This is an audio transcript of The Economics Show with Soumaya Keynes podcast episode: ‘Why even a PhD isn’t enough to erase the effects of class?’
Soumaya Keynes
You’ve heard about racial inequality. You have heard about the glass ceiling. Today we are going to be talking about something that in the US hasn’t had as much attention — the class ceiling. A recent working paper argues that we really need to think about it because independently of race or gender, people’s family circumstances seem to be holding them back. And that’s the case even after they have done enough work to get a doctor in front of their name. This week, we are going to talk about the finding that even a PhD isn’t enough to erase the effects of class.
[MUSIC PLAYING]
This is The Economics Show with Soumaya Keynes. I’m joined today by Anna Stansbury of MIT Sloan School of Management and one of that study’s authors. Anna, hello.
Anna Stansbury
Hi. Thank you for having me.
Soumaya Keynes
Thanks so much for being here. OK, so first question, you are a Brit and you live in America. So on a scale of one to 10, how much of a problem do you think that Brits perceive class-based inequalities?
Anna Stansbury
Seven.
Soumaya Keynes
Seven? OK, OK. So pretty present. What about the US?
Anna Stansbury
Four.
Soumaya Keynes
Ooh. OK. So the US is not doing so great on class consciousness. I mean, what do you think the main kind of qualitative difference is in how people talk about class between those two countries?
Anna Stansbury
Well, so when I first moved to the US, which was 11 years ago and talk to fellow grad students at the time, they would say in the UK, the problem is class, in the US, the problem is race. And that’s a pithy way of explaining kind of the big salient factor that people think about when they think about social and socio-economic inequality in each country. And there are good reasons why race is a lot more salient in the US. But class is also a factor independent of race, and I think people are aware of that in the extent to which it determines whether you go to school in a good place, but not really later in the life course.
Soumaya Keynes
Right. OK. OK, well, look, let’s step back for a second, though. What exactly do you mean by class?
Anna Stansbury
So class is one of those words that is hard to define very specifically. Typically, when sociologists talk about class and they’re the academic discipline that thinks about it the most, they’re talking about the set of qualities and resources you had that determined your opportunities. So I’m talking specifically here about class background. So what was the class that you were raised in? And that’s some combination of the income and wealth that your family had as a kid because that determines your resources and your opportunities. That’s also the education that your parents had and the kinds of occupations that they worked in, because that determines some of the slightly less tangible aspects of tacit knowledge about how elite careers work and how education works and aspects of cultural capital. So whether you have access to the kinds of cultural knowledge that give you social status in certain groups, all of those factors together in some combination are what we talk about when we talk about class.
Soumaya Keynes
OK, so class as distinct from income, essentially?
Anna Stansbury
Class as much more than just income, yeah.
Soumaya Keynes
OK. All right. Well, what made you want to look into this question of whether class holds people back? I mean, I know you were struck moving from the UK to the US, but what made that into a research question?
Anna Stansbury
So part of it was this, I think being from the UK makes you think a lot more about how these factors matter even later in life. Obviously in the UK class background can be a lot more salient in things like accent than it is in the US. But I think I believed that it probably mattered in similar ways in the US as well. And then what really got me into the research question was I was doing some kind of activism, advocacy-type work on gender and race in economics, diversity. Economics is bad on gender and race, as you’ve noted in prior work. And I wondered if this was also true about class and found some data that enabled me to investigate that and that sort of set me off on this route of trying to figure out whether class matches for people even once they’ve done their education, even once they’ve got a PhD, even once they’ve got into an elite career.
Soumaya Keynes
Yeah, because I mean, one thing your paper obviously does focus on is people who are pretty much in the elites, right? You’re looking at people with PhDs. And I suppose one question is why should we care about them? Right? We know that they’re doing pretty well.
Anna Stansbury
There’s a couple of different reasons we should care, and this is true of any kind of diversity. One is equity. We should care if people have opportunities to fulfil their talents for reasons of equity and justice. But the other is a very kind of banal economic reason, which is efficiency. If you assume that talent for something is equally distributed, then we should care if people that are talented aren’t getting to fulfil that talent because it’s worse for overall productivity and overall outcomes.
Soumaya Keynes
OK. And that’s essentially the high-level question that you’re asking in your working paper, which is if you compare people, do people from different backgrounds perform differently later on in their academic careers, having got that PhD? So can you just tell us a bit more about the methodology? How do you start going about answering this question?
Anna Stansbury
Yes. So we have data from the National Science Foundation in the US which surveys a representative sample of people that got their PhD in the US. So we have basically a very large representative sample of everyone that got a US PhD since about 1993. The sample goes back to them. And so what we do is we ask the question conditional on getting your PhD in the same field from the same institution, does your class background affect what kind of job you end up in later in your life?
