Podcast

Related Links

• Opportunity Insights
• Investing in Rural America

Transcript

Tim Sablik: For the seventh annual Investing in Rural America conference, the Richmond Fed worked with the Federal Reserve System, Rural LISC, and other community partners to bring together community development practitioners, financial institutions, policymakers, researchers, and rural leaders from across the country. The theme of the three-day event in Asheville, N.C., was "The Strength of Rural Places," highlighting the people, institutions, and strategies that power rural economies.

In this episode, I'm speaking with Raj Chetty, the William A. Ackman Professor of Economics at Harvard University and the director of Opportunity Insights. I caught up with Chetty at the conference just before his keynote presentation on "Economic Mobility Across Rural America."

I'm glad that you could take some time to join us here in the on-the-road podcast studio to talk about your research.

So, your research explores the factors that influence economic mobility in the United States. How do you define economic mobility and what drew you to that research topic?

Raj Chetty: Economic mobility sounds like a bit of a technical term. But it's at the heart of what America is about, going back to the founding principles of this country — the idea that we aspire to be a land of opportunity where anyone has the chance to achieve the American dream of moving up, no matter what their background. That's what drew my own parents to come to this country. That's what has drawn countless other immigrants to come to the United States.

So, the way I think about economic mobility is if you grow up in a family with a lower level of income without all the resources, what are your chances of making it in life and succeeding? One simple way to measure that is if you're growing up in a low-income family, what are your odds of reaching the middle class or beyond? What is your average income in adulthood?

Sablik: What are some of the characteristics that you found that really matter for economic mobility?

Chetty: The way we study this question in our research group, Opportunity Insights, is to use the tools of big data — for example, information from anonymized tax returns and Census data covering the entire U.S. population. That gives you a lot of information on who is able to rise up. We're able to link kids to their parents and ask, "How are they doing relative to how their parents were doing in the previous generation?"

Importantly, we're able to look at that data geographically. If you ask what factors matter for economic mobility and is there a lot of economic mobility in America, is the American dream well and alive, the answer varies greatly depending upon which part of America you're talking about. There are some parts of the country, like the rural Midwest for example, where kids growing up in low-income families have great chances of rising up. There are other parts of the country like much of the Southeast where, unfortunately, if you're born into poverty, your odds of escaping poverty don't look that great.

That kind of background gives us a lens to understand the question you asked, which is, "What is it that leads to economic mobility? What's driving that variation across areas?" This is a complicated problem, so obviously there are many different factors at play.

We find that a number of things matter. Some things you might expect intuitively, like the quality of local schools, access to higher education, access to job training programs, things like that. But another important factor that I think might actually be the most important one is what we call social capital or economic connectedness — basically, the degree to which low-income people are connected to high-income folks. We measure that using Facebook data on millions of people, basically asking among low-income folks on Facebook what fraction of their Facebook friends are high-income.

It turns out that [social capital] is the single strongest predictor of economic mobility that we or anybody else has identified. Perhaps, if you're connected to higher-income people, you get jobs that help you get ahead or your aspirations change because you see career pathways you wouldn't otherwise have experienced.

Really, it's a complex mix of factors. But I think some of these factors, like who we're connected to, are extremely important to think about.

Sablik: You mentioned there are these geographic differences across regions. Has economic mobility also changed over time for the U.S.?

Chetty: If you look at the long sweep of history going back to the middle of last century, your odds of doing better than your parents used to be extremely high. For kids born back in the 1940s and 1950s, it was a virtual guarantee that you were going to do better than your parents. If you look at kids entering the labor market today, people around 30 years old now, it's become a 50-50 shot as to whether you're going to do better than your parents did.

So, in that sense, the American dream has receded. But it hasn't receded equally for all places for everyone. As I was saying, there are still many places in America where if you're born to a low-income family, you have great chances of rising up. But on the whole, I do think it's getting harder to do better than your parents. That underlies a lot of challenges that the U.S. is facing at the moment that people are frustrated about.

Sablik: Thinking about some of these key factors that you mentioned like education and social capital, what does economic mobility look like in rural places? What are the characteristics of rural communities that positively influence mobility and ones where maybe there are challenges that are harder to overcome there than in urban places?

