Equitable Growth in Conversation is a recurring series where we talk with economists and other academics to help us better understand whether and how economic inequality affects economic growth and stability. In this installment, Kate Bahn, the director of labor market policy and interim chief economist at the Washington Center for Equitable Growth, speaks with Jonathan Fisher, the new research advisor at Equitable Growth. His areas of research include inequality, mobility, and personal bankruptcy. Previously, Fisher was at the Stanford University Center on Poverty and Inequality, where he was a research scholar and a senior researcher on the American Opportunity Study. He has also worked at the U.S. Census Bureau and the U.S. Bureau of Labor Statistics.

Kate Bahn: Thank you for taking the time, Jonathan. I’m really excited to get to know your research better and also dig in to how it fits in to Equitable Growth’s mission. But to start, I want to look at a really broad overview of your research, where you’ve examined the interrelationship between inequality of income, wealth, and consumption. I think not a lot of economists are tying together all three phenomena at once, and so I think that’s a really important contribution. And I want to dig into why it’s so important to connect these different economic phenomena when we’re trying to understand trends in economic well-being.

Jonathan Fisher: Thanks, Kate. I’m excited to be here, too. Economists understand that income inequality is going up. Consumption inequality has been going up, too, though there is a bigger debate about it in the economic literature. And wealth inequality is clearly going up.

What I want to understand is whether this one-dimensional view of inequality—looking at income or consumption or wealth—is it really reflective of overall economic inequality? And the sense I get is that inequality is going up across all three of these dimensions among the people who are already high income. But how is that changing over time in terms of higher wealth and high consumption among these individuals?

What I and my co-authors are finding, using the Federal Reserve Board’s Survey of Consumer Finances, is that the percent of people who are in the top 5 percent of all three of these measures—income, consumption, and wealth—has gone up since 1989. And I can make it more concrete by putting the percentages to it.

Take the people who were in the top 5 percent of income in 1989. About a third of them were also in the top 5 percent of consumption and wealth. In 2007, just before the Great Recession of 2007–2009, half of those in the top 5 percent of income were also in the top 5 percent of consumption and wealth. So, this group has become more exclusive. Who is at the top of the income, consumption, and wealth ladders are really closer to being the same people.

Now, this convergence of economic inequality among the top 5 percent did decrease slightly after the Great Recession. In 2016, the most recent year for which we have complete data, around 44 percent of the top 5 percent of income were also in the top 5 percent of consumption and wealth. So, our economy and our society is growing apart not just in one dimension, but in all three of these dimensions simultaneously.

Bahn: I want to dig into one of the aspects that you’re looking into, which is consumption, in particular. You note that there’s a debate in the economic literature around consumption inequality. This is an area that probably I know the least about because it’s generally understudied. But, that being said, consumption is an indicator that really gives a sense of someone’s well-being and overall standard of living. In doing your research, what illuminating conclusions have you found about how consumption patterns interact with other factors—in particular, racial and ethnic differences? Does one’s demographic background really shape these outcomes for consumption?

Fisher: You’re exactly right about consumption. I think what really motivates research on consumption is that when economists think about the poverty threshold, we measure it using income. We measure poverty in income, but then we ask, “Well, what’s the minimum amount of money people living in poverty need to consume, to spend, to meet their food needs, their clothing needs, their needs for shelter, their utility needs?” Economists measure these things using income, but what we really are thinking about is how much money they have to spend. And that’s why I’ve been particularly interested in looking at consumption.

One recent paper that I researched with [Georgetown University economist] Bradley Hardy looks at consumption volatility. We measure how much consumption changes from quarter to quarter. Using the U.S. Bureau of Labor Statistics’ Consumer Expenditure Survey, which interviews people in four consecutive quarters, we look at how consumption changed over those quarters. One particular focus of ours was across the income distribution. We find that spending on food is much more volatile for low-income households than it is for high-income households. It’s particularly volatile among the bottom 20 percent of income earners.

Food doesn’t last. That’s why we think this spending really does capture differences in the level of spending and how much people are actually able to eat. But we also see volatility in spending on apparel and in personal care items—things that help people maintain their appearances.

