Nobel laureate in economics turns statistics into insight


Correlation is not causation. There is an important fact behind the clichés. For example, in January this year, the United Kingdom implemented one of the strictest lockdown measures and is also one of the countries with the highest Covid death rate. There are no deaths in New Zealand and almost no restrictions. However, no matter what your favorite YouTube conspiracy theorists say, the blockade will not cause a wave of Covid. The wave of Covid led to the blockade.

But while “correlation is not causation” is an important warning, it is not a response when policymakers ask questions.

For example: Why do people who are educated tend to have higher incomes? Is it because education brings higher income, or is it because smart, energetic people can thrive in schools and workplaces?

Why are there so many foreign-born workers in affluent places? Is it because immigration has increased income, or is it because people go to places with money?

Where there are many storks, there are also many babies. Is that because the stork gave birth to a child, or is it because there is room for two children in the big country?

The example of stork and baby is a warning story, as I explained in my book, How to make the world add upIn 1965, the famous statistical communicator Darrell Huff stated in a US Senate hearing that the correlation between smoking and cancer is as false as the correlation between stork and baby.This is a grim example that illustrates Healthy doubts It can be condensed into cynicism.

All of this explains why I announced on Monday Nobel Prize in Economics. Winners David Card, Joshua Angrist and Guido Imbens led the “credibility revolution” in the field of economics.

Facing chaotic real-world data, economists can easily shrug their shoulders and avoid key questions such as “Can education increase income?” And “Can immigration increase productivity?” Card, Angrist, and Imbens showed careers that we can be more ambitious.

In 1992, New Jersey raised the minimum wage from $4.25 per hour to $5.05 per hour. Will this make some fast food restaurant employees too expensive to hire? Card and Alan Krueger discovered a natural experiment: Pennsylvania is adjacent to New Jersey in the east, with similar economies, but Pennsylvania has not changed its minimum wage. Kruger Comparing the employment situation in New Jersey and Eastern Pennsylvania, we found no signs of job losses in the fast-food industry when the minimum wage in New Jersey increased.

This is an extremely influential discovery, but perhaps the most important part of it is not the result, but the proof that economists can find data to answer serious policy questions.

Angrist and Krueger solved the education income problem by observing a quirk in the American education system. Consider two children, one born in late December and the other born a few weeks later in early January. Children in December start school a full year in advance. However, both children can legally leave school on their 16th birthday, a few weeks apart.The difference seems trivial, but in 1991 Angrist and Kruger It shows that the baby’s time in school in January has decreased significantly, and the income has also decreased.

Of course, only some children drop out of school at the age of 16; most do not. This is a typical natural experiment: instead of randomly assigning drugs and placebos, natural experiments randomly assign something more vague, such as the opportunity to drop out of school early.

This is a headache for statistics, but Imbens and Angrist have developed a toolkit to help researchers discern clear causality from vague natural experiments. Economics has become a field full of clever experiences and discoveries, most of which are based on Angrist-Imbens.

This year’s Nobel Memorial Prize is bittersweet.This is a reminder of suicide Alan Kruger in 2019Kruger co-authored several papers cited by the Nobel Committee.

This is also a stark example of the gap between political speech and the best data detective work. For example, one of Card’s most influential essays touched on the hottest topic in British politics today: Can you increase wages by restricting immigration? British Prime Minister Boris Johnson said he can and he will.

The data shows a different story.Card studied Mariel ship liftIn 1980, 125,000 people moved from Cuba to the United States. Most of these people arrive and stay in Miami, and most of them are relatively unskilled.

Although Miami’s unskilled workforce has increased by nearly 20% in a few months, Card found no signs of depressed unskilled wages in Miami. Instead of using the influx of workers to drive down wages, Miami’s companies have found a way to hire these new workers.

This is just a study, but Card’s work prompted economists to rethink the simple model of immigration. The current balance of evidence shows that migrant It is more likely to increase productivity rather than inhibit it.

The world is full of interesting data, but not full of strictly controlled experiments. It is too easy to use treacherous statistics to argue that the blockade caused the new crown virus. But denying the evidence altogether is not much better, reassurance that cigarettes may be safe, because correlation is not causation.

We can do better. As Kruger once said: “The idea of ​​turning economics into a real empirical science and that the core theory can be rejected is a great revolutionary idea.”

that’s all. It is indeed possible to turn statistics into insights. We must try.

Tim Harford’sThe next 50 things that will make a modern economy“It’s a paperback now

follow @FTMag Get to know our latest story first on Twitter


Source link

Leave a Reply

Your email address will not be published. Required fields are marked *

Powered by BeaconSites