Particle physics experiments

How Card, Angrist, and Imbens Used Natural Experiments to Answer Society’s Most Important Questions

This year’s Nobel Prize in Economics was jointly awarded to academics David Card, Joshua Angrist and Guido Imbens for their pioneering work in using natural experiments as a reliable research method and the revolution in empirical research in economics. .

13 October 2021, 13:20

Last modification: October 13, 2021, 01:25 PM

David Card, Joshua Angrist and Guido Imbens (left to right). Photo: Bloomberg via Nobel Prize Outreach.


David Card, Joshua Angrist and Guido Imbens (left to right). Photo: Bloomberg via Nobel Prize Outreach.

When it comes to proving the reliability of a theory or hypothesis, fields like physics, chemistry or medicine are ahead of the social sciences because of their ability to control different variables through randomized controlled trials (RCTs) and to establish causality.

For example, if researchers wanted to test the effectiveness of a vaccine, they could easily administer the vaccine to one randomly selected group and the placebo to the other to see if there was a significant improvement.

The same cannot be done in the social sciences because of ethical and practical limitations. For example, to study the impact of poverty on the health of citizens, we cannot randomly subject a large group of people to economic hardship to see how their health is changing.

Likewise, other variables could have an impact on both poverty and health that we cannot know or control. So how can sociologists establish causality between real wages and living standards, lockdowns and infection rates, or the impact of different government policies?

The answer is natural experiments – a policy or event that randomly segments individuals into different treatment and control groups that helps researchers establish causation with minimal assumptions.

In fact, this year’s Nobel Prize in Economics (or the Sveriges Riksbank Prize in Economics in Memory of Alfred Nobel 2021) was jointly awarded to academics David Card (UC Berkeley), Joshua Angrist (MIT) and Guido Imbens. (Stanford University) for their pioneering work in using natural experiments as a reliable research method and “revolutionizing empirical research in economics”.

Famous, in the early 1990s, David Card with the late Alan Krueger studied the impact of a higher minimum wage on the employment of the workforce by analyzing a natural experiment (conventionally the assumption was that higher wages created unemployment because they increased the cost of production for companies).

To mimic the terms of an RCT, they collected data regarding employment at fast food restaurants in adjacent areas on either side of the Pennsylvania-New Jersey border, assuming that businesses in both jurisdictions would be quite similar.

Comparing the results of the two regions before and after an 8% increase in the minimum wage was issued in New Jersey, the researchers found that unemployment had not increased as expected despite the introduction of a minimum wage as the theory predicted.

While we now know that the negative impact of a higher minimum wage was quite small as companies simply pass the costs on to customers, the work of Card and Krueger has sparked countless studies and launched the use of natural experiments in economics which were subsequently adopted by other fields of social science.

By studying natural experiences, Card was later able to publish remarkable findings such as the impact of immigration on employment (natives actually improved) and the impact of investing in schools on future student income (yes, better teachers, books, and facilities increased the earning potential of students).

However, it was still difficult to ensure causation in a natural experiment because, unlike an RCT, the researcher cannot control who receives treatment and who does not, making the results difficult to interpret.

To illustrate this problem, let’s explore a scenario. Imagine two similar companies A and B in the same industry where A gave bikes to their employees as bonuses and the other did not. Obviously, this can be a useful natural experiment in finding causality between variables such as time spent cycling and health conditions.

However, the problem arises with the level of participation of the participants themselves. For example, even though they have received a bicycle, some employees at A may never use it to get to work. Plus, there are likely those in A who would have cycled to work, regardless of the corporate giveaway. Finally, there could be employees in B who decide to cycle without being gifted.

These possibilities made it difficult to draw precise conclusions about the causality of the experiments until the work of Joshua Angrist and Guido Imbens in the mid-1990s.

Using probabilities and a limited set of assumptions, they developed the LATE (Average Localized Treatment Effect) framework. This framework could estimate the causal effect among participants who changed their behavior (that is, who decided to ride a bicycle only because it was offered to them) as a result of the natural experience.

This framework was more credible and transparent because it encouraged researchers to state their hypotheses and was widely adopted by other researchers in the field of economics and beyond.

It’s been over 30 years since the findings of the three researchers mentioned above transformed empirical research, provided better policy advice, and helped researchers answer questions we couldn’t study by controlling variables in a lab.

Opportunities to study natural experiences have become relatively abundant in the context of the pandemic and the rise of despotic regimes and political instability.

In this regard, during a call with the scientific director of Nobel Media, Adam Smith (right, I know!), While being congratulated on his Nobel Prize, David Card said: “Crazy political regimes have a lot of disadvantages, but one of their advantages is that they create very good conditions for causal analysis.

Source link


Your email address will not be published.