Chetty: We've been spending a lot of time thinking about that question the past few weeks as our team has been preparing for this presentation I'm giving today. It has been very interesting to look at the data for rural America.

The pattern you see is not uniform. In the rural Midwest, the center of the country — think of places like Iowa or the Great Plains —those rural areas look terrific, among the best in the U.S. [and] best in the world in terms of your chances of rising out of poverty. If you look at the Southeast, unfortunately, in much of the rural Southeast you have very low levels of economic mobility.

What's driving that difference? It actually turns out that it goes back to the very factors that I was describing earlier. People have praised places like Iowa for the quality of its schooling system. That is, fortunately, increasingly changing in places like the Southeast. You may have heard of something that people call the Mississippi Miracle, with improving test scores in states like Mississippi in recent years, So, education plays a role.

Another key explanatory factor is social capital and this measure of cross-class interaction. When we look at the Facebook data we see that in many rural areas, particularly in the Midwest, you actually have a lot of cross-class interaction.

Imagine you're growing up in a small town. My wife grew up in a small town in rural Illinois, for example. Her dad was the town doctor. She went to the local high school. Her friends came from a variety of different backgrounds because there was only one school and everyone knew everyone. If you imagine that same family living in New York City, as a doctor, you're sending your kid to a very different school than kids from different backgrounds. You can see how there's much more splitting by class in a dense urban area.

So, one thing I think rural areas have going for them is a natural interaction that is created simply because you're in a smaller community where people are more connected to each other. Now, that appears to play out very favorably in the rural Midwest. In the rural Southeast — partly because average levels of incomes are lower — even though people are interacting with each other, the degree of exposure low-income people have to higher-income folks is much less, and so economic connectedness is much less. Segregation tends to be higher in the rural Southeast and I think that's part of what contributes to the lower levels of economic mobility there.

Sablik: Speaking about the Southeast, the Richmond Fed's district encompasses some of that area. It includes Maryland; Washington, D.C.; West Virginia; Virginia; and the Carolinas. Did you have a chance to look at rural communities in those states? How do they compare to the whole picture?

Chetty: When we zoom in on those states, we find a picture as what we see in the Southeast: currently relatively low levels of economic mobility compared to other rural areas in the country.

But I want to emphasize that we should not think of this as fixed. These things can change over time. We are starting to see changes in some parts of the Southeast. We're starting to, unfortunately, see some reductions in economic mobility in parts of the Midwest and the West Coast. What that shows us is all of this is malleable. By making the right investments and the right policy changes, we can potentially create opportunity where it's currently lacking.

Sablik: Should we think of economic mobility and geographic mobility as the same or intertwined? Or, are there key differences between those that you would like to highlight?

Chetty: There is a link between those two. Much of the high mobility you see in the rural areas, where we see very good outcomes, is achieved by people who are leaving those areas as adults. With our data, we're able to trace people back to the place where they grew up. A lot of those kids who are achieving those really good outcomes in a place like Iowa that I was citing before moved to Chicago. They've moved to New York. They've moved to other cities when we're measuring their incomes as adults.

That said, we're still finding better outcomes in the rural Great Plains than in the Southeast among the people who stay. But a big part of the advantage is driven by those who leave. More generally, as you'd expect intuitively, rates of leaving are much higher in rural areas than in urban areas, and in particular in the rural Midwest than in the Southeast.

Sablik: Yeah, I wanted to ask about that because it seems in your framework, outmigration is counted as positive economic mobility. Some of the rural communities that we've talked to might view outmigration as negative. So, I'm wondering how you think about that.

Chetty: It depends upon what kind of question you're asking. It's not that, per se, we view outmigration as a plus. All we're doing is just measuring people's incomes as adults no matter where they live. We're just observing that a lot of these kids who are growing up in these rural areas and doing well, they're doing well somewhere else.

From a national point of view, it's great that people are doing well wherever they live. Of course, from a rural community's point of view, you have a brain drain concern. These folks whose talent you'd like to retain, maybe help the next generation build new local businesses, have now moved somewhere else. They've taken their talents and their resources to other places.

