Raj Chetty's Opportunity Insights wants to fight recessions with new tools, data : Planet Money A research group at Harvard came up with a faster way to check the economy's pulse. It may change how we fight recessions. | Subscribe to our weekly newsletter here.
NPR logo

What's Next for the Economy?

  • Download
  • <iframe src="https://www.npr.org/player/embed/931496888/931511082" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
What's Next for the Economy?

What's Next for the Economy?

  • Download
  • <iframe src="https://www.npr.org/player/embed/931496888/931511082" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


AMANDA ARONCZYK (HOST): Well, that happened.

ROBERT SMITH (HOST): Happened? Amanda, it is still happening.


SMITH: As of midday Wednesday when we're recording this, we still don't know for sure who will be sworn in as president on January 20.

ARONCYZK: So instead of fixating on, say, vote counting in Lackawanna County in Pennsylvania, we want to spend today's show on what we do know about the economy.

SMITH: With all that has been happening - and it has been a lot - we may have forgotten that we are still in the midst of an economic catastrophe, and the next Congress and president have to do something about it.

ARONCYZK: But figuring out exactly where we are in the economy is harder than it may seem. The drop in March and April was so quick, and then the partial recovery was so fast. The numbers that we do have seem to go out of date really quickly.

SMITH: Once every three months, we get these advanced numbers about the GDP, the gross domestic product. That's the sum total of everything we buy and sell. And that number comes out, and when it comes out, it's just an advance number. And then it gets updated a month later.

ARONCYZK: And then there is all this data we only get a glimpse of, like, once every blue moon - consumer spending, housing. So how does the next Congress, the next president, design a plan to help the economy figure out who to send money to and exactly how much when, honestly, we don't even know what's going on right now?

SMITH: Hello, and welcome to PLANET MONEY. I'm Robert Smith.

ARONCYZK: And I'm Amanda Aronczyk. Today on the show, a ragtag team of brilliant economists have been working on a way to better understand exactly where we are this instant using data that we have never seen before.

SMITH: We'll give you a quick tour of the state of the economy, provide the warm comfort of some cold, hard data and ask, what do we do now?


ARONCYZK: So to get a better understanding of where we are right now with the economy, we're going to look at this one group of economists who seem to have some answers. So, Greg Rosalsky, you're the guy who writes our newsletter, and you are always nerding (ph) out on the latest research. Hello there, Greg.

GREG ROSALSKY (BYLINE): Hey, Amanda and Robert.

ARONCYZK: So this group is called Opportunity Insights. Can you tell us a bit more?

ROSALSKY: Sure. So Opportunity Insights is this relatively new research and policy institute based at Harvard University. It's led by Raj Chetty and John Friedman and Nathaniel Hendren, and they've been doing some really amazing work. So kind of the original idea of the group is to sort of use, like, cutting-edge economic research to figure out ways to fix poverty in America. So then COVID hits, and I spoke with John Friedman about this.

JOHN FRIEDMAN (CO-DIRECTOR, OPPORTUNITY INSIGHTS): Some of our projects had to be halted due to COVID. And it was clear this was going to be not just a public health event but a major economic event that held the possibility of driving many people into poverty. So we put together a call to try to brainstorm around this and how we could contribute.

ROSALSKY: So they're on this call, and they start talking about one of the big problems with responding to this economic crisis. The numbers we all rely upon - the official economic statistics like GDP and the unemployment rate - they're done using surveys, calling people up on the telephone or sending them forms. I mean, now, first of all, this data isn't very detailed. And even more pressing, it takes a long time to get this data. And the virus is moving really fast.

FRIEDMAN: You know, imagine driving a car except you're blindfolded and you don't get any information about whether you're doing it right or not until a minute has passed, right? It's going to be really difficult to get where you want to go. And I think it's very much the same situation with the economy. We didn't get the first estimate of what GDP looked like in April until the end of July.

ROSALSKY: That's four months of waiting. And think about what's happening in the country at that time. I mean, the economy is cratering, so they come up with a solution. Let's kind of turn to private companies. Like, private companies have all this valuable data.

SMITH: Because if you think about it, a company like Starbucks - like, they know if they're going to run out of coffee at 1 o'clock in the afternoon at a particular store. Or Amazon or credit card companies can tell you exactly, like, how good was the morning in the U.S. economy.

