Famed British economist John Maynard Keynes would have found this year’s Nobel economics laureates “splendid.” David Card, Joshua Angrist and Guido Imbens all are microeconomists working on questions of direct importance to businesses and households by measuring, analyzing and evaluating what occurs in the real world.
All three got the prize for using “natural experiments” to answer pressing applied questions: Do increases in minimum wages decrease the number of jobs or not? If yes, by how much? What does the arrival of many immigrants in a specific locality do to wage rates? Does military service really change life outcomes in terms of income, education or health? Why or why not?
Note the prize was not awarded to each individually for work on a general technique — as were the 1994 prizes for game theory were given to John Harsanyi, John F. Nash (the “beautiful mind” of movie fame) and Reinhard Selten for individual, but complementary work. Nor was it awarded to the three as a team, as is common in pure sciences.
Instead, Card got half of the prize himself and Angrist and Imbens split the other half for work often done together. There also was this tragic irony: Card was not a loner; he did much of this with Alan Kreuger, another brilliant scholar who died by suicide in 2019. Nobels are never given posthumously. And it is a statute of the Nobel Committee that no prize can be given to more than three people. So if Krueger still were alive, it is likely that either one duo or the other might have gotten the award, but all four could not.
But what about the work?
Let’s start by understanding differences between theoretical versus applied research, and between natural versus designed-and-controlled experiments.
Consider physics: Albert Einstein and Nils Bohr were theoreticians who worked in their own brilliant minds through observation and calculation. They did not have laboratories where they connected wires to sensors or invented detectors of various atomic particles. On the other hand, Berkeley’s Ernest O. Lawrence’s cyclotron atom smashers, Lise Meitner with her makeshift tabletop apparatus first detecting nuclear fission or J.J. Thompson, who discovered the electron, were all experimentalists spending long hours in laboratories.
In economics, Friedrich Hayek, Gerard Debreu or Minnesota’s own Leonid Hurwicz all principally focused on theory with little work on applied problems; whereas another series of economic laureates in 1970s-1980s sifted reams of data to develop computer mathematical models for forecasting what the economy would do. Their data all came from existing measurements of national economies.
Recently however, the Nobel committee has turned to economists who apply experimental data to specific real-world issues. The 2019 award to Abhijit Banerjee, Esther Duflos and Michael Kremer was for experiments testing different approaches to relieving poverty.
Now we have Card, Angrist and Imbens.
The difference between the two years is natural and controlled experimentation.
In 1916, Einstein published his full General Theory that upset accepted ideas on gravitation. It predicted that gravity could bend light much more than predicted by Isaac Newton’s work. This caused great controversy that could not be resolved in a laboratory.
However, a British astronomer, Arthur Eddington, recognized that observations of a major solar eclipse due in 1919 would produce data proving or disproving Einstein’s new theory. At great expense, expeditions of scientists went to northeastern Brazil and the tiny African island of Principe across the South Atlantic, to take measurements. These proved Einstein correct. Measuring data from this already-occurring event was a natural experiment.
In 1938, two German chemists bombarded uranium with neutrons and discovered that barium appeared. Physicist Meitner and her nephew Otto Frisch immediately showed this resulted from uranium atoms being split. All of this was a designed, controlled experiment, not a natural one.
Duflos and her 2019 co-laureates designed controlled experiments of programs to help poor people: Randomly sort poor children or families into two groups, give one group a basic monthly payment or school meals or some health measure, don’t give it to the other group. Then, after some interval, measure how things turned out for each group, as a pathologist might with test plots of two wheat varieties.
In contrast, his year’s laureates’ natural experiments measured data from things occurring in the real world, similar to the eclipse measurements.
Here’s what they found:
New Jersey raised its minimum wage. Adjoining Pennsylvania did not. There were large single urban areas divided between the two states by a river. What happened to the number of jobs and to prevailing wages on either side of the river? Card and his then-colleague Krueger showed the higher minimum wage on the New Jersey side did not cut jobs relative to the Pennsylvania side, the way simple economic theory predicted.
Some students get admitted to elite private high schools, others do not. Some graduates of these elite schools get into top universities and others do not. Ditto for ordinary high schools. Are the higher top-university admission rates of elite grads due to superior teaching? Or just to the way such schools screen who they admit? Angrist and Imbens demonstrated the latter factor is more important.
Historically, military service was connected to better life outcomes in some terms of education and earnings. Was it the cause? Some people enlisted in the military, some were drafted, some never served at all. Comparing the outcomes for these distinct, already-determined groups can tell us something.
Thus, the 2021 awards for analysis of natural experiments complement the 2019 award for controlled experiments.
There also is a salient point in the award announcements that many miss.
Card primarily got his award for the policy implications of his outcomes: Do minimum wage increases kill jobs? Card showed not necessarily. Does a sudden influx of immigrants lower wages? Card showed it did not in at least one localized case. Angrist and Imbens were honored for the innovative statistical techniques used to analyze data, but not as much for the results obtained.
Now for Keynes, who once said, “If economists could manage to get themselves thought of as humble, competent people on a level with dentists, that would be splendid.”
The work of some grand theoreticians and forecasters who got economics Nobels in the first 40 years of the award has proven pretty useless. These recent winners, the more “humble, competent” scholars, have provided us with work that is at least as useful to humanity as dentists.
St. Paul economist and writer Edward Lotterman can be reached at stpaul@edlotterman.com.
Real World Economics: Nobelists honored for work and methods - TwinCities.com-Pioneer Press
Read More
No comments:
Post a Comment