Sequential Decisions 517how much load should he shed: enough (70 percent) to ensure against a com-
plete disaster, or a minimum amount (25 percent)? If you were in the opera-
tor’s place, what kind of decision analysis would you undertake in 20 minutes?
In his place, what would you have done?^3(^3) This account is based on Robert Behn, “The Shed Load Decision,” Management Case(Duke
University, 1983) and “Con Ed Control Had Fifteen Minutes to Pull Switch,” The New York Times,
July 18, 1977. A short description of Con Ed’s actions and their results follows the Check Station
Answers at the end of the chapter.
Business Behavior:
Risky Decisions
Research shows that individuals have difficulties identifying and evaluating
risks. Too often they rely on intuition, rules of thumb, and experience to make
risky decisions. Managers’ most common pitfalls include:
1.Seeing too few possibilities.That is, they take a too narrow or “myopic”
view of the future. While successful firms astutely foresee possible future
consequences and act appropriately, many firms suffer losses by failing
to foresee coming events. Too often managers simply extrapolate the
current status quo into their forecasts for the future, thus ignoring
upside and downside possibilities alike. Professor Max Bazerman of
Harvard Business School calls these risks “predictable surprises” (the
disasters you should have seen coming). It is like drawing a decision
tree with whole sections of chance branches missing (because those
possibilities have been overlooked) but not knowing it.
2.Relying on verbal expressions of probability.Losing the patent case is
unlikely. There is a reasonable chancethat our product will beat our
rival’s to market. Although expressions such as these come naturally,
they are surprisingly imprecise. Researchers have asked scores of
individuals, including businesspeople, what a host of such expressions
mean in terms of probability. For instance, “unlikely” conveys a
probability of anywhere between 11 and 39 percent, with a median
response of 25 percent. In turn, “a reasonable chance” can mean a
probability as high as 80 percent or as low as 50 percent. As decision
trees remind us, determining reasonable probabilities for the risks
that matter is crucial for crafting profit-maximizing decisions. A
pessimistic view of “unlikely” could well lead to a very different
decision than an optimistic view. It’s far better to try to pinpoint and
agree on reasonable probability estimates in the first place.
3.Holding optimistic beliefs.Here, optimism means overstating the
probability of favorable outcomes and downplaying the chances of
unfavorable ones. By nature confident, many mangers unconsciously
engage in wishful forecasting: What they want to have happen they
believe is likely to happen. Clearly, overoptimistic, unrealistic beliefs
can lead to poor or even disastrous decisions. A constructive remedy
c12DecisionMakingunderUncertainty.qxd 9/29/11 1:34 PM Page 517