Rational decision making proceeds on the belief that managers can transform a complicated web of facts, assumptions, objectives, and educated guesses into a clear decision that people at the organization can act on. There is a strong faith in all this that the world can be influenced through manager’s mental capabilities. Decision making, then is an effort to exercise control over the organization’s destiny. This has been a distinctive management belief for more than a century.
This faith has been challenged in recent years. More and more, an adaptive approach to decision making has emerged as a way to think about what managers can and cannot expect to accomplish. This adaptive approach turns on the assumption that the link between an organizational action (the result of a decision) and the consequences of that action is far messier and far more unpredictable than rational decision makers believe. According to adaptive thinking, the results of a decision action are jointly produced by what your organization does and what other organizations are doing at the same time.
A classic example of a jointly-produced outcome has been occurring in recent years in the U.S. passenger airline industry. Airlines often fly relatively empty planes. Yet airline companies have to pay for expensive jet aircraft and for large numbers of highly-trained employees whether flights are full or not. So for any one airline, it might make rational sense to cut air fares to induce more passengers to fly on that airline. What happened in the spring of 1992, however, is that managers at several leading airlines-including American, United, Delta, and Northwest-all decided to cut fares at the same time. But there was not sufficient growth in the total number of passengers flying to make up for the forgone revenue from lower airfares. The jointly-produced result of these separate, rationally-derived decisions at each individual company was a so-called “price war” in which every airline suffered financial losses.
Two versions of the adaptive approach to decision making are game theory and chaos theory. Game theory is the study of people making interdependent choices. A game is a situation involving at least two people in which each person makes choices based, in part, on what he or she expects the other to do. Game theory highlights the explicit role of human relationships and interactions in decisions.
Games were first used in planning nuclear armament build-ups after World War II. Currently, game theory is used in such business decisions as competitive pricing. We experience games all the time. Two people must “play” a game when they meet at a revolving door at the same time, or when one is driving on a freeway and another attempts to enter the flow of traffic. For our understanding of decision making, the key point is that in a game the outcome is jointly produced. We make it through the revolving door, or we get delayed, because you do something and I do something.
Hence, a game theory perspective requires that we view decision making as a process of two decision makers adapting to each other’s presence at the same time. Each can decide rationally but also adaptively. This is what we were just describing about the modern U.S. airlines industry.
Chaos theory is the study of dynamic patterns in large social systems. Chaos theory thus is a descendant of systems theory. Chaos theorists pay close attention to the turbulence of a system. Under conditions of turbulence, not only is the future completely unpredictable, but present circumstances are likely unstable, too. In this way, turbulence differs from certainty, risk, and uncertainty, which are comparatively stable conditions under which managers can at least choose an attainable objective.
Chaos is a pattern of three states: equilibrium, disequilibrium, and bounded instability. The task of decision makers is to keep the organization in the third state because that is where organizations can innovate. Decision making becomes a continual process of adaptation to forces largely beyond a decision maker’s control.