Lessons of the Beer Game

LESSONS OF THE BEER GAME

The beer game simulates ordering and deliveries along a supply chain. Each player manages a position with limited information about what is happening elsewhere.

• Structure Influences Behavior

Different people in the same structure tend to produce qualitatively similar results. When there are problems, or performance fails to live up to what is intended, it is easy to find someone or something to blame. But, more often than we realize, systems cause their own crises, not external forces or individuals’ mistakes.

• Structure in Human Systems Is Subtle

We tend to think of “structure” as external constraints on the individual. But, structure in complex living systems, such as the “structure” of the multiple “systems” in a human body (for example, the cardiovascular and neuromuscular) means the basic interrelationships that control behavior. In human systems, structure includes how people make decisions—the “operating policies” whereby we translate perceptions, goals, rules, and norms into actions.

• Leverage Often Comes from New Ways of Thinking

In human systems, people often have potential leverage that they do not exercise because they focus only on their own decisions and ignore how their decisions affect others. In the beer game, players have it in their power to eliminate the extreme instabilities that invariably occur, but they fail to do so because they do not understand how they are creating the instability in the first place.

People in the business world love heroes. We lavish praise and promotion on those who achieve visible results. But if something goes wrong, we feel intuitively that somebody must have screwed up.

In the beer game, there are no such culprits. There is no one to blame. Each of the three players in Senge’s account had the best possible intentions: to serve his customers well, to keep the product moving smoothly through the system, and to avoid penalties. Each participant made well-motivated, clearly defensible judgments based on reasonable guesses about what might happen. There were no villains, but there was a crisis nonetheless—built into the structure of the system.

Over the twenty years covered by this account, the beer game had been played thousands of times in classes and management training seminars. It has been played on five continents, among people of all ages, nationalities, cultural origins, and vastly varied business backgrounds. Some players had never heard of a production/distribution system before; others had spent a good portion of their lives working in such businesses. Yet every time the game is played the same crises ensue. First, there is growing demand that can’t be met. Orders build throughout the system. Inventories are depleted. Backlogs grow. Then the beer arrives en masse while incoming orders suddenly decline. By the end of the experiment, almost all players are sitting with large inventories they cannot unload—for example, it is not unusual to find brewery inventory levels in the hundreds overhanging orders from wholesalers for eight, ten, or twelve cases per week.3

If literally thousands of players, from enormously diverse backgrounds, all generate the same qualitative behavior patterns, the causes of the behavior must lie beyond the individuals. The causes of the behavior must lie in the structure of the game itself.

Moreover beer game-type structures create similar crises in real-life production-distribution systems. For instance, in 1985, personal computer memory chips were cheap and readily available; sales went down by 18 percent and American producers suffered 25 to 60 percent losses.4 But in late 1986 a sudden shortage developed and was then exacerbated by panic and overordering. The result was a 100 to 300 percent increase in prices for the same chips.5 A similar surge and collapse in demand occurred in the semiconductor industry in 1973 to 1975. After a huge order buildup and increases in delivery delays throughout the industry, demand collapsed and you could have virtually any product you wanted off any supplier’s shelf overnight. Within a few years, Siemens, Signetics, Northern Telecom, Honeywell, and Schlumberger all entered the business by buying weakened semiconductor manufacturers.6

In mid-1989, General Motors, Ford, and Chrysler, as the May 30 Wall Street Journal put it, “were simply producing far more cars than they were selling, and dealer inventories were piling up … The companies already are idling plants and laying off workers at rates not seen for years.”7 Entire national economies undergo the same sorts of surges in demand and inventory overadjustments, due to what economists call the “inventory accelerator” theory of business cycles.

Similar boom and bust cycles continue to recur in diverse service businesses. For example, real estate is notoriously cyclic, often fueled by speculators who drive up prices to attract investors to new projects. “The phone would ring,” Massachusetts condominium developer Paul Quinn told the MacNeil-Lehrer Newshour in 1989, “in our offices, and we said ‘How are we going to handle this? We’ll tell everybody to send in a $5,000 check with their name and we’ll put them on the list.’ The next thing we knew, we had over 150 checks sitting on the desk.” The glut followed quickly on the boom: “It was a slow, sinking feeling,” Quinn said, interviewed in a seaside town full of unsold developments. “Now’s the time to start building for the next boom. Unfortunately, the people in the real estate industry are too busy trying to address the problems they have left over from the last one.”8

In fact, reality in production-distribution systems is often worse than the beer game. A real retailer can order from three or four wholesalers at once, wait for the first group of deliveries to arrive, and cancel the other orders. Real producers often run up against production capacity limits not present in the game, thereby exacerbating panic throughout the distribution system. In turn, producers invest in additional capacity because they believe that current demand levels will continue into the future, then find themselves strapped with excess capacity once demand collapses.

These production-distribution dynamics illustrate a principle of systems thinking: structure influences behavior. Ordering rules and delays can generate a crisis through the interaction of decisions that seem reasonable to each participant.