Saturday, June 21, 2008

Index Fund Fees, Expensive Tastes, and the Dijon Mustard Paradox

(Felix Salmon) You're asked to allocate $10,000 between four S&P 500 index funds. The first has a front-end of 2.5% and charges fees of 60bp per year; the second has a front-end of 5.25% and charges fees of 64bp per year; the third has a front-end of 5.75% and charges fees of 75bp per year; and the fourth has a front-end of 2.5% and charges fees of 70bp per year. The obvious answer is "bugger that, I'm buying an ETF". But if you have to confine yourself to the four funds in question, you'd simply put all the money in the first one, which is the cheapest one, right?

Of course you would. But boy are you in the minority. When a bunch of Harvard professors and Wharton MBA students were asked to do exactly the same thing - and would make real money by picking the cheapest fund - the overwhelming majority of both groups put most of their money in gratuitously expensive funds. (The way the experiment was designed, there was no benefit at all to investing in one of the more expensive funds.)

The full paper, by James Choi, David Laibson, and Brigitte Madrian, is here, and is well worth reading. The pointer is from Leonard Mlodinow's very good new book The Drunkard's Walk, which has other great facts too. For instance:

A few years back "The Penguin Good Australian Wine Guide" and "On Wine's Australian Wine Annual" provided an Alexei test: they both reviewed the 1999 vintage of the Mitchelton Blackwood Park Riesling. "The Penguin Guide" named the wine Penguin Best Wine of the Year. "Wine Annual" rated it the worst vintage of the decade.

Not just the worst vintage of the decade, but also the worst wine of all the wines reviewed: that paper is here.

But back to those index funds. The authors show quite convincingly that investors look at the wrong things when judging funds, and don't pay nearly enough attention to fees. But what they didn't do, unfortunately, is run an experiment where the only information that the subjects had was the total fees. I'd love to know what the result would be in that case, because I have a suspicion that MBA students, especially, have internalized demand curves to the point at which they reckon that a more expensive product must be superior.

In other words, the subjects might not have been choosing the more expensive funds despite their fees, but rather choosing them because of their high fees. The logic goes something like this: if the funds were genuinely identical but for the fees, then everyone would simply choose the cheaper fund. But the more expensive fund is clearly still going strong, so there must be some reason why it's superior to the cheaper fund. I don't know what that reason is, necessarily, but it still can't be completely stupid to put at least some of my money in the more expensive fund.

This kind of thinking helps explain, I think, what I consider the "Dijon mustard paradox". The best Dijon mustard available in New York is Trader Joe's; it is also, by some margin, the cheapest Dijon mustard available in New York. Yet even people who already shop at Trader Joe's often end up getting more expensive Dijon mustard, either at Trader Joe's or elsewhere. And the reason, I think, is that they're in some sense put off by the low price. If it's that cheap, they think, it can't really be all that good. Price signals are so important that they make people doubt their own taste.

All of us do this. For many years I didn't "get" Ed Ruscha. I knew his work, I pretty much understood the ideas behind it, but I thought it was all a bit silly and/or superficial. Eventually, however, the drumbeat of the art market was so loud and so incessant that I changed my mind, and now I really like Ruscha, and think he's a great American artist. If it hadn't been for the market, that change of mind would probably never have happened.

Some people naturally have expensive tastes. Many more, however, end up ajusting their taste to what is expensive. It's weird: wouldn't we be happier if we adjusted our taste to what is cheap?

No comments: