Prospect theory is a concept from behavioral economics related to framing, that explains why humans do not always behave rationally when making economic decisions. It states that humans value a certain gain more than a probable gain of equal economic value, and prefer probable loss to a certain loss of equal economic value.

Q. This term is suddenly cropping up at CIO conferences… what’s it mean?
A. It’s a favorite idea of behavior economics that’s picking up a lot of steam lately. One of the reasons its interesting is that it lays down a fundamental challenge to the classic Chicago school of rational investors and efficient markets. The best way to illustrate is to ask a couple of questions, OK?

Q. OK! I think we have these up on a screen for the viewers:
PROBLEM 1: Which do you choose?
Get $9,000 for sure OR
90% chance to get $10,000

PROBLEM 2: Which do you choose?
Lose $9,000 for sure OR
90% chance to lose $10,000

A. So nearly everybody takes the first option in the first question, and the second option in the second. It’s really the same choice in both, of course. We choose different options because one is a gain situation, and one is a loss situation. We love sure things, and we hate losses.

Q. So we don’t value gains and losses equally. And I guess that can lead to some poor investment decisions.
A. Yes! We’re simply wired to be too conservative when facing positive outcomes, and too aggressive when facing negative ones. One very simple consequence is that investors take gains too early but let losses run, because they want to get back to level before they sell a stock. Anything to avoid recognizing that loss!

Q. And how does all this play in the professional investing world?
A. Interestingly, studies show that professional traders are less subject to this bias… they’ve learned better. So it’s not hopeless, but it does illustrate an important idea for less experienced investors. Also, as awareness of these kinds of behavioral biases grow, institutional investors are trying to move towards building internal decision making processes that take all this into account. More process, less intuition.