Presentation Speech by Professor Karl Gustav Jöreskog of the Royal Swedish Academy of Sciences, December 10, 2000.
Translation of the Swedish text.
Copyright © Nobel Media AB 2000
Photo: Hans Mehlin
Your Majesties, Your Royal Highnesses, Ladies and Gentlemen,
What determines whether an individual chooses to work and, if so, how many hours? How do economic incentives affect individual choices of education, occupation, place of residence and travel mode? What are the effects of labor market programs and educational programs on income and employment? Reliable answers to such questions require access to microdata, i.e., detailed data on the economic behavior of individuals. This year’s Laureates in Economic Sciences have developed methods for solving fundamental problems arising in the analysis of microdata. James Heckman and Daniel McFadden have made pioneering research contributions to microeconometrics, an area in the borderland between economics and statistics. Their methods have become standard tools of microeconometric research in economics as well as in other social sciences, and have been applied to solving many important problems in society.
Social scientists cannot, as do natural scientists, assume that the effects of a given intervention or treatment on certain individuals will generalize to other individuals. Unlike atoms and molecules, the behavior of individuals depends on specific characteristics of the individuals which researchers cannot observe, much less control. One can estimate the average effect on a population in those rare cases where the results of a random sample can be compared to a control group. But the researcher is usually confined to analyzing existing, non-random microdata. For example: Suppose we want to estimate the effect of education on income. The problem is that the available data on wages constitute a selective, rather than a random sample, because individuals who choose to work or engage in education may have other, unobserved characteristics than those who choose to abstain from education or remain outside the labor force. Heckman has developed methods for handling selective samples in a statistically satisfactory way. He has also shown how similar methods can be used to evaluate the effect of public labor market programs and educational programs, and to estimate the effect of length of unemployment on the chance of getting a job. The methods can also be used to study discrimination. Heckman has made significant empirical studies in all these areas. His work has had a decisive influence both on subsequent methodological development and on applied research in microeconometrics.
McFadden has shown how we can handle another fundamental aspect of microdata, namely data on the most important decisions we make in life: the choice of education, occupation, place of residence, marital status, number of children, etc. Such decisions are called discrete choices since they are choices among a few alternatives. Empirical studies of such choices lacked a foundation in economic theory. McFadden’s breakthrough in the mid-1970s was to develop new statistical methods based on a new economic theory of discrete choice. His prize-winning contributions had an enormous impact on modern micreconometrics and fundamentally changed the way researchers viewed empirical studies of individual economic behavior. His methods are very useful. One of many examples is traffic planning, where McFadden’s models are used in order to study what factors determine individuals’ choice of travel mode, for example, whether they choose a car, bus or subway to go to work. The models can be used to predict changes in the percentage of all travelers who choose a given alternative, for instance, if the price of gasoline goes up, if the subway becomes more easily accessible or if longer life span, lower birth rate or a larger influx of people changes the demographic structure of the population. One of McFadden’s original applications concerned the design of the Bay Area Rapid Transport (BART) system in the San Francisco area. He later studied investments in telephone service and senior housing. One of his most recent applications concern the valuation of natural resources, where his model was used to estimate the environmental damage caused by the Alaska oil spill from the tanker Exxon Valdez in 1989.
Professor Heckman, Professor McFadden:
Your pioneering contributions to the analysis of microdata laid the foundation for modern microeconometrics. The methods you developed, together with new data sets and powerful computers, made it possible to study individual economic behavior in a statistically correct way. In your own applied research you demonstrated how solid empirical knowledge can help address important social problems. It is a great honor and privilege for me to convey to you, on behalf of the Royal Swedish Academy of Sciences, our warmest congratulations. I now ask you to receive your Prizes from the hands of His Majesty the King.