A Business Cycle
Economic "crises" occurred throughout the 19th century. By the latter part of the century, a number of observers had come to the conclusion that there was a regular, wavelike pattern to crises, and they called this pattern the business cycle. A cycle with its regular and periodic fluctuations needed an explanation. What caused the pattern? What forces determined the length and amplitude of the waves? Could the cycle be eliminated?
Research on business cycles expanded rapidly early in the twentieth century, but receded after the 1930s. Although the peak in study of the cycle has passed, many of the ideas developed during the boom years of business-cycle research remain as major ideas in macroeconomics.
Although a large number of people once believed that a business cycle existed, they agreed on little else. The literature is notable for the wide variety of contradictory ideas it harbors and for the lack of commonly-accepted terms to describe basic ideas. Also, many writers were ambiguous and unclear in what they wrote, and there is sometimes a tendency to read more into what they wrote than they may have meant. Two very good summaries of the literature are Gottfried Haberler's Prosperity and Depression (Geneva: League of Nations, 1941) and Robert A. Gordon's Business Fluctuations (New York: Harper, 1952). They form an interesting contrast not only because they group theories differently, but also because their conclusions about what was valuable and what was nonsense in the literature are very different.
The picture below illustrates a framework in which business-cycle theories were often constructed. Starting at point A, the economy grew very rapidly in a self-feeding growth. Because of this "self-feeding" process, the economy would develop momentum. Once started, growth would continue until the system hit a limit or boundary that would stop it. The economy would then turn around and enter a contraction that was also self-feeding. This downswing would continue until a lower boundary was encountered, which would stop the contraction and start a new upswing. Notice that in this view the economy is never in equilibrium, but perpetually adjusting.
Although many business-cycle theorists accepted this basic picture, they differed on what provided the upper and lower bounds and why the self-feeding process would take place. Sometimes their differences were matters of emphasis. What one took as part of the institutional structure or as a "given," another would emphasize as the central causal force. To use an analogy, suppose two people are trying to explain why a windmill is turning. One argues that it is the wind that causes everything and the other argues that the rotation depends entirely on the pitch of the blades. This disagreement would be more apparent than real. They differ in what factors should be taken as given and which to consider as causal. One person takes the wind as given and then examines the structure of the machine, while the other accepts the machine as given and looks instead at outside forces.
In the business-cycle framework illustrated above, lines of causation are difficult to test because of feedback. To see this, consider a simple cycle, the temperature in a house heated by hot-water radiators. When it is cold outside, the rooms lose heat. Eventually the house reaches a temperature so that the thermostat lights the boiler and starts water circulating to the radiators. This does not immediately stop the decline in temperature because it takes time for the heat to be transferred from the fire to the water to the radiator to the air near the thermostat. Nor does the furnace shut off once the decline has stopped. Rather it continues to heat water until the temperature of the air at the thermostat has reached a shut-off level. After the furnace has shut off, the temperature in the house will continue to rise for a while because at the time of shut off the water in the radiators had been heated to its highest value.
It is clear that the behavior of the furnace determines the temperature in the house. It is equally clear that the temperature in the house determines the behavior of the furnace. Systems in which variable A determines variable B, but in turn variable B determines variable A contain feedback. Systems with feedback can be difficult to explain if only their outcomes are available. For example, suppose someone who had no idea of how heating systems worked was given a chart of temperatures in the house and the times when the furnace was on. Perhaps the most obvious relationship in these data would be that when the house was coldest, the furnace was on. A person with no knowledge of the system might conclude that the furnace made the house cold, and turning it off heated the house. Unless one knows what to look for, or when one looks for the wrong things, strange conclusions can emerge from studies of complex systems.
Not all economists working in the late 19th and early 20th centuries accepted the business-cycle framework, and some business-cycle theories do not fit it. Irving Fisher is an example of an economist who had a theory of business fluctuations, but had serious reservations about cycles. Although he agreed with many of the ideas mentioned above, he stressed the differences of each episode, and did not believe a regular, periodic cycle existed. Other economists, such as Joseph Schumpeter, had much more elaborate frameworks. Schumpeter thought there were three cycles within the economy: a long, 60-year cycle, a moderately long, 10-year cycle, and a short, 40-month cycle.
The best examples of theories of business cycles that do not fit the above framework are theories relying on sunspots or other astronomical phenomena. Several prominent economists developed these theories in the 19th and early 20th centuries. According to sunspot theories, changes in sunspots caused changes in harvests, which in turn affected the overall performance of the economy. Perhaps the weirdest of these theories was that of Henry L. Moore, an American economist who was an early user of mathematics. Moore, after considerable statistical work, concluded that there was an eight-year cycle and attributed it to the planet Venus that every eight years passes directly between the earth and the sun. None of these theories involved feedback, and they did not have self-feeding processes that some boundary stopped. All have predicted so poorly that today they are as dead as economic ideas ever get.
Feedback is an important concept, and we can look at it in more depth.
Copyright Robert Schenk