Last month I interviewed psychologist Daniel Kahneman, who won the Nobel Prize in economics in 2002 and recently authored the book Thinking, Fast and Slow.
In this clip, Kahneman and I discuss why losing money hurts more than making it, a concept called "loss aversion." (Transcript follows.)
Dr. Kahneman: Well, why it is, I think, we've called loss aversion probably has evolutionary roots. There is an asymmetry between threats and opportunities. That is if it's a severe threat, it has higher priority than opportunity, assuming that there'll be more opportunities than the threat is more urgent. So that asymmetry is probably built in. There is an asymmetry I think between pain and pleasure and in general pain is a more urgent signal than pleasure. They both are signals, but one is more urgent than the other.
Now is it rational? Well that's a really complicated issue, because if you have an individual whose objective is to maximize wealth at a certain future point in time, then loss aversion is very bad because loss aversion will cause that individual to miss out on many opportunities. But if it is rational to live with whatever your nature is and to try to enjoy life as much as possible, then loss aversion is just a fact of life, like regret. You may think that regret is a foolish emotion, but if you know that you're going to be susceptible to regret, it is not irrational to anticipate it and to act accordingly.
Morgan Housel: So we were talking about overconfidence and how we're talking about loss aversion. Those seem to conflict a little bit in that people are optimistic in the risks that they take, but then they have a loss aversion when any risk shows up. Those seem like very conflicting views. How do they mesh together?
Dr. Kahneman: Loss aversion is on the value side, and optimism is on the odds side and the judgment side. You need both value and judgment to lead to a decision. The output of those two biases, they do tend to work in opposite directions, but there is absolutely no guarantee that the output is the correct decision that is unbiased, but they do work against each other. And the result is that people take risks that they would not take if they knew the real odds. That's one implication.
And another implication is that it's dangerous to fiddle with one of these biases if you're not fixing the other at the same time, because you could generate too much optimism or too much loss aversion, if you fix the other bias but not that one.
Morgan Housel: So the balance between optimism and loss aversion is one about calculating the odds of a future event happening. What is the evidence that we are good or bad at calculating those odds?
Dr. Kahneman: Well there is massive evidence actually that people are not very good at predicting the future outside some domains, you know. They can predict the patterns in their daily life that are stable, but when it comes to predicting future changes, people are really not very good at it. And they are overconfident because they see patterns, and they see only one pattern. We generally see the world as much simpler than it is.
The evidence is almost too massive to list here, but some of the nicest evidence of overconfidence and predictions come from the study of chief financial officers in large corporations that was published a few years ago. They were required once a year, a lot of them, I forget the exact number now, and they were required to predict what would happen to the S&P 500 over the following year. The correlation between these predictions and what actually happens is slightly negative. People had absolutely no idea what's going to happen, but they're very overconfident. That is people are asked to set an 80% confidence interval so that if they do it right, they shouldn't be surprised 20% of the time. But in fact, they were surprised about two-thirds of the time, so there's huge overconfidence.