Last week the FCC released new proposed rules for how to handle traffic on the internet. This has created a bit of a firestorm and a whole lot of debate, because the proposed rules change from “everybody gets equal access” to “you can pay for priority access”, or what is known as a fast lane. As is typical in these debates, you end up getting two sides, vehemently defending their position and accusing the other side of being big old doodie heads.
I strongly come down on the side of “net neutrality” – the idea that all traffic should be given the same priority, so I want to dispel some myths about what the other side is saying. In order to dispel these notions, I think it is helpful to look to the debates around health reform that we just went through in the United States. Much like that debate, arguments on both sides were heated, and people in the middle just weren’t sure what to think. Fortunately, just like with health reform, we have actual data that can say which side is right (spoiler alert: it’s net neutrality).
While the right wing in this country was decrying health reform as socialist, that prices would soar, that quality would drop, all you had to do was look “across the pond” to England and Europe, or over the Pacific towards Asia, and you could find data that dispelled that notion. The hellish landscape of socialized forms of medicine had already been “means tested” – i.e. implemented in places, and the results were encouraging. And thus, changing the system to be more socially oriented from purely profit oriented was not the end of the world. It was not a massive experiment in social engineering. It was, basically, blasé. It was simple. It worked. And it worked better than our system.
When you look deeper at the arguments against net neutrality and for the new FCC proposal for fast lanes, sure enough, you see the same thing. So, let’s begin.
The Myth of Free Market Innovation
First off, you have the typical Ayn Randian “free market” blathering that goes on in whenever some of us ask for government to regulate things. There is this idea that if the government is involved at all, it will somehow take over or stifle innovation. This was laughably discussed in a blog post on Re/Code, titled “The Internet is Not a Rotary Phone”. In it, the author tries to paint the growth in high speed Internet as something that was only possible through lack of government involvement, by stating that:
“Consumers are fleeing POTS (Plain Old Telephone Service) in droves. In 1996, when they had no other choice, 94 percent of American households relied on POTS as their sole means of telecommunications. Today, fewer than 5 percent do so. That is not surprising. There are things regulators do well, but innovation is not one of them.”
The problem with this statement, however, is that the author apparently doesn’t understand how much POTS innovated during its lifespan. When telephone service was first put into place, and when it became regulated as a monopoly, your phone service was a direct wire from your house to a central building where a human would make a connection for you to some other human. You may remember seeing pictures of men and women moving cables between connectors to make that connection for you.
Now, if what the author was saying was correct, there would be no reason to do anything but that. Maybe you tinker, but why? You just capture the regulators, keep your system inefficient by claiming it “creates jobs”, and sit back and collect your guaranteed return. But, we know this isn’t true. While being a regulated monopoly, the phone system changed from humans moving connections between copper wires routed directly to homes, to a computer controlled, switched network. Your phone changed from being a thing that generated sparks with a rotary dial, to a set of buttons that made tones that computers would use to create connections. Speeds also improved. The backbone of the phone system became high-speed, digital, packetized, and ran over coax, fiber optic lines, and satellites. If this hadn’t happened, you wouldn’t have been even able to create computer connections that this author then summarily derides.
Was it perfect? No – it is entirely possible that the phone network could have improved faster without regulations. But can we at least admit that the idea that government involvement “stifles” regulations is just crap? Innovation happened!
Now, to lay the groundwork for how awesome the world would be without regulation the author of this piece tries to show how wonderful the wireless technology we all use today is, and says how this is because of how the wireless industry constantly innovates:
“Nearly two-thirds of consumer Internet connections in the U.S. today are over mobile devices, up from zero in 2005. Those nifty smartphones and tablets consumers enjoy can download at speeds that average between 14 megabits per second and 19 Mbps, and can peak as high as 57 Mbps, because wireless carriers have moved through six generations of network technology in seven years.”
What the author doesn’t say, however, is where these wireless technologies were created, or where they were first deployed. The simple fact is, the US lags in such deployment. LTE (currently the fastest wireless technology) was first proposed in Japan, by DoCoMo – in 2004. It was approved in 2008, and first deployed in Stockholm in 2009. (Source: Wikipedia) It wasn’t first deployed in the United States until 2010, in small, select cities. As of today, it still isn’t very ubiquitous in the United States – we all know of at least one dead spot around our where we live, and companies throw mud at each other about how their deployment of LTE is better than the competitor’s deployment of LTE. And it is 2014, for Pete’s sake.
LTE, was a cooperation of governments and private industry, codified into an interoperable standard under spectrums licensed by governments, deployed first outside the US in heavily regulated markets, and is still not widely available in the US in 2014. How, exactly, is the “free market” innovating here?