ROSALSKY: Right, Robert. I mean, it's kind of known in academia as the big data revolution. And it's had this huge impact on the economy. And, I mean, like, the whole business model of Google and Facebook is just to harvest our data. And Friedman says we're basically, like, just trying to do this for policymaking in this country.

FRIEDMAN: So what we're trying to do here is take all that data, except instead of trying to figure out how to, you know, sell you on a particular brand of coffee, we're trying to help everyone understand what's going on in the economy.

SMITH: So I get that it's this dream of economists to be able to access this beautiful private dataset, but why would companies give it to economists? I mean, this is the most valuable thing they own - their secret proprietary data. Why would they want to share it with the world?

ROSALSKY: Well, first of all, I mean, there's a national crisis going on. And they're making this pitch that, you know, we need this data to help the country. But beyond that, I mean, Opportunity Insights is a pretty prestigious organization. They're funded by the Bill and Melinda Gates Foundation and the Chan Zuckerberg Initiative. These are all fancy-shmancy (ph) economists at Harvard and Brown University. And they're able to, you know, start making some calls and getting their hands on these datasets.

ARONCYZK: So what are some examples of the kinds of data they were able to get?

ROSALSKY: We're talking credit card spending data from a company called Affinity Solutions, small business revenue data from a company called Womply. They got employment data from, like, Earnin, Paychex and Intuit. You may have heard of that last one. I don't know. Do you guys use TurboTax?

SMITH: I do.

ARONCYZK: I do not.

ROSALSKY: Intuit owns TurboTax, so they might - Robert, they might have your data (laughter). But...

ARONCYZK: They won't have mine.

SMITH: That's OK.

ROSALSKY: If they use it for good.

ARONCYZK: OK. So, Greg, let's - what we're going to do now is we want to go through all the different areas of the economy and find out what the researchers are finding in the new data that they've got. So let's start with something like unemployment. We know there are about 12 million people who are unemployed right now. What was this group able to tell you about the details in those numbers?

ROSALSKY: They're able to sort of break down what unemployment looks like for different groups and different ZIP codes. So let's start with, like, breaking down groups by income. And actually, they're able to tell us something very specific about that. So for high-income people, those making more than $60,000 a year, Friedman says they're actually doing really well right now. Their job market has completely recovered.

FRIEDMAN: So for high-income workers, the recession is really over. It was over by June. But for low-income workers, employment really only recovered about halfway, and then starting in July, it started to flatline. And in the last month or so, it's even started to get worse.

ROSALSKY: Low-income workers, the poorest workers in our society - they make up the vast majority of the over 12 million Americans who remain unemployed. And according to OI's data, their employment rate is at Depression levels. We're talking over 20%. It's down over 20% since the pandemic began.

FRIEDMAN: There's no question in our data that this recession is widening the inequalities that we have in this society.

SMITH: And this is also not equal across the United States, obviously, right?

ROSALSKY: It's not. It's not. And that's one of the great things about the Opportunity Insights data - is not only can they break it down by different groups. They can actually break down these numbers by different ZIP code, even different neighborhoods. So take, for example, Midtown Manhattan.

FRIEDMAN: Employment rates in Midtown Manhattan have absolutely cratered. Employment there is down 70% from what it was before the pandemic.

ROSALSKY: But Midtown Manhattan is really kind of like two places when it comes to employment. I mean, there are high-income people who work in offices, and there are lower-income people who work providing face-to-face services like, you know, making people sandwiches during lunch or mixing martinis for, you know, people after they get off work.

And what the Opportunity Insights data shows is that after the pandemic hit, I mean, rich people stopped spending. I mean, most of them, you know, could do their job remotely. Some of them even left New York City. Meanwhile, lower-income workers whose jobs depended on being face-to-face with people and, you know, richer people coming in and buying sandwiches and martinis - those lower-income workers' jobs just disappeared.

SMITH: I'm glad you mentioned spending, Greg, because I feel like we're getting these weird mixed signals about how people are spending in the economy right now. On one hand, we're seeing retail sales have almost returned to normal. We're seeing a lot of sales of real estate. Durable goods - so things like refrigerators and stoves - that's come back to normal. It seems like there's a lot of buying out there, but if you look closer with this Opportunity Insights data, what do you actually see in spending?

ROSALSKY: If you go back to, like, the beginning of the recession - like, back in late March, early April - consumer spending just really fell off a cliff. It was down 30% from what it was before the pandemic. And there's actually some good news here, as you were just talking about, with retail numbers and with houses and, you know, refrigerators and that sort of stuff. You know, with government assistance in the CARES Act and sort of just this broad societal adjustment of living in the pandemic, consumer spending has really come roaring back. It's now down about 5% from what it was before COVID hit.