And all of this increased volatility could come at a cost to those individuals. Take food. When people aren’t eating enough, it has negative health effects and negative health outcomes. Or consider clothing. If people need to spend money to have sharp appearances at job interviews and yet don’t have sharp clothes for those interviews, then it could affect their chances of getting a job or retaining a job. And simply having clean clothing that is not degraded is associated with boosts in self-esteem and school attendance for kids. So, consumption volatility in apparel can have real negative consequences for low-income individuals.

Hardy and I then look for differences in consumption volatility by race. Unfortunately, the Consumer Expenditure Survey is relatively small, so we can really only look at differences between White individuals and Black individuals. But what we do find is that even at the same income level, Black individuals have higher consumption volatility in food, apparel, and personal care than White individuals.

We think that this has to do with differences in wealth. There’s been discrimination, historical and ongoing, which prevents Black people from accumulating wealth. And we think this lack of wealth is leading to higher consumption volatility for Black individuals. If there were a higher-quality measure of wealth in the Consumer Expenditure Survey, then we could look at it as well, but the survey has really limited information.

Bahn: I’ve also seen research showing how personal grooming matters a lot for wage levels, too, and it’s not just racial differences but also gender differences. Basically, this research shows that if you are from a more subaltern group, the wage premium associated with attractiveness is partially due to higher personal grooming standards, particularly for women workers. Personal grooming standards matter for getting a wage premium, but they matter more if you are less likely to be seen as professional in the first place, which means you need to groom to an even higher standard.

Fisher: Yeah. That is super interesting.

Bahn: At Equitable Growth, one way we try to demonstrate how inequality subverts, distorts, and obstructs economic growth is to zero in on economic mobility, or the ability of either individuals to move up the income distribution within their own lifetimes or of families to move up the income distribution across generations. What does your research find about the impact of economic inequality on economic mobility?

Fisher: In ongoing research I am doing with [University of Michigan economist] David Johnson, we look at intergenerational mobility using income, consumption, and wealth. And then, we look at differences in income mobility and consumption mobility by the level of wealth of the parents. What we find is that kids who grew up in high-wealth households are more likely to have higher incomes than their parents at the low end of the income spectrum, and they’re less likely to have lower income than their parents at the top end.

So, if you’re living in a high-wealth household, but your family’s income was low, then you’re more likely to move up. And if you’re living in a high-wealth household, and you’re at the top of the income distribution, then the kid is less likely to move down. Wealth and income together have that reinforcing effect. Wealth has this reinforcing effect on income that is not captured in income alone, and so it’s really adding a lot of explanatory power to understanding differences in intergenerational mobility than if you just look at one of these measures alone.

Economists have always hypothesized that wealth can have this effect. If you’re living in a high-wealth household, then your parents can just support you when you’re an adult, or they can help you pay for college. They can help you start a small business and help you buy a first home. All of these things can help kids and are missed when you’re just looking at income or consumption alone.

Bahn: Digging a little bit further into your research you’ve done, I’d like to discuss your work with Stanford University sociologist David Grusky on occupational mobility. You two are developing a new longitudinal dataset using the U.S. Census Bureau’s American Community Survey. Could you tell us a little bit about that project and what new insights you’re able to bring to our understanding of people’s labor market experiences by doing these new data linkages? As someone who researches occupational mobility, I’m really excited to hear about what you’re doing with better data that you’re creating than I’ve been able to tap before.

Fisher: What we’re doing is using the Census Bureau’s 2000 decennial long form, which no longer exists, and its successor, the American Community Survey, which replaced it in 2003, to look at parents and kids living together when the kids were 12 to 17 years old in 2000 and what the parents’ occupations were at the time. And then, we can observe the kids, in the American Community Survey, when they were in their 30s and look at their occupations.

When looking at intergenerational mobility, sociologists tend to focus on occupation. It captures a lot that may not be captured in income. It captures class. Obviously, it does capture income as well, something economists tend to look at.

So, the first paper Grusky and I are doing together is research looking at occupational differences by race and ethnicity. With this much larger dataset, we can look at differences for not just Black individuals but also for ethnicity because the Census Bureau’s decennial long form and the American Community Survey have this question: “What’s your ancestry?”

This means we can see Black individuals who say their ancestry is Jamaican, or Hispanic individuals who say they are Cuban. In this way, we can separate out those who identify as Hispanic and Cuban and who tend to be more highly educated from Hispanics who identify as Mexican or Central American and who are more likely to be lower-educated. And we can do the same thing with Asian Americans. We can separate out those who say their ancestry is Japanese and who tend to be more highly educated from those who are Vietnamese and tend to be less so.