Soumaya Keynes
OK. So that’s interesting because you might think that your background would affect the kind of subjects that you would study and which institution you go to. But you’re kind of trying to essentially wipe all of that and say, no, we’re comparing people who are basically very, very similar, right? They’ve got their PhD from the same place in the same subject and what happens next?
Anna Stansbury
Exactly. So you can think about it as a very high bar to pass because we’re saying it’s quite plausible to believe that your class background might affect whether you manage to get into a PhD program at a really elite university because of all the prior factors in your life. But it might be much less simple to believe that your class background is gonna continue to affect you after that. And I think this is what has surprised people about these findings is that we find that it does continue to matter even if you get your PhD at the same program.
Soumaya Keynes
So just starting off, how exactly do you measure class?
Anna Stansbury
So we use parental education background. As I said before, this is only one of the aspects that feeds into class background. Unfortunately, we can’t measure the rest in our data. And this is actually why class background is rarely studied. It’s very rare for surveys to ask this kind of question, particularly in the US. So we have parental education and we cut it into four groups, people who are first-gen college graduates. So they had no parent that got a four-year college degree, people who had a parent with a college degree but no graduate degree, people who have a parent with a graduate degree that’s not a PhD. So that would be MD, doctors, JDs, lawyers, MBAs, masters in education. And then finally, people that have a parent with a PhD, and we look at those four groups separately.
Soumaya Keynes
Just out of interest. If you got to be God for a day and could gather whatever data you wanted, what would your dream measure of class include?
Anna Stansbury
Great question. I think I would ask four things, and this tends to follow what I think a lot of sociologists would ask when they ask about class. It would be parental education, ideally not just the level, but also where they got that degree. It would be family income or wealth when they were a child. Again, ideally, both of those things because you might have low income but high family wealth. It would also be your parent’s most common occupation when you were a kid. And then finally, there’s a question people often ask that is trying to capture some combination of all of these which asks you to think about social status as a ladder and then place where you think your family was on that ladder. So that’s a sort of self-perceived measure of where you stood as a child in the overall socioeconomic distribution.
Soumaya Keynes
OK. I mean, that sounds attainable.
Anna Stansbury
Absolutely. It’s definitely attainable. It’s just rarely asked.
Soumaya Keynes
OK. Well, in this survey, we have data on parental education, so that’s what you use. OK, let’s move on to what outcomes you look at, right? Because you’re looking at whether family background matters for success, what counts as success here?
Anna Stansbury
So what we look at mostly is success in academic careers. And we define this in a couple of ways. First, we look at do people end up in a tenured academic job? And then we look at where they end up, what kind of institution? So we use a couple of different measures for how prestigious their institution is. One is how research intensive it is, which reflects how much research funding you get and other kinds of opportunities an academic would want to have. And the other is the rank of the institution.
Soumaya Keynes
OK, so you’re actually looking at placement after they get their PhD? Can you look at anything else like pay, for example?
Anna Stansbury
Yes. So we actually do also look at pay. In academia specifically, the sort of currency of success is less clearly only pay than it is in many other professions, which is why we mostly look at the kinds of institutions and whether people have tenure. But we also look at pay, and then we also look separately at the PhDs who go off into private sector careers instead of academia, which is a very large share of PhD recipients. And for those people, we look at pay as the primary measure, and then we also look at whether they end up later in their career in managerial positions.
Soumaya Keynes
OK. So you’ve got a fairly wide range of things that you look at. So what do you find?
Anna Stansbury
The top-level finding is a big class gap in career progression, even for two people from the same PhD program. So we can break that down into looking at the folks that went into academia and the folks that went into industry, private sector. In academia, we find that there is a big gap in the prestige or rank of the institutions that people end up employed at as professors. So a first-generation college grad is about 13 per cent less likely to end up as a tenured professor at a research-intensive university than someone from the same PhD program who is from a more advantaged background. And specifically, this more advantaged background that I’m referring to here is someone with a parent that had a non-PhD-graduate degree. So a parent that was like an MD, a JD, an MBA. We cut out those PhD parents because we think they might have academia-specific advantages that aren’t really about class background.
Soumaya Keynes
OK, so you see this class ceiling, are you able to look at any trends over time? I mean, I would really hope that perhaps this had become better as there was more thinking about inclusion, diversity.
Anna Stansbury
Yeah. So I don’t want to be the bearer of bad news, but we actually don’t see any improvement over time. We have a relatively long time period. We have data from 1993 to 2021, and we don’t see any obvious improvement in this class ceiling or this class gap in where people end up conditional on that PhD program.