In a new study we're going to be putting out shortly — looking at where America's business founders come from, where do America's entrepreneurs come from — there is a very interesting pattern. People who grow up in rural areas are much more likely to start businesses than people who grow up in urban areas, but many of those businesses they start are in urban areas. They move to those urban areas. So, that's a clear example of that brain drain effect. And so, I think rural areas are absolutely right to be concerned about that.

I think there's a further concern in the context of the tax base that's very important. These folks who you've invested in, they've done really well because of the rural community. They are now living in a different place, possibly in a different state, paying high income tax rates on what they're now currently earning. But none of that money is flowing back to the rural areas. That's a fundamental fiscal problem that I think we need to solve.

Sablik: So, if you were to sit down with rural leaders who want the residents in their community to succeed, but they also want their community to succeed, where would you tell them to focus their efforts?

Chetty: As with a lot of these things, I think the solution varies depending upon the context.

Let's start with the Southeast and the areas in the Fifth District that you all are most focused on. What we're seeing in the data is kids are not having very high rates of upward mobility and the economic prospects for adults could be improved. So, I think there you need to take kind of a two-pronged approach of figuring out how you invest in local schools [and] job training programs, giving people the skills that may then attract businesses to want to invest in local areas.

Sometimes people think of this as I just need to give tax credits or attract the businesses directly. But businesses follow talent. If you see that there's a talent pool of people who you could hire potentially at a lower wage rate than in urban areas, you may want to come to these places. Especially with remote work and modern technology which has reduced the costs of distance, I think this is very doable in the modern era.

In places like the Midwest where it seems like kids are doing pretty well but they're moving away, it's kind of a different problem that you need to solve. I think the school system is working. Social capital seems to be good. Fortunately, some of those problems have been addressed. How do we then retain that talent? How do we provide job opportunities for folks in those local areas? How do we draw capital to invest in those places? Those are the kinds of issues to focus on.

Sablik: What's the current frontier of your research? What questions are you grappling with right now?

Chetty: One thing I'm very excited about is a new set of methods we've developed to measure the impacts of various programs that can help people advance economically — for example, workforce training programs or education programs. We're able to isolate the value added of those programs without having to run a randomized trial, which is the traditional way economists and policy analysts would figure out if something works.

Just to give you a concrete example, we have 5,000 different job training programs in America. Many of these operate in rural areas and a number of them, obviously, in urban areas. We really don't know which of those programs work and which of them don't. The challenge is that traditionally we've thought we have to run a randomized control trial to figure out which program works and which program doesn't. Those trials are incredibly expensive. There are ethical challenges to doing that.

What we've recently found is, using the very large scale data our team is working with combined with modern machine learning [and] artificial intelligence techniques, we can form what we call digital twins for each person in the data. Say you participated in a workforce training program. I can basically construct a synthetic digital twin for you using data from a bunch of people who look like you, drawing from the full population [of] tax data [on] 270 million American adults [to find] somebody very similar to you who did not participate in the program. I can then follow both of your earnings over time and then figure out when you participated in the job training program or higher education program [and] you got a particular degree, how did that change your earnings trajectory and change your life relative to your digital twin?

What we're finding is that that approach gets you results that are very close to what you get from a true randomized experiment. That's incredibly powerful because now we are able to implement this on scale and evaluate across rural areas, across urban areas, which things are really working. The next step is [to] allocate money towards the programs that are actually effective and retool the programs that are less effective, really create a market basis here that I think could greatly amplify our efforts to increase economic mobility.

Sablik: Sounds super interesting.

I mentioned we're at the Investing in Rural America Conference where you're giving the keynote address. What about this conference caught your interest, and why do you think it's important to talk to this audience about your research and your findings?

Chetty: I spend a lot of time studying economic mobility and essentially all of the talks I've given in the past, say, 10 years on this issue have been in urban areas. Fifteen percent of Americans live in rural places — 40 million people, roughly — so you would think that we should be devoting at least 15 percent of our time to thinking about the challenges in that setting. So, I welcome the opportunity to think about those problems, which I felt our team and the research enterprise and the public more generally does not spend enough time thinking about. I'm looking forward to sharing what we've learned about opportunity in rural America.

Sablik: Well, Raj, thanks so much for joining me today.