Mauboussin on the Wisdom of Crowds

Ryan Leggio: Hi, I'm Ryan Leggio. I'm a mutual fund analyst here at Morningstar, and with me today is Michael Mauboussin, the chief investment strategist of Legg Mason Capital Management. He's also an adjunct professor at Columbia University, the author of numerous books, and his newest book is entitled "Think Twice."

Michael, thanks so much for joining us today.

Michael Mauboussin: Ryan, great to be with you.

Leggio: I thought we'd talk a little bit about the title of your book and how you basically used a "wisdom of crowds"-type process and an Amazon tool to come up with it.

Mauboussin: Exactly, well, I try to practice what I preach a little bit. It's an interesting thing. We debated about what the title should be, and I thought "Think Twice" would make some sense because it's active. It's literal. It's a good title.

My editor was a little bit uncomfortable with it, and it turns out I realized why. About a year ago they published another book called "Think Again," which is not quite as good a title, but you can see how that would cause some confusion.

So I said to her, "Just give me some title ideas, and we'll see."

She'd throw them at me, and I would say, "Why do you like that one?"

They would say, "Well, we just think it sounds good."

There was basically no foundation for it. So we decided to take matters into our own hands, and we created a title tournament. You're exactly right. We went to Amazon.com. It's got a feature called Mechanical Turk where you give what are called Human Intelligence Tasks, or HITs, to people for small micropayments.

<TRANSCRIPT>

I said, "Give me all the titles you want to list," and we listed them in a tournament format. We paired them off, and we just went down the line.

It's a great feature, by the way, because you leave work, you hit a button, it goes around the world, and hundreds of people answer your question. Then you come in the morning, and the answers are there.

We ran the tournament, and "Think Twice" won the tournament hands down. I do like to say, as you pointed out, the title of this book was determined by hundreds of random people around the world willing to answer a question for 10 cents. That really is the truth. We really did find that that title was by far the most popular.

Leggio: You've written a lot about the wisdom of crowds and if there are certain conditions that are met, you oftentimes get great decisions by diverse people, et cetera.

I guess I'm wondering how that concept might apply to money management. We see a lot of times managers or consultants or mutual funds choose, say, five mutual funds and allocate 20 percent to each instead of picking the best ideas of those five managers and building a portfolio that way.

Are there any principles from the wisdom of crowds you think the money management industry could be a little smarter in using?

Mauboussin: That's a great question. Just take one step back. As you point out, there are certain conditions when it works and conditions when it doesn't work. It's worth taking just a moment to talk about those.

The conditions are typically three. One is diversity. We need to have, for markets to be smart, diverse people, and that is technical, fundamental, long term, short term, different techniques.

The second is a properly functioning aggregation mechanism, which of course exchanges do well.

Then the third is incentives, which are rewards for being right and penalties for being wrong. Of course it's financial in our world, but it doesn't have to be. It could be reputation or other things.

When those conditions are happening, you tend to get very efficient results. So what we're looking for as money managers is when one or more of those conditions are violated, and by far the most likely to be violated is diversity.

Rather than people thinking differently about a topic, they tend to get on the same side of the ship, which leads to excesses and fundamentals that get out of sync with expectations.

One of the challenges with doing a meta-wisdom of crowds is the market itself is already doing this. So it's hard to say I'm going to pick the best ideas because the market itself is doing this now.

To me, the better way to conceptualize that is to find people that consistently focus on that fundamentals-expectations gap and the varying perception. Often the source of the varying perception is some sort of diversity breakdown that you can identify.

Leggio: A lot of times, with a lot of Legg Mason funds, it could be concentration or other avenues to differentiate yourself from the crowd?

Mauboussin: Yes, I think the ultimate objective for us and all of our funds is consistent, which is delivering excess returns for our fundholders.

So things like concentration or even turnover, these I would say are tactics to serve the ultimate objective, and we happen to believe that more concentrated portfolios and relatively long time horizons ultimately serve those objectives, but there's nothing immutable about those. Those are means to the end.

But we do think that, indeed, that is the case, that taking longer-term perspectives, three-to-five year perspectives, on things and typically more concentration. We still want to diversify the portfolios, but say, 30, 40, 50 names instead of 100 or 200 names, again, best serves those purposes.