Soumaya Keynes
OK. Well, that’s not very happy, is it? I suppose it could be something to do with preferences, though, right? I mean, you know, the obvious conclusions draw from that is there’s some sort of discrimination, implicit bias against people who don’t grow up with the same kinds of resources as others. But what if it’s that, some types of people from a particular family background just have different priorities in their career? Maybe they want to go to an institution that may not be the highest-ranked institution but may have more of a social mission. Is that possible?
Anna Stansbury
Yeah. So we do a lot of work to try and understand if this is driven by preferences and we can’t find any strong evidence that it is. So one strong reason you might think, preferences or constraints play a role, is differential financial backgrounds. Academia in the US is relatively lucrative, but it’s not as negative as going into the private sector.
So you might think that people from less advantaged backgrounds are choosing to leave academic jobs or leave academia altogether and go get better-paid private sector jobs. We don’t see that that’s happening on average. You might think it’s what you raised, which is this idea of preference for social mission and being willing to trade off rank and prestige of institution for serving a less advantaged population. Sure, that happens to some extent, but we aren’t able to find evidence of that when we look at is there a difference in terms of the income of the student body that the school serves.
We also thought about whether geographic constraints might play a role. If someone from a less advantaged background, there might be more constraints in terms of living far away from family. For example, if they’re a breadwinner or a carer for a family member. But again, we conditioned on distance from home as proxied by where they went to school. And again, we don’t find that that explains the gap. So we tried quite hard to see if preferences can explain it, but on average, it doesn’t seem to.
Soumaya Keynes
But that’s interesting that people from different socioeconomic backgrounds are just as likely to sort of stay in academia, right? It’s not that, because in say, gender, it looks like women are disproportionately dropping out of academia. And that is one thing that contributes to gaps in their outcomes. But more generally, how does this class gap operate differently to the gaps in gender or race?
Anna Stansbury
Yeah, this is a really interesting one, and I think we’re just scratching the surface here. I think it’s a super fruitful topic for research. So one thing to emphasise before I answer that question specifically is that when I’m talking about the class gaps that we estimate, we’re estimating them conditional on race and gender. So what that means is that our effects aren’t driven by, for example, in the US context, African-American PhD students are also more likely to be first-gen college grads. When we’re comparing outcomes of first-gen college grads, the people from more advantaged backgrounds, we’re effectively comparing them within racial groups. So within African Americans, within Asian Americans, within white Americans, and so on.
So having said that, our results aren’t driven by race, but we can compare them to the gaps by race and gender. And as you said, with gender, the big phenomenon is what you call the leaky pipeline. Women are falling out of the academic pipeline at every stage. They’re less likely to go on to get a tenure-track job, they’re less likely to get tenure. But actually, if you look at the women that stay in academia, there’s not that much of a gender gap in what kinds of institutions they end up employed at. They’re at similarly ranked, similarly prestigious places.
The class gap looks the opposite. There’s no leaky pipeline. There’s no sense in which first-gen college grads are less likely to stay in academia. But when they do, they end up employed at these less prestigious places.
Race gaps. It’s shameful to say, but unfortunate, but still relatively few African American and Hispanic professors in our sample. So we estimate these with a lot of noise because it’s hard to get a precise estimate if you have a smaller sample. But the race gaps look more similar to the class gap, where you see these big gaps in the institution, rank and prestige.
Soumaya Keynes
But less so in the staying in academia?
Anna Stansbury
Yes, exactly.
Soumaya Keynes
Just a quick question. Can you do anything to compare between subjects? Are some subjects worse than others? I’m thinking about economics, that is the name of the show.
Anna Stansbury
So we do try and compare across subjects and we just don’t really find much going on there. Our data’s Stem and social sciences. So we compare the physical sciences, the biological sciences and the social sciences. The patterns look pretty similar across all three groups. We don’t have a super large sample of economists in this data, so we can’t say for sure that economics is not worse. But we don’t see any strong evidence that economics is worse.
This is different, by the way, to some of my prior work, which is just looking at the composition of people by class in fields. And in that sense, economics is worse. So the way you can think about it is when you look at PhD students, economics really stands out as having the lowest share of first-generation college grads of any PhD field in the US. So it’s doing really bad in terms of class diversity. But once you’re taking people who’ve already got their PhDs and looking at how they progress, economics then doesn’t look any worse than the other fields in terms of having disparate rates of success.
Soumaya Keynes
So just on that, do you have a theory as to why economics is so bad in attracting a diverse group of students?
Anna Stansbury
Yeah, I think there are probably three things going on and I don’t know what role proportionately to this place. One of them I think is that economics is a subject where it’s not exactly obvious what it is or what it involves. And so you see that even at the college level, first-gen college students are less likely to major in economics in the US or to take an economics degree in the UK. Probably in large part because it’s not clear what economics is, what it does, what it’s useful for, and studies that sort of just provide information and nudges do find big enrolment increases from all kinds of minority under-represented groups, not just first-gen college grads, also women, also racial minorities. When there’s more awareness of what economics is and what it does.
