s hard to see why these companies or their shareholders will want to do anything except collect profits.

Call it trickle-down broadband: Pai’s arguing that making rich ISPs richer will somehow benefit everyone. Of course, just like trickle-down economics, the mechanism by which this will actually happen is completely unclear. And until Pai can articulate a specific set of causes and effects that will make his ideas work the way he says they will, the burden lies on him to prove it out.


INTERNET SERVICE PROVIDERS ARE NOT THE INTERNET

If I were to ask you if the government should regulate the internet, it’s pretty likely that you would say no. I know this because the cable industry’s lobbying group, NCTA, asked a bunch of people that question last month and 78 percent of them said no.

But 61 percent of those people said they supported net neutrality rules like no throttling and no paid prioritization.

Why? Because what NCTA did not ask — what the internet access industry is terrified of asking — is if people think Comcast and Google and Verizon and Twitter and AT&T and Reddit should all be regulated the same way.

Because while it’s useful for the ISPs to frame all those companies as “the internet,” it’s obviously not correct. Americans value — even love — the content and services they get on the internet, and do not want them meddled with. They do not, however, love their internet service providers. The conflation of internet access with the internet is dangerous and misleading, and it is a fundamental feature of the ISP lobbying platform.

This is a point that tech reviewers make repeatedly. I made it in 2014. My colleague Walt Mossberg has been making it since 2007, when he referred to wireless carriers as “Soviet Ministries.” In 2009 David Pogue caught Verizon inflating bills by silently charging users a tiny fee if they pressed the wrong button on their phones. Real tech companies are often aggressive to the point of self-destruction but few of them would venture that far.

Why? Because if Apple pulled that stunt, people could switch to Google or Microsoft. If Amazon starts cheating you, you can buy stuff from Walmart. If you hate Uber, you can take a Lyft. And on and on. The idea of tech companies getting too big to feel competitive pressure is an obvious problem to those of us who routinely cover the industry. Mossberg warned against growing oligopoly in his final column, while The New York Times’ Farhad Manjoo is writing a book called The Frightful Five that considers the growing influence of Apple, Google, Microsoft, Facebook, and Amazon.

But none of these companies would even be in the position to be frightful if not for open and unrestricted access to the internet. They have had to compete in a wide-open marketplace and they’ve each found so much success that we’re beginning to grapple with the consequences.

Why would we allow broadband providers in an uncompetitive market the ability to act as the gatekeepers to this much economic activity? Why would we threaten the market conditions that have allowed our most innovative and valuable companies to thrive? Why are we even having this debate, but for the fact that Verizon and AT&T feel left out?


SUCCESSFUL INTERNET COMPANIES EVENTUALLY STOP CARING ABOUT THE OPEN INTERNET

In 2014, the Netflix team was an aggressive champion of net neutrality. They put out charts, they paid lobbyists, they told people their videos would buffer endlessly, they had PR people call reporters day and night — in short, Netflix waged war.

A couple weeks ago, at the Code Conference, Netflix CEO Reed Hastings somewhat casually said that Netflix wasn’t paying attention this time around, and that smaller companies could carry the water.

This is sad — Netflix only exists because of net neutrality — but not surprising. As companies get bigger, their ability to demand access and special treatment grows accordingly and becomes a competitive advantage. “We’re big enough to get the deals we want,” said Hastings at Code. If you’re Netflix, you’re happy to cut a deal with Verizon to stream video outside the data cap — it means more people will watch House of Cards than I Love Dick. Every other major CEO at Code that talked about net neutrality struck a similar tone — this fight doesn’t matter if you’re big enough to get the deals. The brilliant analyst Ben Thompson has written about this extensively — a world without net neutrality is in Netflix’s best interests, because it insulates it from competition.

But this is miserable for smaller companies that wish to compete, which is the entire reason net neutrality exists. My colleague Jake Kastrenakes charted Netflix’s public positions on net neutrality against its subscriber numbers, and found a direct correlation between its decreasing commitment to an open internet and its size.

Netflix is now beginning to react to the backlash against Hastings’ statements and is participating in the day of action, but the simple fact is that once you’ve built an economy where giants are cutting deals to get ahead, the ability for smaller companies to disrupt them fades fast. And if every new company has to pay to play, you’re sucking dollars away from innovation and using them to line ISP pockets.

The same is true of Apple, Google, Facebook, and every other large beloved tech giant. Their interests are not the same as your interests, or the interests of smaller potential competitors. They advocate for open access when it aligns with the needs of their various platforms, but when their interests demand it, they rationally run in the other direction. That legislation to strip privacy protections from ISPs that most people hated? Facebook and Google lobbied for it in the background, because they don’t want comprehensive privacy rules that apply to them as well.

Companies exist to make money for their shareholders, usually by providing value to their customers. Asking any company to idealistically act against those interests is risky, although sometimes it works. But trying to create a national broadband policy based on that risk is just stupid.


AJIT PAI HAS ALREADY MADE UP HIS MIND

Chairman Pai and the FCC can’t simply erase the net neutrality rules on a whim. They have to provide substantial evidence that the rules need to be changed to survive an inevitable court challenge, and they have to show that their minds aren’t already made up. But it’s clear that Pai’s mind is already made up, and has been since he voted against the rules in 2014.

