Risk Aversion

"I'm always thinking about losing money as opposed to making money. Don't focus on making money, focus on protecting what you have." - Paul Tudor Jones, Tudor Investment Corporation

It’s common sense to believe that avoiding risk and limiting loss is good and that we make conscious, logical decisions to do just that. But, we may not be as rational as we think, according to Nobel Prize winner Daniel Khaneman and his associate Amos Tversky.

Indeed, Khaneman and Tversky found that economic behaviors were inconsistent with rational behavior. They discovered that investors feel more pain from losing a dollar than they feel happiness from gaining that same dollar. In other words, making money does not bring with it enough happiness to offset the pain and disappointment of losing money. This is known as loss aversion.

Status Quo Bias and Risk Aversion

If you experienced a financial windfall of $1 million and a casino offered you the opportunity to keep the money or flip a coin on a double-or-nothing bet, what would you do?

There is extensive empirical evidence that says you would decline the bet and not lose a moment’s sleep over it. In “proving” that most people would be unwilling to gamble away a sure thing—a conclusion that sounds right even to those of us who have never had the chance to carry out our own research in behavioral finance—two well-known economic principles are revealed—status quo bias and risk aversion. In other words, most people assume that the way things are is better than the way they will be if circumstances change. Furthermore, most people have little interest in assuming risk, even when the risk/reward ratio is in their favor. 

The Experiment

Therefore, it is surprising to find out that when Khaneman modified the circumstances of this very question, but not the outcome, the vast majority of respondents reacted much differently than one might expect.  In their initial study, they posed two scenarios:

Scenario 1 - Choose between the following:

1. You have a 100% chance of winning $3,000, or
2. You have an 80% chance of winning $4,000 and a 20% chance of winning nothing

Results:

80% of the respondents chose the certain $3,000, notwithstanding the fact that they had an excellent chance to win significantly more.  A textbook case of status quo bias and risk aversion, this outcome is in line with traditional expectations.  

Then, however, the subjects were presented with a different scenario, and the results were diametrically opposite to traditional expectations.

Scenario 2- Choose between the following:

1. You have a 100% chance of losing $3,000, or
2. You have an 80% of losing $4,000 and a 20% chance of breaking even

Results:

92% of the respondents chose the second answer—to risk losing $4,000 or trying to break even—even though they had the chance to lock in a significantly lower loss ($3,000), and it was highly likely (4:1) that by taking the gamble they would end up losing considerably more. 

What is most surprising about the outcome of this experiment is that while the verbiage used in Scenario 2 suggests it is a different case than Scenario 1, in fact, it is identical in terms of the risk undertaken.

Both cases present a choice of a sure thing to be weighed against a gamble. On the surface, the cases seem dissimilar because one offers the great likelihood of walking away a winner and the other offers, at best, only a small chance of breaking even. Nonetheless, in both cases, the mathematical expectation of assuming the gamble is $3,200 ($4,000 x 80%).

In the first scenario, where one is likely to be rewarded for taking the gamble, the vast majority was more concerned about averting the 20% chance of getting nothing. In the second scenario, however, in which one is likely to be punished for gambling, all but 8% of the respondents chose to put themselves in harm’s way, ignoring the status quo and risking an uncertain outcome of dubious potential benefit.