A Freakonomics Contest: The Coase Theorem Online

Freakonomics schwag is up for offer at the end of this post. As such, it may actually be worth slogging through the brief economics lesson that follows.

The Coase Theorem is a somewhat rare species of beast: an economic theory that is both completely counterintuitive and yet often right in practice. The idea is named after Ronald Coase, one of the University of Chicago’s many Nobel Laureate economists. Nearing the age of 100, he may not hear quite as well as he used to, but otherwise he is still going strong, sitting on the Board of Directors of the Becker Center, giving lectures, writing academic articles, and attending the occasional seminar.

The basic idea of the Coase Theorem is that no matter who is assigned property rights, as long as transaction costs are not too high, the efficient outcome will be achieved. In his original article nearly fifty years ago, Coase motivated this idea by writing about the problem of sparks from railroad trains setting wheat fields on fire. We will assume that these fires are very costly, and as such it is best to take action to prevent them from occurring.

Let’s say there are two ways to avoid the risk of fire: adding some attachment to the train that catches the sparks, or having the farmers not plant wheat close to the railroad tracks. The naive view of the problem would be this: if the law dictates that railroads are responsible for the losses to farmers from the fires they start, then the railroad will invest in the spark-catching attachment, since it is a cheaper alternative to paying the farmers if a fire occurs. By this logic, if the law says that the farmers are responsible for the losses, then they will not plant wheat next to the tracks.

The Coase Theorem states that this logic is wrong. Regardless of who is liable, the two parties should bargain to the efficient solution. Let’s say it is cheaper to stop the fires with the attachment. Then even if the railroads don’t have to pay for fire damage, the farmer should offer to pay for the railroad attachment, plus give the railroad a little extra money. The railroad is now better off (it doesn’t mind having the attachment, and it has more money) and the farmer is better off (the attachment was a cheaper alternative than not planting next to the tracks). Everyone is better off, so we expect that this will be the outcome.

The obvious obstacle is what economists call “transaction costs.” If there are many farmers and many railroads, or private information about costs or how much care the parties are taking, bargaining can break down and an inefficient outcome can occur.

With that as background, the actual point of my post relates to how well the Coase Theorem seems to hold in online transactions. My understanding is that more or less anyone can purchase any URL that they want. In that setting, you might expect chaos on the Web, with Web addresses that you would think should belong to one company or organization actually leading somewhere completely different. But the Coase Theorem predicts otherwise: regardless of who originally purchases the right to a URL, ultimately it will end up in the hands of the user who values it most highly. In some cases, when a Web address is really valuable, there may be large transfers of money — according to Forbes, the domain name “Seniors.com” went for $1.8 million at auction.

But there must also be some stark counterexamples in which the Coase Theorem fails. One example is the following: what do you think happens when you enter www.american.com, as I did yesterday, while searching for a plane ticket. American Airlines? Nope — a new magazine that features all sorts of interesting and intelligent stuff, like an article by Alan Krueger about terrorism and a profile of economist Jesse Shapiro. It is hard for me to believe that this URL wouldn’t be more valuable to American Airlines. I’d love to know the story of how it came to be owned by an upstart magazine instead.

So, here is the reader challenge: find the best examples of the Coase Theorem failing on the web. What URLs seem like they should logically take visitors to one place, but in fact lead them somewhere completely different? We’ll give some Freakonomics schwag to the best reader entry.


Keith Cambpell

Under this definition, "What URLs seem like they should logically take visitors to one place, but in fact lead them somewhere completely different?"

I would submit whitehouse.com which directs you to a porn site, not the White House. This is a very well know example, but I know some very embarrassing stories of good intention people, my mom, who have learned the valuable lesson of the difference between .com and .gov.

Chris

I can't verify what website is there these days since I'm at work, but my younger cousin was once trying to reach Dick's Sporting Goods and naively went to www.dicks.com. You can imagine his surprise, and I'd think that Dicks would want to take care of this (common?) mistake.

Mike

http://nissan.com/ is not owned by the car company, but some computer company with a really bad web designer.

Also, http://www.theeconomist.com is not owned by "The Economist" magazine (www.economist.com).

Eddie Camp

www.cambridge.edu So Cambridge is in England and has an English URL. Still, Cambridge career college in Yuba City?

Kevin C

I am an 11th grader in high school taking my first year of economics, and I would just like to start by telling you how much your book has changed my perspective on life and what is going on around me. But back to the reason I'm posting. I found these two websites, martinlutherking.org and tvs.com. The Martin Luther King website is racist and does not actually tell about him, but portrays him in a negative light. The second website I would think a big company would have like Panasonic or a TV retailer, but instead it is just a generic search engine.

YC

CoaseTheorem.com has yet to be registered...

David Newberry

I believe that you might be operating on an incorrect assumption. The value companies place in urls closely matching a particular name shouldn't be that high.

Top ranked searches on Yahoo and Google are usually just the names of a company to direct the searcher to the company website. If you don't know the url, you search for it. Few internet savvy visitors will simply guess at a url.

Perhaps a more interesting survey would be to look at how a company works to ensure a high placement in internet searches for its name.

Roman

Cosi.com is owned by COS, Incorporated rather than the restaurant.

Sam

www.brucespringsteen.com

The Boss tried to get this one but had to settle for www.brucespringsteen.net

Justin

Nissan.com is occupied by a small computer company in NC and not the multinational car company.

matt gulde

www.whitehouse.org, despite the absence of a .gov extension, is a pretty good example. Someone looking for the official site of the president's residence could easily land at this site, labeled the "officious" web site of the white house and president Bush, and bearing the presidential seal with a vulture in place of the eagle. A google search of WHITE HOUSE returns the whitehouse.org site as the third result.

Pablito

http://www.citistreet.com/

The joint venture between CitiGroup and State Street in the defined benefits industry is actually called CitiStreet

Devan

In any case, the URL is slowly becoming the second-class way of getting to a web site. Now (and more so in the future), you won't need to have the most apt URL- you will just need the top listing on the most popular search engines. I don't care what the URL for American is as long as I can search it easily.

squidbot

Typosquatting in general seems contrary to this theory.

http://en.wikipedia.org/wiki/Typosquatting

Willy

Whitehouse dot com used to be a porn site, it has since changed hands, but still has not proven the Coase Theorem correct, since it's not Government affiliated (not sure if they only use .gov TLDs).

Kevin

Email addresses? Especially for people with common names. I guess if you are rich enough you would just host your own email.

Andres

television.com

Adam

There are brokers who buy up domain names of athletes and celebrities with the intent of selling the domain names to those celebrities. They don't always buy.

Try warrensapp.com, for instance.
Sapp refused to pay the outrageous price the broker wanted and instead set up qbkilla.com as his official site.

Chris

I would have thought "www.wood.com" would have something to do with, you know, wood. As it turns out, it is the site for an investment bank.

Dscottgrrl

just found an article on digg about most misleading website URLs (television.com is mentioned)

http://www.omglists.com/article/41593/the-top-7-most-misleading-urls/