Risky Business

I ran across this great Helen Keller quote:

"Security does not exist in nature nor do the children of men as a whole experience it. Avoiding danger in the long run is no safer than exposure."

Recently, I have found myself in two rather lengthy discussions (translation: I started two arguments) about how people’s inability to gauge risk leads to irrational behavior—or, at least, irrational justification for their decisions.

My basic argument is this: Ultimately, good decisions will be based on preferences over outcomes and objective, precise measurements of the probability of those outcomes. People tend to be quite risk averse, but they are also very bad at measuring both absolute and relative risk. If mistakes are random, then this just leads to random error in decisions…no big deal so long as the error is reduced to a minimum. But if the measurement of risk is driven by a personal bias such as fear, then the error will be systematic. In other words, you will be making bad decisions.

Now, one reply is that fear is subjective, but it is a real emotion and therefore a meaningful barometer for risk. This, I think, is either sloppy reasoning or sloppy language and both lead to bad decisions. Ex ante risk and probability are known—or at least knowable—so disagreements should be minor. (At a bare minimum, two people should be able to agree on relative risks.) When someone expresses "fear" as justification against something, what they are really revealing are preferences, not probabilities. Consider the following:

• Suppose someone opposes traveling to foreign country X because they might become a victim of crime or terrorism when, in fact, the likelihood of something bad happening to them is probably greater in parts of some U.S. city they have been to (or some remote part of the U.S. wilderness). What this person is really saying is that travel to country X just isn’t that appealing to them for whatever balance of reasons including, but not especially, danger.
• Suppose someone opposes adventure activity Z (skydiving, rock climbing, etc.) because they might get hurt. True, they might get hurt, but that same person engages in a number of routine activities with a larger probability of injury (e.g. driving). What they really mean, is activity Z just doesn’t sound like very much fun even when the risk is small.

One possible explanation is that if you have a strong taste for the known, then you may be willing to literally risk your life for peace of mind. I think this is what we observe in human nature. But it seems really perverse to let fear be such a dominate force in your decision making that you are willing to accept greater real risk to reduce falsely perceived risk.

Admittedly, feeling afraid sucks, but if we get in the habit of couching decisions and measuring probabilities in terms of fear then we risk making systematically biased and bad decisions. That, is a risk I suggest eliminating.

Update: If interested, check out this brilliant podcast with Cass Sunstein on worst case scenarios. (And by "brilliant" I mean he supports my argument.) Among other things, he discusses climate change, terrorism, healthcare and shark attacks.

Malcolm Gladwell's book "Blink" also talks about this idea, but more in the context of instinct. He basically presents several exceptions to the idea that well-informed, well-reasoned decisions lead to the best outcome. He concludes that we should trust instinct and decisions made in the blink of an eye more often--a conclusion I reject. Finding exceptions in this case does not disprove the rule. Rather, it proves that the outcomes were just probabilistic. Showing that one person always ends up winning the lottery does not make playing the lottery less of a gamble. Pointing to one calamitous event does not make that event more likely to happen.