From Max Bazerman’s “Judgement In Managerial Decision Making”
Acquire Expertise (P. 223)
Many of the biases we have examined in this book were identified in experiments with student participants who were not rewarded for accurate performance and who were making decisions in task domains unfamiliar to them. Thus, one optimistic possibility is that experts or experienced decision makers facing important real-world decisions might be far less affected by biases than most research participants. Does this book unfairly exaggerate the prevalence of judgment biases?
This is certainly an important question, since experience and expertise might be useful tools for improving decision making. Some researchers believe that the process of improving judgment will occur naturally as individuals receive feedback about their past decisions. This view is represented by Kagel and Levin (1986, p. 917) in their analysis of the winner’s curse in competitive bidding discussed in Chapter 4: Given sufficient experience and feedback regarding the outcomes of their decisions, we have no doubt that our experimental participants, as well as most bidders in “real world” settings, would eventually learn to avoid the winner’s curse in any particular set of circumstances. The winner’s curse is a disequilibrium phenomenon that will correct itself given sufficient time and the right kind of information feedback. In fact, Kagel and Levin (1986) do show a reduction in the winner’s curse in the auction context as the market (but not necessarily specific players) “learns” over time. However, much of this learning can be attributed to the phenomenon in which the most aggressive bidders go broke and drop out of the market. Additional learning occurs by observing the consistent losses being suffered by “winners” in the auction.
Clearly, life experiences help us to improve numerous skills and abandon many bad habits. Unfortunately, our judgmental distortions might not be among them. Tversky and Kahneman (1986) have argued that basic judgmental biases are unlikely to correct themselves over time. Responsive learning requires accurate and immediate feedback, which is rarely available in the real world because: (i) outcomes are commonly delayed and not easily attributable to a particular action; (ii) variability in the environment degrades the reliability of feedback . . . ; (iii) there is often no information about what the outcome would have been if another decision had been taken; and (iv) most important decisions are unique and therefore provide little opportunity for learning (see Einhorn and Hogarth, 1978) . . . any claim that a particular error will be eliminated by experience must be supported by demonstrating that the conditions for effective learning are satisfied.
Even if accurate and immediate feedback is available in a given situation, we face another crucial challenge: we are likely to misremember our own forecasts (Meyvis, Ratner, & Levav, 2010). We often anchor to current states and fail to accurately recall our prior predictions. Thus, it is common for us to underestimate the extent to which our prior predictions deviated from actual outcomes, and this underestimation leads to us inadequately learn from prior experience (Morris &Moore, 2000).
Using the “Acquiring a Company” problem described in Chapter 4, Ball, Bazerman, and Carroll (1991) tested the ability of individuals to learn to avoid the winner’s curse by incorporating the decisions of others into their decision making. Participants in this experiment played for real money, played in 20 trials, and were given full feedback immediately after each trial based on a random determination of the value of the firm up for sale; in addition, they could observe changes in their asset balance (which virtually always went down). Thus, when compared to the limitations cited by Tversky and Kahneman, ideal conditions existed for learning from past mistakes. The only limitation that was not eliminated—namely, the variability of the environment (ii above)—is a natural part of the winner’s curse phenomenon. Thus, we were able to look at whether or not the ability to consider the cognitions of the other party in a bilateral negotiation problem can be learned in a highly favorable environment.
Remembering that $0 is the correct answer and that $50 to $75 is the answer typically obtained when decision makers ignore the cognitions of others, examine the mean bids across the 20 trials in Figure 12.1. Across the 20 trials, there is no obvious trend indicating that participants learned the correct response. In fact, only five of 72 participants from a leading MBA program learned over the course of the trials. Our general conclusion? Individuals are unlikely to overcome the winner’s curse simply through experience or feedback
This evidence paints a pessimistic picture of the idea that experience will cure the decision biases identified in this book. In fact, Bereby-Meyer and Grosskopf (2008) documented that even hundreds of trials do not lead most study participants to solve the Acquiring a Company problem. This evidence is consistent with the documentation of extensive bias in decision making by actual investors, real-estate agents, medical doctors, and numerous other “expert” groups. Neale and Northcraft (1989) proposed that biased decision-making outcomes could be eliminated or ameliorated through the development of expertise. While we often think of experience and expertise as closely related, Neale and Northcraft defined experience simply as repeated feedback. By contrast, they assert that expertise results when individuals develop a “strategic conceptualization” of what constitutes a rational decision-making process and learn to recognize the biases that limit rationality.
Neale and Northcraft’s experience/expertise distinction is highly relevant to the question of whether or not experienced decision makers can benefit from the study of decision making. Northcraft and Neale’s (1987) study of anchoring and adjustment among real-estate agents suggests that experienced decision makers can be very biased. In addition, while most “effective decision makers” are successful in a specific domain, experience without expertise can be quite dangerous when it is transferred to a different context or when the environment changes. Evidence from Chapter 2 suggests that as the amount of ignorance increases, individuals become more overconfident regarding their fallible judgment.
If you think that experience should help negotiators do a better job of understanding the other side’s reservation price, think again. Larrick and Wu (2007) find that, when it comes to estimating the size of the bargaining zone, experience will only help us correct one type of error: overestimation of the bargaining zone’s size. When you think the bargaining zone is much bigger than it is, your negotiating counterpart will help you identify and correct your error by refusing to agree to deal at the price you propose. When, on the other hand, you underestimate the size of the bargaining zone, you will end up offering the other side more than was necessary. Though she probably will be anxious to accept your offer, she may try to get you to concede a bit more first, so that you will think that your offer is close to her reservation price. This type of experience will generally lead negotiators to believe that bargaining zones are smaller than they actually are and that they need to make more generous offers to their negotiating opponents.
Stressing the drawbacks of relying on experience for knowledge, Dawes (1988) notes that Benjamin Franklin’s famous quote “experience is a dear teacher” is often misinterpreted to mean “experience is the best teacher,” when in fact Franklin was using “dear” as a synonym for expensive. After all, the quote continues, “yet fools will learn in no other [school].” Dawes writes,
Learning from an experience of failure . . . is indeed “dear,” and it can even be fatal. . . . moreover, experiences of success may have negative as well as positive results when people mindlessly learn from them. . . . People who are extraordinarily successful—or lucky—in general may conclude from their “experience” that they are invulnerable and consequently court disaster by failing to monitor their behavior and its implications.
Or in the words of Confucius: “By three methods we may learn wisdom: First, by reflection, which is noblest; Second, by imitation, which is easiest; and third, by experience, which is the bitterest.”
This view of experience reiterates the comparative value of gaining a conceptual understanding of how to make a rational decision rather than simply depending upon the relatively mindless, passive learning obtained via experience. Expertise requires much more than the unclear feedback of uncertain, uncontrollable, and often delayed results. Rather, it necessitates constant monitoring and awareness of our decision-making processes. The final benefit of developing a strategic conceptualization of decision making concerns transferability. If you ask experienced decision makers for the secrets of their success, they routinely insist that their skills have developed over years of observation and experience that cannot be taught. This obviously reduces their ability to pass on their knowledge to others. Thus, experience without expertise limits the ability to transfer knowledge to future generations. A key element of developing a strategic conceptualization of decision making is to become aware of the many biases in individual and group contexts that we have discussed in Chapters 1 through 11. However, awareness is just one step in the process. Another strategy, debiasing, is the topic of the next section.