With these data in hand, we can look at differences in occupational mobility between different race groups based on their ancestry and not just compare Asian American individuals to Hispanic individuals to Black individuals. And we, not surprisingly, find a lot of variation within race and ethnic groups that is missed in the typical research that groups these individuals together more broadly. We also do the same thing with those individuals who identify as White and find a lot of variation within what typically is thought of as a single race or group.

Bahn: You may not have gotten to this in your research yet, but have you found any correlation between occupational segregation within these different categories and the likelihood of occupational mobility across lifetimes or generations?

Fisher: No. We haven’t. It’s something we’ll have to think about, if there’s a way to measure that. Another paper that we have planned may get at this. We plan to examine immigration status by generation. We would be able to see, for instance, whether the parents were born in the United States and if their kids were born in the United States, and then see if there are generational differences. If, for example, both you and your parents were born outside of the United States, then does that make you less likely to change occupations?

Bahn: That would be interesting, too. Some of those occupational segregation patterns are generational too. We know people are more likely to go into the same profession or occupation as their parents. There is a generational component to occupational segregation. I’m really excited to see where this research goes. This project really demonstrates why data advances have been really critical to our understanding of economic inequality. Being able to look at much more fine-grained analysis of different groups based on their country of origin, their race, their generation, all of which previously was not available to a lot of researchers, is very informative.

Fisher: Yes.

Bahn: But there still remains a need for improving data provided by U.S. statical agencies. Can you describe some of the shortcomings of our current statistical information, and what you think some key reforms would need to be so that economists like you can do better research?

Fisher: Let me begin by offering two preambles. First, I worked at the U.S. Bureau of Labor Statistics for 5 years and the U.S. Census Bureau for 4 years, and I know everyone is really mission-driven at both of these agencies. I don’t think outsiders necessarily see that dedication, so I want to say everyone is striving toward better data and better access to the data as well. So, what I have to say next is not criticism but, rather, ways we can improve upon the data and ways internally that the staff and leadership at these two agencies probably want to improve.

And my second preamble is this—it’s not just economists who are using the data. A lot of other disciplines are using it. Sociologists are using the data, and political scientists and other disciplines, too. So, there’s a large group of people who are using it. With those two thoughts in mind, I see three ways that there can be improvements to the federal statistical system and the statistical agencies.

One is something we’ve already discussed. It’s hard for smaller data samples and smaller surveys to be able to speak to differences by race. If you have to call Asian Americans all the same race, or you say Hispanic Americans are one ethnicity, then you’re missing a lot. And you’re also missing if you can’t look at Hispanic men versus Hispanic women, or Asian men versus Asian women. Obviously, researchers across many disciplines would like to go even deeper than the smaller sample sizes that hinder a lot of deeper understanding of the sources of economic inequality and who’s experiencing inequality, or who’s experiencing the benefits of economic growth.

Second, in my research that looks at consumption inequality, wealth, and income together, I have to impute consumption to the Survey of Consumer Finances. There is no federal dataset that captures all of these components in the same survey. The University of Michigan’s Panel Study of Income Dynamics does, but it’s a private dataset, and even though it has lots of amazing strengths, it is a relatively small survey. It’s been following the same people since 1968, so that’s amazing for the work that I’m doing, but it’s not part of the federal statistical system, and it would be fantastic if one survey could have all of those data together so we could really understand, from the federal statistical system, economic inequality along these dimensions.

And the last thing in the way of improvements I would suggest is more data sharing among agencies and then sharing that data with researchers. Right now, the Census Bureau can access IRS tax records but other agencies cannot, which hinders some important research that the Washington Center for Equitable Growth is supporting, such as looking at the distributional national accounts, or our GDP 2.0 project.

If the U.S. Bureau of Economic Analysis had access to tax data, for example, then it would really aid those efforts to understand where the fruits of economic growth are accruing in the United States—where, more specifically along the income and wealth distributions, are those fruits accumulating and where in the country is that growth occurring. And then, if researchers had access to that data in a confidential and secure manner, it would help further our understanding of inequality and how inequality is hindering growth in the United States.

Bahn: That would be great. As someone who does not work for a university, I would love that, too. It’s hard to get access when you’re not at a university.

