As chief investment strategist of Legg Mason Capital Management, Michael Mauboussin's job is to understand the world and then make money off of it. What he's found over the years is that investors, like any other group of people, are prone to make mistakes that stem from faulty approaches to decision-making. In Think Twice: Harnessing the Power of Counterintuition, Mauboussin also an adjunct professor of finance at Columbia Business School pulls from fields such as psychology, statistics and complexity science to explain how we might do better. TIME's Barbara Kiviat spoke with him.
You just wrote an entire book about how smart people make bad decisions. Maybe you could shed some light on the financial meltdown.
Increasingly, we as human beings create systems that are more complex than we can comprehend. If you're designing an airplane you create redundancies to try to mitigate bad outcomes, but as we saw earlier this year with Air France 447, when certain cascading events line up, they can lead to a disaster that you just can't anticipate. In the realm of the social, we create institutions like global, interconnected markets. By and large when investors are operating with some degree of diversity, when they're thinking about things differently time horizons, trading strategies the markets are basically efficient. But what we know is that people episodically coordinate their behavior. They all become optimistic or pessimistic, and that leads to extremes. Even with the best of intentions, you can't anticipate all the things that are going to happen.
What does that mean for regulation?
You can set up certain decision rules that try to keep people on the straight and narrow, but too much regulation about the details is going to get you in trouble. The big banks should probably be fairly boring and reasonably well regulated. And their core businesses are but they were able to get away from their core activities and move further out on the risk spectrum and jeopardize the whole operation. So much more stringent capital requirements for banks as they get larger, to make sure that as they get larger they get more boring. For compensation, one way to mitigate risk is to defer some chunk of compensation so that it's only success over time that you get paid for. Having a compensation czar dictate who should earn what is probably not a great solution.
Let's zoom in to the level of the individual. How do you approach investing differently, knowing what you do about the psychology of decision-making?
One theme is contemplating different outcomes and attaching probabilities to those. There is this notion called the inside-outside view. The inside view, which is how almost all of us plan, is that when you're thinking about the future, you gather all the information around you and do your analysis and go from there whether you're launching a new product, putting an addition onto your house or forecasting the markets. The outside view asks the question, what happened when other people were in this situation? It looks at things from a larger statistical sample. Almost always the outside view gives you an answer that's more pessimistic than the inside view. An analyst sent us a report about Amazon.com that said I think they can grow 25% a year for the next 10 years. The analyst looked at the category, at international expansion he had a story but the question is, how many companies in history have ever done that? The psychology of this is that there are all these illusions that we have, like the illusion of superiority and the illusion of control.
Speaking of which, you write that money management is one of the best examples of the illusion of control in the professional world ...
I was hoping no one would read that.
Well, honesty never goes unpunished. Could you give me a good reason then why investors wouldn't be better off with index funds?
By and large, investors would be better off with index funds. The question to me is, if you are an active manager, what makes you think you can do better than others over time? We like to think about four distinct building blocks. One is thinking about capital markets more properly. Sometimes markets are efficient and sometimes they aren't. The insight is knowing what mechanisms lead you from efficiency to inefficiency. The second is being very disciplined in valuation, in figuring out if there are differences between fundamentals and expectations. The third thing is competitive strategy work, being able to apply tools from microeconomics and strategy to figure out which companies are going to do better or worse down the road. And then finally there is all this stuff on decision-making. Very few people are willing to allocate the time to grasp the things they need to know to make good decisions consistently.
You talk about how winning strategies aren't transitive. What's the lesson for managers?
One mistake managers make is they'll read a book that says: Here are the eight things great managers do. They figure if they just live up to those attributes then they'll succeed. Well, no. It's not the attributes that matter, it's the circumstances what's going on around them. Almost all successful strategies are circumstance-based, not attribute-based. Boeing, for instance, a very smart company, had outsourced a lot of their products, but then they outsourced the design of the 787 and it was a complete disaster. The attribute was "outsourcing is good," but in this case, the circumstances didn't serve it well. The key was that you should only outsource when your product is modularized, when you can take the components and click them together.
Any other big takeaways?
A huge one is reversion to the mean [the tendency of an extreme event to be followed by a less extreme one]. Really high performance is a function of skill and luck. And luck, by definition, is transitory it comes and goes. The next time you do something, even if you sustain your skill, maybe you'll have a little less luck, and you'll mean revert. When you apply that to the world, you see it everywhere corporate performance, sports-team performance, individual performance. It gets you to think about how to evaluate outcomes. For example, when people do performance reviews for employees, what are they thinking about? Often those performance reviews focus on outcomes, where what you should try to separate is the skill component, what people can control, and the luck component wind to their back or wind to their face. For companies, there are some outliers some do better or worse for longer but it's not a large number of companies and we don't understand why. Random processes can explain an enormous amount of what we see in the real world. This idea of a best team, you should always be circumspect about that.