SMITH: Which we should say, like, sounds small. But that also means that we're behind two or three years.

ROSALSKY: Right. I mean, like, 5% drop in spending is still catastrophically bad. That's, like, a really bad recession. But what the 5% drop is missing is there's also been this huge shift in spending, and, you know, we're just spending differently than we did before. All those high-income people who were once upon a time going out to bars and restaurants and yoga studios - they now have all this extra cash to burn. And how they're spending it is much differently - so more online delivery, buying gadgets and gear. And Opportunity Insights can see this in the spending data, and it's having this really lopsided effect on the economy.

ARONCYZK: So what I imagine you're seeing is that, like, the Walmarts and the Amazons and all of the online delivery places are doing just fine. They probably are doing even better. And then so who is doing poorly, then?

ROSALSKY: Well, I talked to Friedman about this.

FRIEDMAN: At its root, people still aren't willing to go back and consume services that require being in close proximity to other people because of the pandemic. People still aren't willing to really go back and eat in restaurants. They're not shopping in stores with the same frequency that they used to. All of those shifts tend to move people away from small business.

ROSALSKY: Their data shows that small business revenue has just been completely eviscerated during the pandemic. In late March and early April, it fell, like, 50%. Small business revenue has rebounded a bit since then, but it's still down over 20%. And this finding has really helped solve a puzzle that has been kind of bothering me. I mean, we have this really high unemployment rate. We have this decline in spending. We have all these failing businesses. And yet the stock market is kind of shrugging it all off.

SMITH: Yeah.

ROSALSKY: But this weird disparity kind of makes sense when you think about it because all of these in-person services that have taken a huge hit, like going out to a bar or getting your hair cut - it's generally a small business that is providing you with that service.

SMITH: Who is not listed on the stock market.

ROSALSKY: Who is not listed on the stock market. But this is still a huge problem for the economy. I mean, half of all private sector employment in this country is provided by small businesses. There are literally millions of jobs on the line.

ARONCYZK: Now, of course, the government did do something to try to fix these problems. There was the stimulus package that Congress passed back in March. Coming up after the break - did it work? And what should Congress and president whoever do next?


ARONCYZK: One fascinating thing about the economic situation we are in now is that it's entirely within the power of Congress and the president to make it better. Whatever your politics, trillions of dollars can do wonders to perk up a flagging economy.

SMITH: True.

ARONCYZK: Congress did it back in March. It can do it again.

SMITH: But before we do it, we should ask, did the original stimulus work? Let's recap what Congress did. It was called the CARES Act. It passed in March - $2 trillion for all sorts of things. There were the $1,200 checks for a lot of Americans, enhanced unemployment payments and the PPP, the Paycheck Protection Program, which was supposed to support small businesses. So we're going to bring Greg back in. Greg, what do we know about what worked and what didn't?

ROSALSKY: So let's start with the checks. So Friedman says their data suggests that the checks were not exactly well-targeted. So high-income people - they were able to look in their data and see that high-income people generally did not spend the checks. They tended to save the checks. Low-income people did spend the checks. So the CARES Act, Friedman says, is - these stimulus payments, also the massive boost to unemployment insurance - this really helped low-income people. In fact, according to their data, their income was, like, better than ever.

FRIEDMAN: It really provided a buffer where low-income households, who were experiencing historically high unemployment rates, actually had more income and more spending, in some cases, than ever before. April was an amazing month in terms of income because - in part from all these stimulus payments.

ROSALSKY: There's actually another study by a group of researchers that found that the poverty rate declined in April because of these payments. But by the summer, low-income folks had mostly spent their checks, and because of how the CARES Act was structured, the federal boost to unemployment insurance ended.

ARONCYZK: OK, so that is the consumer spending piece of it. One of the things that we did an episode about and I've been wondering about ever since is, like, what happened to PPP? Was that helpful for small businesses?

ROSALSKY: So Opportunity Insights, they're pretty negative on the program. They've been crunching the numbers, and their most recent estimate is that PPP boosted employment by only 2.4% or about a million and a half jobs.

SMITH: Well, that's not nothing.

ROSALSKY: Right. But Friedman says you have to keep in mind how much all that cost.

FRIEDMAN: What we got from spending half a trillion dollars was just maybe a million and a half or two million jobs, and so that works out to about $300,000 per job.

