Two new research techniques in economics appeared in the post-World War II era. Macroeconomic models of the economy were developed as a new tool for economic forecasting. Consisting of hundreds of equations that could be solved simultaneously using the new high-speed computers, these models are used for economic forecasting by both governments and the private sector, predicting such economic variables as total output, consumer spending, changes in the general price level, employment and unemployment, output of major industries, exports and imports, flows of capital— indeed, any variable for which the model has an equation.
Macroeconomic models were originally based on Keynesian theory. They include equations such as Y = C + S (national income equals consumption spending plus savings) and identities such as S = I (savings equal investment). Other equations and identities define the component parts of these simple relationships until a highly complex, multiple-equation model is produced. The theoretic model is then tested and revised using actual economic data, until reliable forecasts are obtained.
Early steps toward development of macroeconomic models were taken by Jan Tinbergen (born 1903), a Dutch economist whose two-volume Statistical Testing of Business Cycle Theories (1939) used a system of equations based on data from the American economy for the years 1919-1932 to evaluate theories of business cycles. Laurence Klein (1920- ) began to experiment
with multi-equation econometric models in the late 1940s. Under his direction, the Research Seminar in Quantitative Economics at the University of Michigan produced the first economic forecast from such a model in 1953.
The Michigan model was initially quite simple, with just sixteen equations and four identities, using data for the period 1929-1952, leaving out the war years of 1942-1945. Since then the model has grown to 291 equations and identities. The latest available data are used to make both quarterly and annual forecasts, and the entire model is updated ("reestimated," to use the econometrician's lingo) every five years. The Michigan model, and the economists who use it, produce some of the most accurate of the multitude of economic forecasts that are made every year.
Klein left Michigan in 1954, a victim of the McCarthy era witch-hunts during the Cold War. He continued his work on macroeconomic models soon after at the University of Pennsylvania, where he and his assistants developed an even larger model than the one at Michigan. Klein received the Nobel Award in Economic Science in 1980.
The success of the Michigan model stimulated the development of other large-scale macroeconomic models in the 1960s and 1970s. One model at the Brookings Institution in Washington, D.C., had almost 400 equations but didn't work well and was abandoned in 1972. Models were built not only in the United States, but in many other countries throughout the world, in advanced industrial nations, developing countries, and centrally planned economies, and for systems of world trade. By 1990 there were eleven such models in the United States alone. Macroeconomic models have become important tools in the planning of both government policies and business decisions.
Survey research was the second economic research instrument developed in the years immediately following World War II. An interesting problem arose in 1944 and 1945 when it became clear that Germany and Japan would be defeated. Economists in Washington were worried that consumers would cash in their war bonds as soon as the war ended, spend the money, and set off a rapid inflation before the economy could convert to peacetime production. A Division of Program Surveys was established in the Department of Agriculture to find out. A psychologist, George Katona (1901-1981) was brought in to run the project.
Katona was born in Budapest, Hungary. He was a student at the University of Budapest when the communists seized power and closed the university briefly in 1919. Fleeing to Germany, he continued his studies in psychology, taught, worked in a bank, and wrote for a Berlin newspaper. When the Nazis suppressed the newspaper in 1933 Katona came to the United States and became a citizen. A bout of tuberculosis ended a budding career as an investment advisor, and he turned back to psychology. A pathbreaking book. Organizing and Memorizing (1940) showed that organizing material in patterns (gestalts) made it easier to remember and to apply the material to new situations. Another book. War Without Inflation: The Psychological Approach to Problems of a War Economy (1942), emphasized the role of consumer behavior in the economy and brought him to the attention of government officials.
Katona and his associates asked a random sample of several thousand consumers representative of the population as a whole about their expectations and their spending plans. Statistical probability analysis was used to
evaluate the results. The researchers found that most consumers intended to keep their bonds as a reserve in the event that peace brought lower earnings or unemployment.
Katona went on to almost single-handedly develop survey research as an important tool of empirical research in economics and the other social science disciplines. Today survey research is carried on all over the world by universities, research institutes, governments, and private enterprises.
After the war Katona took his entire survey research group from the Department of Agriculture to the University of Michigan where in 1946 he established a Survey Research Center as part of the University's Institute for Social Research. There he and his colleagues, economist James Morgan and John Lansing, statistician Leslie Kish, and others, continued to develop the theory and practice of survey research, building up a huge bank of data on household income, savings, and expenditures. One of their innovations was the use of "longitudinal" surveys that follow cohorts of households over long periods of time to determine how they are affected by changes in the larger economy. The center continues today as an important source of empirical data for applied research in economics and the other social science disciplines. One of its most important findings is that consumer decisions are based not only on price and income but more importantly on their expectations and their understanding of the larger condition of the economy. Despite fads and changing fashions, consumers are highly rational in managing their affairs in the changing and sometimes chaotic modern world.