The Royal Swedish Bank Prize in Economic Sciences, awarded by the Nobel Committee, was shared in 2021 by three economists – David Card, Joshua Angrist and Guido Imbens, whose research focus was primarily on labor markets. However, rather, the committee appreciated their contribution to solving a more general problem, universal for many social sciences: the question of how to identify cause-and-effect relationships in observed events, if experiments are impossible in the literal sense of the term. In the process of studying the problem, Card, Angrist and Imbens investigated the provincial McDonald’s, the madness of Fidel Castro, the schools of the South of the United States, military dodgers in Vietnam and much more, which allowed only a little progress.
The most general explanation for the topic for which in 2021 David Card of the University of California, Joshua Angrist of MIT, and Guido Imbens of Stanford received the Nobel Prize in Economics, is pointless – the study of causality using observable data (this is literally the headline of the “scientific” press -release (.pdf) of the Nobel Committee), strictly speaking, all economists are engaged, as are all people of academic pursuits. But to say (as is done in the “people’s” press release of the committee) that the Nobelists made an important contribution to the study of labor markets and education is also a simplification: the meaning of Nobel-2021, in general, is not the point.
Perhaps the simplest thing to say is that Card, Angrist, and Imbens, thirty years ago, set a new standard for how economists study data from “natural” experiments in economics — a redesign of such studies is worth a premium.
As in all other areas, the Nobel Committee prefers to reward people by celebrating phenomena in science or society, and in “collective” prizes it seeks to illuminate the phenomenon itself from different angles. In the 2021 Nobel Peace Prize, the committee demonstrated two sides of investigative journalism in the world: “attitude journalism” (Dmitry Muratov) and “factual journalism” (Maria Ressa).
The structure of the Nobel in economics is more complicated. Half of the prize went to Card, who in the early 1990s posed the question of the need for a special apparatus and research design to study causal relationships in empirical microeconomic research.
Card largely continued the ideas of the 1989 Nobeliate in Economics Trygve Haavelmo, who in the 1940s – 1950s actively integrated into economics the ideas of “natural experiment” and the methodology of probability theory, in particular the method of instrumental variables. The problem that Card described looks like this roughly. If in a significant part of sciences a true experiment is possible – with a random sample of objects and a control group, the possibility of limiting the influence of factors other than the investigated ones, then in economics and a number of other sciences, for example, sociology, for many reasons, including ethical ones, this is impossible. Yes, life itself provides a huge number of situations that can be perceived as a “natural experiment”, however, the exact indication of even the direction of the causal relationship in the detected correlation in social interaction is complicated by many other factors, the full accounting of which is difficult, if fundamentally possible.
Among Card’s “experiments”, for example, the study of the consequences of the decision of the Cuban dictator Fidel Castro, who allowed for a month uncontrolled (on inflatable boats) emigration of 125 thousand Cubans to Miami in April 1980: with new approaches, the influence of what was happening on the labor market in the United States was studied. Card’s other famous work examines the impact of higher education on income – often co-written with Alan Kruger, who did not live to see the Nobel for two years (he also co-authored with Angrist).
The second half of the award is divided between Angrist and Imbens, on research on similar topics and a new approach to the analysis of quasi-experimental data, the study of the “local mean effect of exposure” (LATE) and the design of such studies. Most of the “advanced” ways to study causal relationships in econometrics, such as RDD and RKD regressions, differences in differences, and synthetic control method (SMD), are based on theoretical findings from the work of Imbens and Angrist. Card’s work on the reasons for the ambiguity of the increase in the minimum wage in the United States is also widely cited.
Card, Angrist and Imbens have spawned a thirty-year wave of research in econometrics – hundreds of papers a month are written about labor markets, education, and inequality. Note, however, that the hopes (traces of which are in the releases of the Nobel Committee) that the new methods will be widely used in other social sciences are not yet coming true. And it is impossible to argue that the Nobelists pointed out the way to study the problem of cause-and-effect relationships in a series of comparable economic data, which society perceived as the main and generally accepted one – in the disputes of politicians about the minimum wage, references even to Card are not found, and it is unlikely that the Nobel Prize in Economics in 2021 will change a lot.
.