Goods, marketing, preferences and the Genius

I've gotten used to ignoring iTunes pining for one upgrade or another - seems to happen every other time I start it up - but, this time, the promise of the Genius was too tempting to ignore. 

The Genius is a new gadget attached to iTunes that is designed to create "smart" playlists at the touch of a button; pick a song from your library and it'll create a playlist made up of songs that fit well with the one you picked. If that sounds familiar, it's because it's precisely the same idea that personalized internet radio (my favorite is Pandora) has been trading on for a few years, with the obvious difference that Pandora and its cousins are capable of creating playlists built on more than one song or genre and that they can play you songs you don't own. Both Genius and Pandora are working on a collective wisdom principle - learning what to recommend based on what you like and on what people like you like, and so on. Actually, that even sounds a bit like the supposed basis for Goggle's search algorithm, right?

Anyway, I've been crying out for this: shuffle in iTunes is precisely useless, and I'm pleasantly encouraged by early returns from Genius. What's all that got to do with economics?

Well, what do we assume about preferences? We've always tried to be agnostic about what people like, which was neat for keeping us honest but made us a little fuzzy on where they come from or how they might change. Genius and Pandora are, arguably, operating on a person with malleable preferences, who's persuadable that she likes what's being connected. For these to be valuable products, there must exist preference for convenience (having playlists made for you rather than doing it yourself) or surprise (for hearing unexpected songs). 

Or could it be as simple as information? Perhaps these things are most valuable as informers, letting us know what goes well with our music or showing us new music that we might like. Again, slightly non-traditional consumer theory, although imperfectly informed consumers isn't a new idea in economics or in life. Strangely, iTunes seems to be at a distinct disadvantage here; sure, this new gadget is designed to recommend things to you to buy at the iTunes store, but it's not a pure free-preview model like Pandora. Nevertheless, we're in the territory of deterministic preferences, and in particular the role of advertising.

The most commonly asked question about advertising in economics has been 'informative or not?', and there's the balance between these two functions: am I being coerced or informed when Genius tells me what songs I'm missing that would go well with one I own? The strict segregation of preferences and the tangible world of goods and budgets that is used in consumer theory is violated a little by this kind of thing; we get into nice philosophical questions like how to define a good. Are my preferences changing when I learn about these awesome songs I'm missing out on?

Irrationality again

I resisted talking about 'Predictably Irrational', a book by Dan Ariely, "the Alfred P. Sloan Professor of Behavioral Economics at the MIT Sloan School of Management and director of the eRationality Group at the Media Lab", when the first wave of columns and reviews appeared about it. I haven't read it, but it is mining the vein of doing experiments to figure out how people behave.

The title, of course, is not palatable to me. It's not possible to test rationality. I see why it's attached to the work that behavioral economists do, but semantically, it's a real pain. Let me dive right in to this article, direct from MIT News.

"Though Ariely's book is often compared to the bestseller "Freakonomics"--both certainly share a quirky, hands-on approach to questions of everyday behavior--he says that in fact his research is almost the opposite of that book's. Those researchers found cases where people's behavior, even in seemingly irrational contexts, was perfectly rational and followed established economic principles. Ariely's work, by contrast, shows the consistently irrational ways people behave in situations where traditional economics predicts they would follow a course of rational self-interest."

Goodness me. Consider the bait taken: what's 'traditional economics' and why is it different from behavioral economics (or are they the same)? What's an 'irrational context'? Here's an example, from the article, of the Ariely book:

"Ariely and his students went around and left six-packs of Coke in randomly selected dorm refrigerators all over campus. When he checked back in a few days, all of the Cokes were gone.

But when he later placed plates of six loose dollar bills in those same refrigerators, not a single bill was missing when he checked back. Even though the value was comparable--and thus the situations were supposed to be equivalent--people responded in opposite ways. Why is that?"

First of all, if I see a plate of loose dollar bills in the refrigerator I'm pretty well out of my comfort zone. Can it be so hard to explain why people don't take dollars from a plate in a dorm refrigerator? Aside from being a bit silly, it's the first misperception of economics at work! 'The value was comparable'. Actually, I'm being unfair: this is worse than the first misperception of economics, because the Coke-dollar game, as reported by the article, is refusing to acknowledge any preferences whatsoever, ignoring, then, the most fundamental building block of modeling in economics. Maybe that's why it's 'not traditional'.

There are other examples in the article, and I'm sure the book is full of interesting experimental results. I can't get past that title, though, and it, like Freakonomics itself and the cottage industry it spawned, is squarely in the making-economics-look-stupid camp with the Christmas stuff. I'm sure insights from behavioral experiments can be informative beyond the triviality - the Obama advisers come to mind - but the press coverage for this new book has instead reinforced the discipline's 'quirkiness' and furthered, in some small and delightful way, the misunderstanding of economics.

Can you test rationality?

Is it possible to test if people are rational? I think the answer, practically, is a very short no: if a rational person tries to achieve his most preferred outcome of the ones that are available, we can't distinguish the rationality or irrationality of his choice from his preferences. That is, if I don't know what you like, I can't tell if you did something because you liked it or because you're "irrational".

Yet rivers of ink have been spilled trying to "prove" or "disprove" models of rational choice. The most famous study of the type is the "Allais paradox", discussed here. It says that when you pose different choices to people, their responses to pairs of choices are "inconsistent" with each other because the two choices really represented the same cash outcomes.

Whether you look at the question being asked by this type of work "are people rational?", or "what do people care about?", it's pretty clear that any observation cannot answer either of these without knowledge of the other. In "The Methodology of Positive Economics" (pdf) Milton Friedman made the valid, general point that

"If there is one hypothesis that is consistent with the available evidence, there are always an infinite number that are."

It just so happens that if we interpret some piece of evidence as being consistent with "people are irrational", one of the "infinite number" towers above all others: "you guessed the preferences wrong". There's nothing wrong with trying to figure out how to better model the decisions of people, but claiming to have proved irrationality is nonsensical.