There’s been a lot of chatter about AT&T and the state of internet connectivity in our country in light of AT&T’s recent decision to start capping bandwidth and merge with the other wireless company, T-Mobile. The short version is that Comcast tried throttling and QoS, which mostly failed due to the whole Network Neutrality movement (which, for the record, is more of an anti-ISP movement than it is a pro-Internet movement, although the two are not mutually exclusive). Since that didn’t pan out, the new plan is caps and usage-based billing, which is still a major rate hike, but is “neutral”. People ask why US internet connectivity sucks relative to, say, Estonia, which is an entirely reasonable question. The conclusion is that “we should have some regulation” to solve this problem.
Now as a card-carrying libertarian I’m supposed to stand here and tell you that the free market solves all our problems and that regulation is bad. This is true in a very technical sense. But, for one, we’re not in anything even closely resembling a free market, and for two, the right answer here involves just a little bit of legislation–no where near the type of legislation people are asking for, which will turn out very badly.
Well, the textbook libertarian answer is that the trouble, no matter what it is, started back with FDR. Under the Telecommunications Act of 1934:
AT&T was essentially granted immunity from antitrust suits with the goal of providing universal service. At the same time, it was also prohibited from telco-broadcast cross-ownership. This, along with the Willis-Graham Act, effectively gave license for AT&T to complete its acquisition of independent phone companies and ultimately build its empire of Bell operating companies (BOCs).
Not only did AT&T maintain a monopoly in voice service, but also maintained monopoly control of the long distance network and all consumer-attached telephone equipment such as phones and wires. AT&T thrived on this vertically integrated monopoly structure for decades before it became a legal and regulatory obstacle for competitors who wished to provide the non-Bell equipment to customers.
The reasoning was that providing universal telecommunications was really freaking important and we were willing to, pretty much, let AT&T do anything it wanted including just handing out federal money and making it effectively illegal for other telephone companies to enter the market.
In effect, every American was henceforth found to be entitled to the right to telephone service, specifically cheap telephone service. To carry out this difficult policy objective, the FCC was given sweeping powers. Beside its powers to regulate rates to ensure they were “just and reasonable,” the FCC was also given the power to restrict entry into the marketplace. Potential competitors were, and still are required to obtain from the FCC a “certificate of public convenience and necessity.” The intent of the licensing process was again to prevent “wasteful duplication” and “unneeded competition.” In reality, it served as a front to guard the interests of the regulated monopoly and the FCC’s social agenda.
For a really fascinating read on how AT&T’s rise to power was effectively bought and paid for by the feds, you should really read the whole article.
Now cable companies basically operate the same way, except the regulating body is typically a local municipality (not the feds) which solicits “bids” from cable companies and grants the one who pays the city the most kickbacks the cable monopoly for fifteen years at a time. Sadly, I’m not joking.
So that’s the history lesson. What can we learn from this?
First, laws last forever. Cable monopolies typically renew every 15 years. Telecommunications law has been majorly overhauled only once, after 60 years. Whatever you decide the law should be, you’d better be happy with it for the rest of your life, because that’s the way it’s going to be, forever. It would shock you to learn how much time and energy has been spent litigating to what extent Skype and friends represent a “telecommunications service” that should be regulated (codeword: excluded) on the basis of some law written 20-60 years ago. So be very, very sure that you’re not outlawing the next Skype when you write your “fix the internet” law.
The next thing to notice is that our “competitive” regulatory landscape is largely written by the people being regulated. Consider this: you’re a congressperson, drafting some legislation about bandwidth caps, hoping to create a “fair and competitive legal framework”. Now obviously we need some kind of usage-based billing, and obviously the cheaper the better. But you can’t just write down “The ISP’s going to give 12 excabytes to everyone, mwahahaha!” because that’s just not going to work, practically speaking. You need a “reasonable” number. And who better to know what’s reasonable than the ISPs? Maybe the EFF files an amicus brief (not the right term, but go with me), but what do they know, they don’t provide internet service to anyone, it’s just formulas on a page. The end result is that whatever Comcast, Verizon, and AT&T can agree to becomes the law of the land. For the next 60 years. That’s what’s going to happen if we “regulate” the ISPs.
Well, for one, we can stop granting companies monopolies over cable and phone lines. The “reason” we do this now is to make sure ISPs are “sufficiently motivated” to build out in rural areas, knowing that they are protected from competition. In all fairness, they’re nowhere near sufficiently motivated with a monopoly right now–rural areas have absolutely horrible connectivity. In our monopoly-free alternate universe, municipalities could always run their own fiber if they felt they were being underserved. As it is, they often try, and unfortunately due to longstanding monopoly awards they often cannot. So the whole line of reasoning that “monopolies are necessary” is an argument in contradiction to the evidence.
Second, we can advocate for fair advertising. “Up to” speeds are completely misleading, which is the way internet service is sold to consumers today. This is an issue of fair and just scales–you should be able to compare the speed and reliability of one ISP to another just like you can compare the active ingredient in the brand name and the generic drug, or like you can compare nutritional information in different foods. The way ISPs advertise right now is almost entirely deceptive, promising things that they don’t actually deliver. This is a contract law problem. I’ve blogged about this in the past. We need to require ISPs to advertise their services in a fair and comparable manner to improve market competition.
