A physicist, a chemist and an economist walk into a bar where they meet a psychologist. He asks them, “If you were on a desert island with no escape and only one can of beans but no tools, what would you do?” “I’d look for a rock to pound the can open,” says the physicist. “I’d build a fire and heat it until the gases and liquids inside forced it open,” says the chemist. “That’s easy,” says the economist. “I’d assume a can opener.”
This hoary economics joke is based on a real problem. Economics, as a science, has until recently not had a reliable way to use real-world evidence to test its theories; it has relied on lots of assumptions. The Nobel Memorial Prize in Economic Sciences this year rightly went to three men who opened a new chapter for testing theories based on “natural experiments.”
For example, economists assume that the more costly something is, the less demand there will be for it. This theory was used for endless studies saying that — assuming no other factors are at work — raising the minimum wage reduces the number of jobs available. The trouble is, there are always other factors involved, and these depend on very particular situations. It is not always the case that raising the minimum wage reduces employment opportunities.
Or consider the impact of immigration on wages. A standard economic assumption is that the greater supply of labor created by immigration depresses wages among the rest of the workforce. But what if, because of language barriers, the influx only reduces the wages of previous immigrants who might not be any more proficient in English than the newcomers?
Half of the economics prize went to David Card of the University of California, Berkeley, for work he did with Alan Krueger of Princeton University in the early 1990s on the minimum wage question. (Mr. Krueger would have been eligible to share Mr. Card’s award had he not died at the age of 59 in 2019.) They realized a good “natural experiment” was going on when New Jersey raised its minimum wage and Pennsylvania did not, so they compared employment at fast food restaurants near the states’ common border and found, contrary to the normal assumption, that employment in New Jersey restaurants remained as steady as in Pennsylvania.
A similar “natural experiment” carried out by Mr. Card showed that immigration did not reduce wages.
But the lesson was not that raising the minimum wage or allowing more inflation would never have a negative economic impact. The lesson was that it depends on other circumstances.
In the case of the fast food restaurants, employers controlled so many other variables, including pricing, supply costs and ways to make their restaurants more efficient, that the New Jersey restaurant owners were able to absorb the higher labor costs without reducing staffing. That might not always be the case.
In the case of immigration, further work showed that wages did fall, but only among previous immigrants whose command of English was not much better than that of the newcomers.
The second half of the Nobel Prize for Economics went to two men who developed a disciplined, reproducible way to examine the effects of these special circumstances by varying them one at a time to identify the critical ones.
That award was shared by Joshua Angrist of the Massachusetts Institute of Technology and Guido Imbens of Stanford University.
The head of the Economics Prize committee, Peter Fredriksson, summed up the winners’ contributions as follows: “(The three) have shown that natural experiments are a rich source of knowledge. Their research has substantially improved our ability to answer key causal questions, which has been of great benefit to society,”
The valuable techniques they developed, noted the award, have spread to other fields and revolutionized empirical research. The prize was refreshing recognition for groundbreaking work in economics that’s rooted more in the real world and less in a theoretical one.