Parag Pathak, recently named the nation’s most impressive economist under 40, has focused much of his career on using economic modeling to study and improve school choice.
The Wall Street Journal spoke with Mr. Pathak, a 37-year-old MIT economist who joined the elite group of recipients of the American Economic Association’s John Bates Clark Medal, to learn more about what works, what doesn’t and why some parents struggle.
Here’s the conversation, edited for length and clarity.
Q. Much of your work focuses on school choice. At MIT, you founded the School Effectiveness and Inequality Initiative. What prompted you to want to study school choice?
A: I kind of fell into the topic. When I was in graduate school, I was studying market design, which is this branch of economics that’s trying to come up with better ways of allocating resources. There was a confluence of things happening around that time involving public schools and school choice.
The first was, in 2003, New York City’s public school assignment system – it kind of fell apart. A team of economists were able to go in and say, Why don’t we treat this issue of school matching as an economics problem? And so we designed a system that works a little like the system used by medical residents in the U.S. It’s both more efficient because it’s letting the market sort students more quickly and it also respects their choices more.
That showed me that economics and market design could be useful in solving real, practical problems.
Q: Your work touches on “school choice” both in public and private schools. What kind of school choice do you find is most effective at improving students’ performance?
A: There is a particular type of charter school that seems to be doing spectacular things for students, they are sometimes called “high expectations” or “no excuses” charter schools. Maybe the best way to describe the schools I am talking about is like the KIP Network – the “Knowledge is Power” program.
They are the largest charter manager across the country. They have a culture of high expectations. Their school day is longer, their school week is longer—they sometimes meet on Saturdays—and their school year is longer.
When there are more applicants to these schools, they run a lottery. That lets you construct an apples-to-apples comparison: you compare kids who won a seat at one of these schools to ones who did not win a seat.
And what we see is the students who won the lottery were performing better at school and going to higher quality colleges.
There’s a question of, Can a school by itself overcome the disadvantages that families face, especially in big cities? These high-performing charter schools are very encouraging. But the important thing to emphasize is that not all charter schools are in this class.
Q: Based on the work you’ve done on vouchers, what does the evidence say about their effectiveness?
A: One thing that got a fair amount of attention was our study of private schools in Louisiana. Louisiana has a voucher program and sends low-income kids to private schools. The state will pay for you to attend a private school. They use a lottery to place students, so we were able to study that lottery system.
Children who won the lottery actually experienced a substantial reduction in achievement. It was one of the most negative effects we’ve seen.
It was a system that involved voluntary participation from private schools, so it’s my suspicion that the schools that participated in the program were not the strongest private schools in the state. The schools that are most likely to sign up to receive voucher students are the ones who needed the money.
But I’ve only been involved in one study on this topic. There’s other work out there that I would say suggests the evidence is fairly mixed.
That should be seen in contrast to the literature on charter schools. Charter schools are a case of much more managed competition. They are much more subject to regulation. And their effects are much more positive.
Q: Several of your studies get at this core question: How much does going to an elite school influence someone’s education?
A: We have a couple of studies that show pretty convincingly that the role of peers at an elite school is kind of overstated in predicting educational outcomes.
We did one study of elite public schools, like Stuyvesant in New York. The question we were looking to answer is: Is a kid’s success a consequence of Stuyvesant or is it a consequence of getting into Stuyvesant?
The way we approached this question is by looking at the cutoff. If you compare kids who just cleared the cutoff to the ones who made it just below the cutoff, you’d expect that if peers are really important for achievement, the kid who got to go to an elite school like Stuyvesant would do better. They don’t.
The title of our paper—I’m very proud of this—is “The Elite Illusion.”
Q: So did you waste your time going to Harvard for college?
A: I don’t think it was a waste. There are other things I got that can’t be measured on a standardized test. But it did make me realize I shouldn’t have stressed so much about it. Maybe it doesn’t matter as much as we thought.
Q: What are the most important ingredients for student achievement?
A: That’s a very hard question. What I sometimes find frustrating in conversations about student achievement is they often get sidetracked from the issue of school quality.
At least for disadvantaged children, we see that what helps is what you might expect: smaller class sizes, longer school days, an environment that emphasizes mathematics and reading. They are schools that are attuned to data—they use assessments to see where they’re at.
For very high-achieving kids, what kinds of strategies and school models are best? There we don’t really know what works best.
Q: How have you found that your work intersects with national politics and the school-choice movement?
A: A lot of our findings are very much in the news these days with the current administration’s push to expand school choice. I don’t participate in political discussions.
On charter schools for example, our work says the results are mixed. We’ve seen that not all charters are doing wonderful things. But advocates for charters often point to our research.
Our job as researchers is exploring the nuances and subtleties, and unfortunately that comes in direct conflict with political discussions.
Q: Do you think giving people more choices in their education or their children’s education is always a good thing?
A: One of the narratives of school-choice reform going all the way back to Milton Friedman is: Giving people choice is inherently good. Now we are starting to see a couple of pretty high-profile scenarios, like the Louisiana voucher system, where giving people choice actually results in people learning less.
Something we tend to see over and over is when we try to make decisions about education as consumers, we don’t have that much information. People struggle to measure value added in their decision-making.
Handicapping the John Bates Clark Medal (April 16, 2014)