Opinion | How to identify cause-effect relationships? Award for credibility in Economics

In the last decades economics has become (for the better) much more empirical and data-driven. From an abstract discipline in which dark theories were developed by characters raised by disciples and creators of schools of thought confronting each other, it has passed to another much closer to reality. Now, as in the so-called “hard sciences”, Any statement about the functioning of the economy, to be taken really seriously, must be contrasted with reality.

The problem is that in Economics the interpretation of the available evidence is complex. What we observe after a certain event, caused by an economic or other policy measure, does not have to be the direct cause of that event. Even being aware that “Correlation is not causation”, There was a time when the empirical analysis of economists consisted basically in the interpretation of statistical correlations as causal relationships under the assumption of certain hypotheses (more or less convincing) derived from economic theories or intuitions (more or less realistic). However, most of the time economic theories are insufficient to achieve this identification because almost always There are many reasons for the existence of causal relationships between economic variables and these run in many ways.

And this is what has come to change the call “The credibility revolution” that advocates the use of experiments to obtain convincing empirical evidence, as in the so-called “hard sciences”. The progress and enormous impact of this revolution in empirical economic analysis has already been doubly recognized by the Committee of the Bank of Sweden Prize in Economic Sciences in memory of Alfred Nobel (for all intents and purposes “the Nobel Prize” in economics). It was first in 2019 when Esther Duflo, Abhijit Banerjee and Michael Kremer were awarded for promoting and teaching the usefulness of “field experiments” based on the comparison between two randomly selected groups (RCT, by its acronym in English Randomized Controlled). Trials), one to which a certain measure is applied (intervention) and another that serves as a placebo (control), imitating the way in which the efficacy of a specific treatment is evaluated in Medicine.

Nevertheless, Nor do RCTs solve all problems. In economics, many of the questions of interest cannot be reproduced in the “field” because it is not feasible or ethically justifiable to implement certain treatments. For example, to measure the economic returns to education, you cannot leave half a generation without studying in order to have the necessary control group to compare with those that do. This year’s Nobel Prize winners in economics, David Card, Joshua Angrist and Guido Imbens have covered this flank with the use of natural experiments, that is, episodes in which it is possible to identify control and treatment groups without the need for RCT. In addition to teaching that there are multiple occasions when different population groups are affected in different ways by the event under study, they have provided the statistical material necessary to make these comparisons reliably and rigorously. After these contributions there is no longer any field of economics in which natural experiments are not a primary source of empirical evidence. For example, Labor Economics has been radically transformed by the pioneering works of David Card (many of them done in conjunction with Alan Krueger, sadly deceased in 2019) on minimum wage, immigration and education, among other topics. It is now a field dominated by research based on the analysis of natural experiments.

As always, the award of the Nobel Prize in Economics has caused some controversy. One of Card and Krueger’s papers from the early 1990s found that the rise in the minimum wage in New Jersey did not make the evolution of employment in fast food restaurants different from that observed in Pennsylvania (a neighboring state where wage minimum did not rise). Hence, it is erroneously concluded that raising the minimum wage “does not destroy employment.” And those who have always reviled the Nobel Prize in Economics claiming that it was a “fictitious” prize created only to promote conservative economic interests, now they have appropriated it to validate any increase in the minimum wage at any place and time and whatever the magnitude. of that climb. It would be paradoxical if an award for credibility ends up becoming an alibi to ignore other results obtained by the same methods as Card and Krueger and to create dogmas or fundamental principles that have no place in economics.

Both in Politics and Economics, credibility consists in thinking (well) what is said, saying (timely and conveniently) what is thought and doing (effectively) what is said. These are three principles that, both in Politics and Economics, are not sufficiently followed with the attachment to reality with which this year’s Nobel Prize winners in economics have built their contributions to economics.

We wish to give thanks to the author of this write-up for this incredible content

Opinion | How to identify cause-effect relationships? Award for credibility in Economics

Hank Gilbert