Risk Tolerance and Risk Neutrality (You Can Live With More Risk Than You Think)

Next week, I’ll be embarking on my first tour as part of the Society of Petroleum Engineers’ Distinguished Lecturer program. I’ll be doing six tours over a period of six months, each of which involves presenting at between three and five different locations – twenty-four stops in all. Three of the tours are in North America, and three are international. It will be hectic, but a lot of fun, too.

My presentation is all about risk management – specifically, why it is important to apply one’s risk tolerance at the portfolio level, rather than at the individual project or asset level. When evaluating individual projects or assets, a risk-neutral approach is most appropriate. “Risk neutrality” means that the economic value one places on an opportunity is equal to that opportunity’s “Expected Value,” or EV. “Expected Value” is a terrible phrase because there is absolutely no reason at all to expect it, but it is commonly used, so I will use it here. Simply, the EV = [the probability of success times the mean value of success] + [the probability of failure times the mean cost of failure]. This may sound like common sense, but unfortunately, it is not the way human beings think.

If I offered you a choice between a 95% chance at winning $1000 or a sure-fire win of $900, the vast majority of you would take the $900, even though the EV of the first opportunity is greater ($950 vs. $900). This is called being “risk averse” – you are willing to give up some EV in order to avoid uncertainty.

Conversely, if I offered you the choice between a 95% chance of losing $1000 or a sure-fire loss of $900, the majority would take the chance (even though the expected value of the uncertain outcome is -$950, vs. -$900 for the sure thing). This is called being “risk seeking” – you are willing to give up some EV in order to have a chance of avoiding a sure loss.

If these were one-off opportunities, you would be well within your rights to decide for yourself which of the two alternatives you would prefer. Our preferences are not just determined by numbers; personal risk tolerance and risk preference come into play.

But businesses fund a large number of projects (a portfolio), and they have a fiduciary responsibility to their shareholders to try to maximize the value of the enterprise. One can easily see that in the long run, if Company A consistently takes the higher EV alternatives regardless of risk (the 95% chance at winning $1000 and the sure-fire $900 loss) and Company B consistently takes the “psychologically natural” alternatives (the sure-fire $900 and the 95% chance of losing $1000), Company A will outperform Company B. Company A is behaving in a risk-neutral manner at the project level; Company B is applying its risk tolerance at the project level.

And things get more complicated still. If I change the probabilities associated with the alternatives from highly likely (95%) to highly unlikely (2%), most people’s decision preferences will change.

For instance, if I offered you a choice between a 2% chance of winning $1000 or $40 for sure, most people would take the chance (even though the EV of the uncertain opportunity is only $20). And if I offered you a choice between a 2% chance of losing $1000 or a certain loss of $40, most of you would simply pay the $40. We become risk-seeking when looking at the potential gain (where before we were risk-averse), and risk-averse when considering the possible loss (where before we were risk-seeking).

Nobel laureate Daniel Kahneman does a far better job than I can of explaining the psychology behind these human foibles in his recent book, Thinking Fast and Slow. Kahneman and a number of his colleagues (most famously, Amos Tversky) have conducted hundreds of experiments over the past few decades which document how poor human beings are at making decisions in the face of uncertainty. This research has formed the foundation of Behavioral Economics, a field of study that is all the rage in economic circles these days. Classical economic theory assumes that, given a choice and sufficient information, people will make decisions that are in their own best interest. This is the foundation of the notion of a rational free market. Behavioral Economics explodes this myth.

A number of psychological characteristics contribute to the phenomena I’ve described, including fear of regret (“I could have had $900 for sure but I got greedy!”) and hope springing eternal, despite the odds (“Why should I take a measly $40 when I have a chance to win $1000?”). These cause all kinds of problems in business, as anyone who has ever tried to kill a bad project can confirm. As long as there is any chance at all of turning the project around and avoiding a loss, people will cling to that hope and continue to throw good money after bad, long after they should have written off their losses.

Even when one is in the happy situation of choosing between alternatives with positive values, businesses often yield sub-optimal results by applying their risk tolerances at the project level. The natural human desire to stamp out all risk leads many companies to enact arbitrary thresholds for projects to meet (e.g., no more than a 15% probability of losing money). They will spend millions of dollars to acquire information (imperfect information, I might add), which may reduce the probability of loss from, say, 18% to 13%, thereby crossing the magic, completely arbitrary threshold. This is a waste of their shareholders’ money.

Now let me be clear: being risk-neutral does not mean being reckless.

>It may very well be that the sensible business decision is to acquire more information (even imperfect information) before pulling the trigger on a major investment. But that decision should be based on a value-of-information (VOI) analysis, not on some arbitrary risk tolerance threshold. VOI analyses have been around for a very long time and are excellent for assessing whether your project is more valuable when you acquire additional information before deciding what to do, or more valuable when you simply make that decision now, despite the uncertainty. If you’re risk-neutral (and you should be), you will opt for the route with the higher value, uncertainty be damned. Note that you are not ignoring uncertainty; the calculation of EV incorporates the chances of success and failure. Being risk-neutral just means not taking any further account of uncertainty or risk when making your decision.

One last point: this essay deals only with how to properly take risk and uncertainty into account when making decisions at the project level based on economic analyses. This is a necessary, but not sufficient, step toward proper portfolio management. Managing a portfolio of assets involves far more than just rolling up the numbers from the individual assets. One must consider how well various alternative portfolios meet one’s strategic objectives, one must balance long-term value against short-term budgets, one must look at portfolio cash flows through time, and yes, one must apply one’s risk tolerance. The portfolio level is the appropriate place to do this.