The Myth of “We Already Prioritize Traffic”
The next bullet in the chamber of the “fast lane is good” gun is to try to confuse you about what a fast lane means. In this next piece on Re/Code, another author describes how we already have different prioritization for traffic based upon traffic class type. For example, you may have heard of VOIP (Voice over IP), which companies like Vonage sell, where your phone is all Internet. Chances are if you work in an office building, all your phones are VOIP already. He states:
“This prioritized service has been around for several years, and has revolutionized the phone industry. Something unthinkable for decades — facilities-based local telephone service — became commonplace in the last few years, and undermined much of the careful industrial planning in the 1996 Telecom Act. If you subscribe to voice service from your cable provider, you are benefiting from fast-lane treatment.”
Thus, the author concludes, because voice traffic is treated differently, we may as well treat other traffic differently, so this fast lane proposal is no big deal.
One problem with this line of reasoning, though, is that voice over IP isn’t some brand name. It was, in effect, already part of the system. Remember above when I said that under the regulation, Ma Bell managed to replace direct copper wires to your house with a packet switched network? Yeah, to do that, you had to invent protocols that allowed those packets to flow in timed fashions, such that it looked as if it was a direct connection like it was before. That was a new set of protocols.
So, this isn’t some new, recent invention where the cable company saw voice connection using IP, and then had to devise a solution for that that was a “fast lane”. This packet concept is baked into the Internet already. To call these new proposals an extension of the fast lane is laughable.
But rather than acknowledge that, the author then says:
“Most troubling, net neutrality turns the regulatory process on its head. Rather than identify a market failure and then take steps to correct the failure, the FCC may prevent commercial agreements that would be unobjectionable in nearly any other industry.”
The packet switching capability that treated “voice” as a streaming service was developed under a strict regulatory scheme. There didn’t need to be a “commercial agreement” that businesses hammered out. It was a core piece of the Internet itself.
A similar analogy can be made to the highway system. Regulations exist that ensure all vehicles can use the road. This means that curves are only build to happen at specific angles. Overpasses are built to be specific heights that allow trucks to pass underneath. Traffic lights are imposed to make sure that vehicles don’t collide. While it is possible to have built a separate set of roads through “commercial agreements” such that, say, UPS could get better service, the idea is just… dumb. Nobody would stand for it.
The Myth of “We Have to Charge Traffic Providers” Because… Netflix!
You may have heard stories that when shows like the new season of House of Cards came out, or when the finale of True Detective happened on HBO Go (or the premiere of Game of Thrones), the Internet became clogged with requests for these shows. Up to 1/3 of Internet traffic became people trying to watch these shows. Internet companies furrowed their brow and cried foul, demanding that Netflix should pay for this.
The one problem with that, however, is that it misplaces who should get charged for what. Yes, people were watching Netflix and HBO Go, but when the day gets hot and we all turn on our air conditioners, the managers of the electric grid don’t go to power plants and charge them, they charge… you. If you use more kilowatts, you pay more. And if you use those kilowatts at specific times, you pay more.
Internet Service Providers (ISPs), such as Comcast or AT&T, are in this bind because they have given their customers a promise of unlimited data usage, and then have purposely not done anything to make sure that unlimited data is actually achievable. In fact, you can read this blog post from an infrastructure insider that shows this explicitly. While the blog post is a bit technical, the key takeaway here is that US ISPs have not done the work to make sure their networks can handle traffic, even though they have seen increases in traffic for a long time.
Going back to the electrical grid as an analogy, it would be as if the grid were privatized, the company managing the grid knows that the grid is loaded and could suffer brown-outs, but rather than enhancing the grid to prevent brown outs, they simply let the brown-outs happen and then shrugged their shoulders about it.
This doesn’t happen (the brown out), because we don’t allow it to happen through… wait for it… regulation. California tried an experiment with deregulation in, and we saw what happened – the grid didn’t get maintained, power plants were purposely taken offline, and brown outs happened. As soon as regulations were put back in place, bam, brown outs disappeared. There were arguments raised that us Californians were all just lazy and left our pool heaters on, but that simply wasn’t the case. Californians behavior prior to de-regulation, during de-regulation, and after de-regulation didn’t change much. What changed was the regulatory environment – that’s it.
The other aspect of this blog post that is interesting to note is that this kind of packet dropping due to congestion doesn’t happen in other countries. Why? Wait for it… it’s because they are regulated. They saw traffic increases starting, and because they are regulated, they were required to spend money to ensure the congestion didn’t break the network.
