Getting the 'Outside View'

Ryan Leggio: ... Switching gears to your book, you talk a lot about the differences between the outside and the inside view. I was wondering if you could explain that concept briefly for people who aren't familiar with it.

Michael Mauboussin: Yes. It's a really important idea, not only in money management but also in life.

The inside view is typically when you have a problem you're trying to solve or a forecast you're trying to make. What you do is you gather lots of information about it, you combine it with your own inputs, and then you forecast.

That could be what's the market going to do. It could be how long is it going to take you to remodel your kitchen. It could be when are we going to launch that new product. Very consistent.

The outside view, in contrast, looks at your problem as an instance of a larger reference class, which allows you to ask a very simple question: When other people have been in this situation, what happened?

It turns out a lot of problems we face or challenges we face may be unique to us, but there's this thing called the database of humanity. A lot of people before have gone through these things, and we tend not to pay attention to that.

The inside view typically leads to too optimistic forecasts about the future, and so what the outside view often does is temper some of those judgments and give you a much better perspective. So that is a big one.

<TRANSCRIPT>

By the way, again, it's a very unnatural way to think. The natural way to think is the inside view because you've worked hard to gather that information. You've got your own input.

The outside view, in a sense you have to leave aside that cherished information and actually defer to a larger sample set. That is a very powerful idea.

I'll give you two concrete examples how one might use it. We have a lot of data on growth rates of corporations. We have a lot of data on margin structures and return on capital patterns.

There was a good friend of mine who's a sell-side analyst, and he sent a report that recommended Amazon.com. We love Amazon, but it suggested that they were going to have 25 percent compounded annual revenue growth for the next 10 years. Now they're off about a $20 billion revenue base.

My simple question was how many companies in the history of mankind have ever done this? The answer is either zero or one, and if it is one, it's going to be Wal-Mart. Again, am I saying it's not possible? No, but when you think about it probabilistically, is it high probability?

The other example I think is going to be much more prominent in the next few months, quarters, is mergers and acquisitions. Same thing. Most executives know that most deals don't create value for the acquiring company, yet they think the deal they're going to do is going to be one that creates value.

So thinking about the broader outside view can really be very instructive in many cases.

Leggio: You talk about how the inside view can sometimes be too optimistic. Currently, one of the views which your firm has talked about is really the inside view maybe being too pessimistic.

Mauboussin: Exactly.

Leggio: Can you talk a little bit about that?

Mauboussin: Yes. The thought is that people take recent experiences and they tend to extrapolate, and this is something Bill Miller's talked a lot about.

One of the best predictors of the magnitude of the subsequent recovery post-recession is the depth of the prior decline. If you go back the last 50 or 60 years and look at the size of declines and the subsequent recoveries, it allows you to draw a pretty decent regression line.

If you place 2009 on that and look at what the 2010 GDP growth rate should be, the regression suggests something north of 8 percent. Now, 8 percent, obviously I don't think anybody believes that. But the consensus is now around 3 percent, high twos, low 3 percent.

Our basic take is if you have to take the over-under on that, at this point, the outside view would point to the over being more sensible. I would talk about maybe a 4 percent or even slightly higher than 4 percent GDP growth, which obviously has implications for corporate earnings and for the general pace of recovery.

I think the basic scenario, which might be a good scenario, is we get a healthy earnings recovery. We get some GDP improvement. Inflation stays tamped down for a period of time, which tends to create a pretty decent backdrop for investors.

Leggio: Another outside view with that implication of all this growth is that we're going to see better returns over the next 10 years because in the previous 13 times where a decade period has had zero or close-to-zero returns, the returns in the next decade have been quite high.

Mauboussin: Exactly. I think this is another one. This is our diversity breakdown. Ten years ago we would have gotten together, and we would have just finished a smashing decade, a fantastic decade, for the 1990s, where we had returns vastly in excess of the historical averages.

Now we finish a decade where we're vastly below that, but people tend to extrapolate their recent experience. So, in 1999, everybody was very optimistic about the market, and today people are bailing basically or very pessimistic. The funds flow for equity managers, it's amazing that people are pulling a lot of money out.

So, yes, the answer is equity markets tend to deliver about 6 to 7 percent real returns over time, some periods better, some periods worse. If you've just come off a bad period, I think you can again reasonably expect in the future to see something back more toward that historical average. That's probably where we sit today.

So, yes, more optimistic.