A Top Econ Prize for a Theory That Works

This economist figured out a better way to assign students to public schools.

By Noah Smith
May 15, 2018, 5:00 AM PDT

https://bloomberg.com/opinion/articles/2018-05-15/parag-pathak-won-john-bates-clark-award-for-a-theory-that-works

What do people think economic theorists do? The pundits who regularly criticize the profession, particularly in the pages of British magazines, seem to think that they spend all their time making abstruse, unrealistic theories about how free markets are the best of all possible worlds. And it's true that there are still a few economists out there who are essentially doing that. But a lot of theorists are doing something much more humble and practical work on small-bore theories that can be immediately applied to make the real world a little more efficient.

Parag Pathak is a theorist of this latter type. The Massachusetts Institute of Technology professor last month was awarded the John Bates Clark medal, which is awarded to an economist under age 40 and is considered the second most prestigious award in the field after the Nobel. Although the Clark medal isn’t given for any specific achievement, the prize cited Pathak’s work on improving “the assignment of students to public schools.” And he did it with economic theory.

Pathak is a student of Alvin Roth, an economist who (along with Lloyd Shapley) won the 2012 Nobel for solving a long-standing problem about matching. There are many situations where two groups of people — say doctors and medical schools or kidney donors and recipients — are trying to match up with each other, and all are trying to get the best match they can. The question of how to pair people up so that no one is dissatisfied and wants to switch was solved by Roth and Shapley, who came up with an idea called “deferred acceptance.” Offers to pair up are either rejected or tentatively accepted, and then more rounds of offers are made. Tentative acceptance can be reversed later on, but it gives people an incentive to reveal who their first choice is. This basic procedure has to be modified to fit particular situations, but it can almost always be set up so as to achieve an arrangement that no two people would willingly change.

Roth and Shapley’s insight was useful for much more than winning big gold medals from Sweden. Thanks to the economic theorists, organ transplants and medical school admissions have become much more efficient. The theory’s real-world success is surprising, given that people can easily lie about their preferences, make stupid mistakes or try to game the system in any number of ways.

In 2004, Pathak, still a graduate student, teamed up with his adviser Roth and Atila Abdulkadiroglu to apply the theory to a new problem — the allocation of students to New York City public schools. The city’s existing matching method had turned out to be a disaster, with top students getting their pick of good schools, and disadvantaged students often getting shut out. Pathak and his co-authors proposed to use a variant of the same algorithm that had worked so well in other areas. Each student would apply to her highest-ranked school, and each school would choose which to reject and which to tentatively accept. The rejected students would then apply to their next choices, and the process would repeat until no one wanted to switch.

The idea worked. The number of students who failed to get into a good school fell instantly from about 31,000 to roughly 3,000, and about half of students got their first choice. That doesn’t mean every student in the city gets to go to a top school, but it’s a lot better than what came before.

Pathak didn’t stop there. He helped design a matching system for Boston public schools, and has continued to work with the city to tweak and improve the program. In 2011, he helped create a process called OneApp in New Orleans, which allocates students to both public and charter schools in a special area called the Recovery School District.

Pathak isn’t just a theorist — in keeping with economics’ age of data, he also does a lot of empirical work, much of it focusing on how various admissions setups affect student outcomes. But the success of the systems he created would have been impossible without theory.

And that fact rebuts a lot of the misconceptions about economic theory. Many claim that cold, precise mathematics can never accurately predict human behavior, and that social science should therefore stick to qualitative or literary analysis. Pathak’s research shows that claim to be false — it’s likely that no qualitative analysis could have designed a system as efficient as the ones created by his mathematical derivations.

Pathak also defies the stereotype that economists always sing the praises of markets — his theories involve human strategizing and interaction, but there are no prices, and nothing is bought or sold. Some economists have even referred to theories like Pathak’s as a victory for the supposedly discredited notion of central planning.

In an age when bashing economics is in vogue, the critics should pay attention to researchers like Pathak. Their theories are not as grandiose as the macroeconomic ideas that appear in the press — but they really work, and every day they improve people’s lives.

To contact the author of this story: Noah Smith at nsmith150@bloomberg.net. ©2018 Bloomberg L.P.

Read more at: https://www.bloombergquint.com/view/parag-pathak-won-john-bates-clark-award-for-a-theory-that-works

Copyright © BloombergQuint

Previous
Previous

A New Way to Measure the Impact of Charter Schools

Next
Next

More on That Study of Bayou Charter Schools