ROSALSKY: Over $300,000 per job - and these jobs are only guaranteed for a few months.

ARONCYZK: Do they think that PPP actually kept some businesses from shutting, though?

ROSALSKY: So there has been some other research on this, and it does suggest that it did help small businesses survive the pandemic. But we don't really know what the long-term effects are because, you know, it hasn't been long-term yet. What we do know is that the program was really expensive because it was not well-targeted. A lot of the PPP money went to businesses that might not have actually needed it.

FRIEDMAN: Many businesses that took advantage of this aid really were not suffering as a result of the pandemic to begin with. Many of these were in parts of the country where the economy was less affected. These were white-collar firms where they were more or less able to shift work to working remotely without much disruption.

ROSALSKY: What economists agree on is that it was extremely expensive for what it did.

SMITH: So regardless of how the election sorts itself out, the president and Congress have to figure out one big thing, which is, should there be another stimulus package? And what exactly should be in it? Like, what is the best bang for the buck?

ROSALSKY: Right. And I should say here the Opportunity Insights - the team - this isn't, like, just, like, a bunch of academics in the ivory tower. They have actually been talking to policymakers in Washington. They've talked to the Senate Finance Committee, people in both parties in the Senate Finance Committee. And John's colleague Raj Chetty - I love you, Raj; please call me - briefed Joe Biden and Kamala Harris about their data and findings back in August.

SMITH: So what did he tell them to do?

ROSALSKY: Friedman says their message to politicians is pretty simple, and it could be boiled down to this. The old weapons against recessions, like widespread tax cuts or stimulus checks - they won't work well in the coronavirus economy. The reason is because, like, people just won't go out and spend it for fear of the virus. I mean, like, there will be some spending, but it's not going to flow to the parts of the economy that are the hardest-hit, which are these small businesses and low-income workers in these particular ZIP codes.

SMITH: So he's saying don't send checks to everybody.


SMITH: Don't have a broad-based PPP every-business-can-apply-for-this loan.

ROSALSKY: Exactly. They're saying, let's target this aid. Instead of, like, a buffet where everybody can just go out to eat, let's do, like, you know, a Seamless delivery to workers and businesses that are actually struggling. And what's so cool about their data is they can actually tell us who those people are.

SMITH: You know, Greg, I'm struck by how, beyond the current COVID crisis, that they're coming up with these new tools that government can use. A lot of the tools we use now - you know, unemployment, GDP - that came out of the Great Depression, when, you know, government panicked and said, we don't know what's going on. We need to invent these things. Is there a sense that, like, these are the tools that we're going to be seeing from the federal government in the future? They're going to be publishing unemployment rates every week, every day?

ROSALSKY: Robert, that's actually, like, why I've been nerding (ph) out so much about this. It's not just that they're helping us understand the recession, like, in real time and have this high-quality data that can tell us specific things about it. It's like they're revolutionizing the way that we all look at recessions. And you know, like, you mentioned the Great Depression. There's a great parallel there. I mean, Simon Kuznets - he was this professor at Harvard, and he invented these survey methods to measure GDP and employment. And at the time, it was this revolutionary thing for the economy and helped people in Washington and everybody, you know, make better decisions about how to manage the economy. It's almost like these researchers are bringing all of that into the computer age. It's like, it's time to bring big data to government.

SMITH: Greg, thank you so much.

ARONCYZK: Thank you so much, Greg. This was super-interesting.

ROSALSKY: Thank you, guys. It's been a pleasure.


SMITH: If you would like to hear more of Greg's insights, he writes the PLANET MONEY newsletter. It's released every Tuesday. I highly recommend it because you can learn about fun things like how gyms are responding during this economic downturn and the recession in office furniture. That's the PLANET MONEY newsletter. Also, we love to hear from you. We are planetmoney@npr.org.

ARONCYZK: You can find us on Instagram, Facebook, Twitter and TikTok. We are @PlanetMoney. We should also mention the Bill and Melinda Gates Foundation, which supports Opportunity Insights, is also a supporter of NPR.

SMITH: Today's show is produced by Alexi Horowitz-Ghazi with help from James Sneed and Gilly Moon. Our editor is Bryant Urstadt, and our senior producer is Alex Goldmark. I'm Robert Smith.

ARONCYZK: And I'm Amanda Aronczyk. This is NPR. Thanks for listening.

Copyright © 2020 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.