Byrne's Blog

A Challenge for Net Neutrality Supporters

The great thing about most critiques of Laissez-Faire economics is that they make great business plans. If “the market” can’t provide something, but that something is in demand, it’s a great opportunity!

This is an easy filter for determining which deregulation critiques have merit: if the business plan falls apart due to high transaction costs (e.g. charging for sidewalks) there’s a sound argument for regulation. But if it falls apart for some other reason (“People are greedy!” or “People—different people this time—are too dumb to advance their own interests!”) there’s a good case for privatization. After all, people are just as fallible regardless of who pays their salaries. But if their fallibility has real-world consequences, they’re likely to face those consequences eventually.

The usual go-to example is the Post Office, which manages to earn a worse return on investment than FedEx or UPS, even though it gets to be a monopoly provider of some services. But in a couple decades, the post office will be foreign to most people, and we’ll need another good example for that Laissez-Faire “Show me the business plan!” theorem. And I think Net Neutrality will provide it.

Here’s Craig Newmark’s argument against Net Neutrality:

Here’s a real world example that shows how this would work. Let’s say you call Joe’s Pizza and the first thing you hear is a message saying you’ll be connected in a minute or two, but if you want, you can be connected to Pizza Hut right away. That’s not fair, right? You called Joe’s and want some Joe’s pizza. Well, that’s how some telecommunications executives want the Internet to operate, with some Web sites easier to access than others. For them, this would be a money-making regime.

I can see how that would work. But keep in mind that the phone company is also out to make a buck. So they’re going to charge for this service. How much will they charge? Well, they’re selling two different products. One is the inconvenience that Joe’s Pizza customers face; that’s going to reflect badly on the phone company, so the company has to charge to compensate for that. And of course there’s lead generation; some folks will hear the message and call Pizza Hut right away. And since Evil Telco can sell those piping-hot leads to anybody, it’s going to be tough for Pizza Hut to argue that it should get much profit from them at all. In fact, Pizza Hut might lose money on the first order, based on the hope that they’ll get a loyal customer out of it.

So from Pizza Hut’s perspective, what we have is an okay deal, not a great one. For one thing, those extra orders they’re getting aren’t especially profitable. Not that they can really say no, since Evil Telco can just sell the same leads to Papa John’s or whoever else. Meanwhile, they have to pay extra every time someone calls up their competitor. Maybe at the corporate level, that’s a low cost, but it’s the decision is made at the regional level; if Pizza Hut is paying to get a leg up on a competitor, that cost will be borne by whichever location is directly competing.

Meanwhile, Joe’s Pizza has an interesting situation. They know that Pizza Hut now has a less forgiving cost structure—they have to count on people who give up on Joe’s, because those people are the only ones who will send higher-margin business to Pizza Hut. But Joe’s also knows that every call they get costs Pizza Hut money. In other words, the faster Joe’s grows, the more (comparatively) profitable they get.

This gives Joe’s Pizza an excellent opportunity: their lower costs can subsidize an ad campaign targeting people with more time than money; those people are more likely to wait their two minutes to order a pizza, especially if it’s cheap. Meanwhile, those folks—who wouldn’t go to Pizza Hut anyway—are still costing Pizza Hut money. And since Joe’s now has a better competitive position, it’s a much easier sell to the local bank’s loan officer—worst-case scenario, Joe’s merely gets new business; best-case scenario, they run Pizza Hut out of town. (Just look at what would happen if Joe’s had a bigger market share than Pizza Hut. Pizza Hut would be paying a fee every time someone ordered pizza, but most of those orders would go to other people.)

There are three temporary equilibria here. First, status quo: Joe’s sells you cheaper pizza, if you’re a little more patient; Pizza Hut sells more expensive pizza fast. Second, the scenario Pizza Hut is counting on: Joe’s goes out of business (of course, this behavior is covered under your standard-issue antitrust laws; it would be very hard to convince a judge that this wasn’t monopolistic behavior, and very hard to convince a CFO that eliminating Joe’s Pizza was worth the risk of a lawsuit). Third, Pizza Hut loses, and either gives up on their phone call rerouting scheme, or leaves town entirely.

But we don’t have to go that far. There have been very few successful schemes that involved company A paying supplier B to provide worse service to all of company A’s competitors. It raises A’s costs, and leaves it beholden to B. And even if it works, it’s likely to provoke an anti-trust lawsuit.

Thus, my challenge to Net Neutrality supporters: Show me the business plan. If you were a big, evil corporation who could strike a deal with a big, evil telco, what would you ask for, what would you pay, and what would constrain your competitors from adapting their business to your strategy in a way that made the original strategy unprofitable?

The horror stories about net neutrality are largely impractical. There’s a reason you see more of them from journalists than from entrepreneurs. (Another big constituency: people who use bittorrent and other filesharing software. Oddly, I don’t hear much net neutrality opposition from people who use iTunes.) Net Neutrality would just make it slightly easier to match the revenue from network infrastructure to the people who earn the most from it. If Youtube and Facebook are paying extra—for the extras they get—the worst outcome is that the laws of supply and demand will more efficiently apply to bandwidth.

The horror.