Unfortunately, I do not think that these two measures (alone) are sufficient, at least in the short run. In the same way that Space X is saving us from NASA, private initiatives like Google’s Fiber project might eventually save us from the ISPs. But it will take a very, very, long time. I think that Google (in the wireline space) and Apple (in the wireless space) are sufficiently motivated to stir the pot a little when it comes to delivering “reasonable” internet speeds, because that’s to some extent in their interest. But really their only goal is to do exactly enough threatening to whip the ISPs into something resembling shape, not to change the landscape all together; they don’t really want to be an ISP. And threatening to compete with ISPs works extremely well, just check out how fast Comcast’s speeds are in a FIOS area vs an area where it’s primary competition is DSL. As long as threatening alone works, they’ll continue to do it, and if Google becomes an ISP at all, they will get there very slowly.
Building on a platform can be a bad idea if the platform decides it doesn’t like you. We all know this, and we continue to build our little Facebook apps, because it’s easy and profitable. App gets pulled by Facebook, out come the whiners saying you shouldn’t have relied on Facebook. Next time use HTML5, you should have known better.
But here’s the dirty little secret: the web is a platform that could potentially be controlled just as tightly as Facebook or the App Store. I’m not talking about the magical land of HTML5 video tags and cross-platform compatibility and unicorn sightings, I’m talking about the small number of ISPs who control almost every data line in the country, I’m talking about Cisco who owns basically all the critical patents on modern-day switching equipment, I’m talking about the feds who are in bed with both companies seizing your domain names and spying on your packets. It’s time we had an adult conversation about the web as a platform. Every web app you write makes the internet more and more important, gives AT&T and the other ISPs more and more power. We’ve not only given them a monopoly, we’ve given them an appreciating monopoly (contrasted with railroads, cotton gins, or any other type of monopoly we’ve authorized). It actually bears compound interest. As organizations, they’ve barely started to work this out. But just because they’ve historically been benevolent dictators does not mean they will continue to be. In fact, if the rate hikes and announcement to acquire their biggest competitor are any gauge, the giant is waking up.
This compound interest thing is a really big problem, because things tend to spiral out of control rather quickly. I don’t believe that reducing a couple of barriers to competition are sufficient to overcome this problem, so I’m going to do something I don’t think I’ve ever done: advocate for government intervention.
You see, we have this thing called the Post Office. Its relationship with the federal government is basically:
Now a libertarian might say “some of these bullet points might be a bad idea” but it’s infinitely better than the government’s relationship today with an ISP, which usually consists of kickbacks, free money, no substantive mandate to provide service, and wide-ranging monopolies on laying any kind of cable. And, in spite of certain complaints, USPS really does a phenomenal job at shipping stuff incredibly cheaply–almost 100% of the time much more cheaply, at the consumer volume, than UPS or FedEx.
So what I’m going to advocate is we simply “PostOffice-ize the internet”. We set up a private, not-for-profit entity to compete with the ISPs and provide a basic level of internet service to the whole country. We let the entity borrow (not take) federal money to run fiber. The entity is federally-mandated to do all the things that we’re currently trying to regulate ISPs into doing (share lines, provide universal access, charge “reasonable” rates, etc.) Then we can sit back and deregulate the ISPs, since they will be pressured to compete with the Federal Internet anyway. Traditional ISPs can focus on providing faster and better speeds to populated areas rather than whining about the costs to support rural ones (which, to be frank, is what they’re doing now anyway). Instead of asking AT&T and Verizon what “reasonable” bandwidth costs are to put into the law, we can simply force them to compete with actual reasonable bandwidth costs, which we now know because we’re running an ISP.
Look, this falls squarely under the Commerce Clause. I know that’s been abused for everything from setting up AT&T’s monopoly in the first place to the War on Drugs, but if anything is interstate commerce, the Internet is. It’s 100% completely within the plain reading of that enumerated power. The only question is–are we going to have good regulation or are we going to have bad regulation? Since FDR, we’ve had bad regulation. I prefer “regulation” which creates a free market instead of regulation that destroys one, “regulation” which creates a competitor that evolves with the times instead of regulation that sits on the books for 60 years and outlaws Skype, which is what we have. If you must, set up a bunch of little state or municipality ISPs if that makes you less nauseous.
Well… if you think about it, I’m really advocating less government intervention than what we already have. Right now, today, this very minute, we basically have an Internet that, behind the cloaks and daggers, is really controlled by the government, either in the form of kickbacks, granted monopolies, favorable regulation, or other complex and nefarious vagaries. All I’m proposing is that we create an entity with even less government involvement than the ISPs of today.
And if it doesn’t pan out, because the government is incompetent? We’ll still have those ISPs to fall back on. So if the government does a terrible job, the fallout might approach the cost of a bomber or two. Oh, noes.
Yeah, the ISPs beat you to it. I mean, I don’t mean to trivialize the issue, because wiretapping is a super important problem. But Federal Internet is certainly not going to make things any worse than they already are. When we finally have the political capital to ban warrantless wiretapping we will, with or without a Federal Internet.
What’s politically infeasible is letting AT&T continue to profit from an exponentially accumulating monopoly that we granted them in perpetuity in 1936. What’s politically infeasible is failing to save the Internet, the biggest driver to economic commerce and growth in the whole US, and pretty much the only way to get our economy to a point where we can start counteracting the defecit. I know that bits and bytes aren’t as exciting as Kate Middleton or killing some bearded guy in Afghanistan but they’re infinitely more important.
It’s a couple PR campaigns away from being feasible. All you have to do is run some attack ads–the ISPs have even written them for you, just read some scary terms of AT&Ts terms of service on TV, play a Comcast technical support call gone awry. Democrats hate ISPs because of the bigCo angle, Republicans hate ISPs because of the monopoly and free market angle. It’s sellable.
And even if it’s not–all the ISPs need is just a whiff of fear and suddenly they start providing better service. Even if it’s not politically feasible, it’s still a really good idea to campaign for it.