Fisher: Yes. Access is designed for university researchers.

Bahn: So, switching gears, I just want to say that I’m really impressed with breadth of all this research you’ve done on economic inequality across all these factors and with all these co-authors and across disciplines. It’s really inspiring to me. I want to understand a little bit of how you first got interested in economic inequality as a topic. And since starting on the journey of becoming an economist researching these topics, how have your research interests changed, or become more refined or more expansive?

Fisher: I first became interested in it when I was working at the U.S. Bureau of Labor Statistics and started working with David Johnson on using income and consumption to measure U.S. economic inequality and intragenerational mobility, and really saw that there was a lot going on and a lot of interesting questions. He and I started working together on our first paper in 2002 and really saw that increasing economic inequality was a particular issue for the United States and deserved more research and had a lot of interesting research questions there.

And as I continued my other research on personal bankruptcy and wage inequality, particularly for Black and White women, I saw how economic inequality was infused in those topics as well. And when I recently applied for the job at Equitable Growth, you asked me to summarize my research and how it fits into what Equitable Growth is doing, I realized that when I really step back and think about it, inequality was the theme through all of my research, even though maybe at the time that I started, it wasn’t planned that way.

It really is the case that economic inequality is affecting all aspects of household behavior and household well-being. So, this kind of research is something that I’m really excited to continue at Equitable Growth. Now that I see how all these pieces fit together, I can do it in my own research and then also help other researchers put all these pieces together and foster that research that can help inform policy and help make a difference. And I’m happy to say that 50 percent of my job is dedicated to continuing my own research, which is one thing that attracted me to the position. It really shows Equitable Growth’s support for research.

Bahn: You recently joined Equitable Growth as our first-ever research advisor and, as you’ve mentioned, our organization brings academic research to bear on pressing policy debates, concerned particularly with economic inequality. So, this, obviously is going to involve you working closely with our academic network to both accelerate and elevate their research. Can you describe how your unique role will help our academic network engage in our mission and what your goals are in expanding what we do as an organization?

Fisher: I think there are several ways that I’m going to be able to do this. I think a good way to start that really excites me is the ability of Equitable Growth to enable research. I’ve realized over the past few years that I really enjoy enabling researchers to do their jobs. It’s something I did at Census Bureau, where I helped researchers find the right data for their questions and then get access to that data. I viewed those research projects like teachers view their students, helping them along to their ultimate goals. I continue to follow those researchers and their research in the specific projects that I helped gain access to Census Bureau data. That’s something I look forward to doing here at Equitable Growth as well—helping find the projects and helping researchers get grant funding.

Equitable Growth, of course, will not be able to fund everything, But there are other ways we can support and help elevate research even if we haven’t funded it. I want to help connect people to the right data and to the other researchers. I was talking the other day with Sheridan Fuller, who is a doctoral fellow at Equitable Growth. He’s using public-use data right now, and I have a ton of ideas for him for restricted-use data that could really help him answer the questions and pursue his research agenda. Being able to connect people to the right data or to the right researchers with whom they can partner is something I’m really excited to continue doing here at Equitable Growth.

There are several other things I’m excited to do that are more specific than just supporting research. One is mentoring either doctoral students or graduate students or early career scholars. It is something I’ve always enjoyed doing and will continue doing here at Equitable Growth. I also want to help Equitable Growth in its mission to support and promote diversity, equity and inclusion in economics. When we have more diverse teams, more diverse voices in economics, we’re going to improve the discipline. There are ways that Equitable Growth is already supporting that, and I want to continue those and also look for ways to broaden and improve the support.

Bahn: That sounds great. And it does take that expertise that you bring, having been someone who has done the research yourself for so long, too. Sometimes those connections may not be so evident, that someone could do maybe a different research subject but knows the dataset really well and so you might want to connect them to that person.

Fisher: You’re exactly right. There are lots of great datasets and researchers at the Census Bureau and the Bureau of Labor Statistics. You just need to know what the right dataset is and who you need to talk to at the agency to be able to use it or to find out how to be able to access it.

Bahn: That’s great. Well, thank you so much. It was really great to hear about your research, really good to hear about what you’ll continue to do at Equitable Growth, and now, I’m even more excited to continue working with you.

Fisher: Thanks, Kate. This was awesome.