So, once again, the free market idea of just letting cable companies and phone companies build out their infrastructure as they see fit hasn’t prevented the problem. This is especially troubling because the same companies that are allowing this kind of congestion are also in the market for streaming services that they would provide to your home. In other words – they offer service that would clog up the network already, thus they could predict this problem, yet they still did nothing about it. Why? As the author states:
“Five of those congested peers are in the United States and one is in Europe. There are none in any other part of the world. All six are large Broadband consumer networks with a dominant or exclusive market share in their local market. In countries or markets where consumers have multiple Broadband choices (like the UK) there are no congested peers.”
While the blog post doesn’t mention companies by name, Comcast is clearly one of them (as it has dominant or exclusive market share in many markets). Comcast has a service like Netflix, called “Streampix”. Nobody uses it, because it sucks. But clearly, Comcast would like you to use it. So… why isn’t Comcast using the fees for Streampix to build out their network for this coming problem? That’s what the “free market” is about, right?
Streampix costs Comcast users $4.99/month. Which is cheaper than what Netflix charges per month. And recently, Comcast used their congestion problem to get Netflix to enter into what is called a “peering arrangement”, where Netflix pays Comcast for priority access.
(Aside: peering arrangements are typically used between owners of various networks, where they agree to let each other’s on to the other network at no, or reduced fees, and they pay each other the difference if one network owner use more of the “peer” traffic than the other network owner. In the case of Netflix, which doesn’t have its own network, this just becomes a cash payment by Netflix to the network owner).
Because of peering, Netflix loses some profit…. which means Netflix may have to charge users more for its service. Sure enough, Netflix announced changes to do exactly that. While this is only for new subscribers, it very well could hit existing subscribers shortly.
But Streampix? Streampix is not charging more for new subscribers. It still gets a free ride.
Let that sink in. An existing company that people like, Netflix, is raising its rates to ride on a set of train tracks (Comcast’s network), while the owner of the train tracks has a competing service that isn’t subject to the same charges. There is something inherently wrong with this. Will Streampix ever have to pay more? Possibly. Would you feel a little bit better if Netflix cost you more if Comcast didn’t also offer Streampix? Yes.
And that is the conundrum here with getting rid of net neutrality and letting fast lanes be developed. Our FCC has created a system where, in the interest of not promoting “winners and losers”, resulted in two dominant pipelines into your house (your phone company and your cable company), pipelines that happened to already be coming into your house when the FCC decided not to regulate the pipes. They’ve let those companies merge with content creation companies such that they aren’t just the pipes, but also the content in the pipes. It is like letting UPS own roads, and charging Fed Ex to use the road (a “maintenance” fee), while letting their trucks ride on it for free.
Bringing it Home
So, just like with health care, the arguments about “government control” that are so worrisome to people are easily dispelled. Innovation we use in this country was created under regulatory systems in foreign countries (LTE), and here (packet switched network). Video streaming services don’t pay more money to use the network, and they don’t get congested or dropped. Things that are supposedly examples of “commercial arrangements” are actually part of the core of the Internet already, and aren’t examples that prove we need more de-regulation.
With health care, the Ayn Randian philosophy of the invisible hand of the market making things work just didn’t always apply. Artificial hips were invented in England under the socialized NIH. Erection dysfunction treatments were developed in the United States. There is more money to be made in erection pills than artificial hips, so that is where an unregulated market will go. But that doesn’t mean the system is “better”. The “invisible hand” of the market goes where the easy money is.
Similarly, with high speed Internet, the free market isn’t just making our lives full of unicorns and rainbows. As I pointed out here, American high speed Internet is slower, and more expensive, than elsewhere. And this is true even for countries that have the same population density as the United States. The brilliant podcast Planet Money did a segment on the history of how we have the system we have (link) and shows how while the US deregulated, England regulated, and the results are much better for consumers in the UK than in the US.
As I already showed, the innovation dilemma more Libertarian minded folks are worried about and use as a cudgel against the FCC doing anything, the US can have plenty of innovation under a regulated environment, and in fact, we as users in America are benefitting from technologies developed under heavily regulated environments. Every time you look at your smartphone and see that it is LTE, you can thank Japan, Sweden and a bunch of academics that developed the algorithms.
Let’s put to rest other arguments around the fear of government control. The only reason you have a cable or fiber going into your house is due to state and local governments creating regulations as to where those cables can be laid. You can only use your smartphone because the government determined which frequencies you were allowed to transmit and receive signals on. The government is already very much in your face about the technology you use. Regulating against fast lanes is not some new, unheard of power leading to the death of freedom and you being placed in a FEMA re-education camp.
Regulating against fast lanes, and ensuring a company like Comcast can’t do things to hurt Netflix while enabling Streampix will, in all likelihood, make your life better. Just like health reform is starting to do.