Look, it’s not entirely clear that Donald Trump knows what net neutrality is. In a 2014 tweet, he described it as the “fairness doctrine” for the internet, which is wrong in the amusing way that Trump is routinely wrong, and I can’t find another instance of him seriously engaging the topic again. There are many reasons people voted for Trump, but I sincerely doubt that reversing Title II classification was among them.

It’s also true that in November 2016 most American ISPs were talking to their investors as though Hillary Clinton would win, and that Title II net neutrality would continue to be the law of the land. And they were entirely fine with this: most of them were saying that broadband investment and usage would continue and even accelerate.

But Pai and his team have taken Trump’s election as a mandate to make sweeping changes and erase rules that protect consumers from companies that have no competition. Granted an immense amount of power they weren’t expecting, they evince no humility or maturity, just a palpable sense of vengeance. A few weeks ago, the FCC’s security guards attacked a veteran member of the tech policy press — not exactly a rowdy bunch — when he attempted to ask the commissioners questions about Title II rollback at the recent vote. The right-wing tech policy think tanks they rely on routinely insist that all pro-consumer net neutrality coverage is radicalized misinformation paid for by enemies of the state — a position Pai himself has taken in speeches.

Pai and his team have learned important lessons from Trump, and it wasn’t to become exciting populists who fight for the people. They’ve learned that wrapping their ideas in aggression precludes engaging them on substance.

Let me make a distinction: The Verge has harshly, severely criticized companies like AT&T and Comcast in the past. We published an entire series called The Worst Company in America while Comcast was mulling a deal to invest $200 million into our parent company. (They did, so here’s the disclosure: Comcast’s NBCU division is an investor in Vox Media, our parent company.) We have ripped AT&T to shreds over its sponsored data package. Verizon’s go90 video service is just a running joke in our coverage, and we have referred to its new Oath unit as a “nightmare.” But when we talk to the people running these companies, developing their products, and running their policy operations, they are all kind and sincere and determined to serve their customers better.

“We don’t want to run the most-hated company in America,” Comcast CEO Brian Roberts said to me at the Code Conference in 2015. I spoke to several ISP executives at Code this year and they all told me that they believe in open internet protections. And you know what? I believe them. But until Roberts’ customers have a real choice of broadband providers, it doesn’t actually matter. And that is a genuinely interesting debate to have with the people who run policy shops across the industry.

It’s the same for virtually every other ISP. We ask hard questions about what our readers — their customers — should expect for their money, and they’re ready to engage in the answers. If these companies were made to compete in a real market, I would bet anything that they’d find ways to win the love and loyalty of their customers. Some of them, like T-Mobile, are competing in a severely limited market and winning that love anyway.

Who are the FCC’s customers? Who is Ajit Pai waking up every morning and trying to please? The FCC relies on a number of public-interest standards to enforce its authority, but every time Pai speaks he addresses corporate interests, not consumer or citizen interests. The American people have repeatedly, loudly said that they feel taken advantage of by their internet providers, and Pai and his team have responded by asking whether ISPs need any rules at all. And his team has repeatedly signaled that they don’t really give a shit anyway.

“What matters most are the quality of the comments, not the quantity,” Pai has said. Well, it turns out that Verizon’s lawyers submit much more detailed telecom policy comments than the everyday Americans who are outraged by this decision. Meanwhile, the FCC says its website went down during the public comment period because it was “attacked,” but the agency refuses to answer any real questions about that attack, where it came from, or the effect it might have had on public participation.

Senior FCC officials have repeated Pai’s line during press calls, adding that any decision “has to be based on reasoned decision making and applying the facts.” But it’s abundantly clear that Pai and his team have already decided what the facts are. I mean, what else does Pai intend to convey when he says “make no mistake about it, this is a fight that we intend to wage, and this is a fight that we are going to win” about reversing Title II? He’s already made this decision. All that’s left is to go through the empty motions of government process.

It should not be surprising that Pai and his team haven’t announced any plans to fix the loss of public comment caused by this “attack.” Like Trump, they are ignoring the opportunity to make real and lasting policy in the best interests of everyone they serve in order to seize the opportunity to attack the credibility of their critics and stoke their base. With Trump, that base is largely disaffected citizens who feel ignored. With Trump’s FCC, that base is massive corporations.


So there it is. We live in an uncompetitive broadband market. That market is dominated by a handful of giant corporations that are being given the keys to shape telecom policy. The big internet companies that might challenge them are doing it half-heartedly. And Ajit Pai seems determined to offer up a massive corporate handout without listening to everyday Americans.

Is this what you want? Does this sound like a path toward better, faster, cheaper internet access? Toward better products and services in a more competitive market? To me, it sounds like Americans need to demand that our government actually hear our concerns, look at our skyrocketing bills, and make real policy that respects us, instead of watching the staff of an unelected official laugh as he ignores us. It sounds like we need to flood the offices of the FCC and Congress with calls and paperwork, demanding to know how giving handouts to huge corporations will help us.

And it sounds like Ajit Pai needs to explain why he’s not fucking up the internet all over again.

Comments

Write a Reply or Comment:

Your email address will not be published.*