The second is that economics has a big drop off from people that major in econ to people that then go on to graduate school to get a PhD, particularly for first-gen college grads. And I think that is probably because economics has really good outside options. So if you get an economics degree, you can get a very well-paid job after college. And that might be disproportionately attractive to people who have come from backgrounds that have less financial security. So I think that in some way, economics is a victim of its own success, specifically on that metric.
The third one, which I think is almost surely playing a role that have not been able to measure, is, I think, the way economics is taught and some of the language and inherent assumptions in sort of econ 101 and a lot of the economics profession can feel quite hostile to people from less advantaged backgrounds. And some of the language we use can feel quite off-putting, some of the assumptions, especially in their core econ 101 style teaching. So just to give one example, referring to workers as unskilled, referring to people that choose not to go to college as low ability in these models. Those kinds of terms feel offensive and inaccurate, I think. And when people have lived experience or family members who would fall into those terms in the econ vernacular, I think that could feel like this is not a subject for me. This is not a subject that describes the world accurately.
Soumaya Keynes
OK. Well, a lot for economists to think about there. Just building on this idea of outside options, though, and returning to your paper about outcomes for PhD holders. I know that you said that there wasn’t really much difference in who decided to stay in academia or not. But how does academia compare to other professions? Do you see what happens to people if they do choose to go into government or the private sector? Is the class gap still there?
Anna Stansbury
So we look at the three other big destinations for PhD recipients. That’s the private sector, government and jobs in education that are not tenure-track academic jobs. So this could be teaching jobs at universities, at community colleges, or the kinds of education jobs. In industry in the private sector, we also find a class gap. Now, our measures here are slightly different because obviously, we don’t see whether someone is a tenured professor because that doesn’t exist in industry. We look at someone’s salary and how that progresses over their career and we look at whether they end up in positions with managerial responsibility. And we find a class gap in both that widens over the course of their career. So when PhD recipients, you know, first get their job in the private sector, there isn’t negative class gap in outcomes, but as they stay in these private sector jobs over their career, you see the salary gap widening and you see that people from first-gen college grad backgrounds are less likely to end up as managers. So that suggests to us pretty strongly that there’s a similar class gap, class ceiling dynamic in the private sector.
Soumaya Keynes
And what about the public sector?
Anna Stansbury
So the story is more optimistic in the public sector. In government and in non-tenure track education, we don’t find a class gap in salary or in managerial responsibilities. So it doesn’t look like, at least on those metrics, there’s a class ceiling.
Soumaya Keynes
Do you have any thoughts on what might be going on there?
Anna Stansbury
So this is just speculation because we don’t have enough data really on what jobs people are doing in these other sectors to know exactly what’s happening. My guess is that particularly in government, there are more standardised pay scales and kind of promotion requirements that make it less easy for disparities to creep in. We also see smaller gender gaps in pay in government than we see in the private sector, although that’s not the focus of our paper. I think it’s an interesting comparator, but I don’t know for sure.
Soumaya Keynes
OK. Well, I think it’s time to throw to a break. But look, this is a controversial topic. And when we get back, I want to ask about the response to your research. So what it has been and what you wish it had been and what you think should be done next?
[MUSIC PLAYING]
[BEHIND THE MONEY PODCAST TRAILER PLAYING]
Soumaya Keynes
We are back from the break. So what has been the most common critique you’ve heard, most common pushback against your paper?
Anna Stansbury
So one of the common pushbacks has been, how does this even happen among two people who’ve got their PhD from the same program? Hasn’t the effective class all kind of been washed out by having got this elite degree at this elite place? And I think depending on your experiences, that is either very obvious that those effects may not be washed out or not very obvious at all. And for me, I think maybe coming from the UK made this more salient, but I come from a relatively advantaged background. My parents both were qualified as lawyers and I was able to see throughout my education and then coming to the US how various factors that I received from my upbringing made it easier for me to basically exist and take up space in these elite places. And so part of the process of doing the paper has been talking to a lot of people who come from different backgrounds, who are in academia to try and ask them what their experiences are about, whether the progression has been affected by that class background. And we’re actually about to run a survey on this as well.
Soumaya Keynes
That’s super interesting. What kinds of things have you been hearing?
Anna Stansbury
So one of the big factors is this sense of ease. Ease is a word that is, you know, often used by sociologists when they’re talking about what does an elite education give you? And I think it’s a really good word for this circumstance, because to have a career, a successful career in academia, but also in lots of other industries, you need basically to seek out mentors who are elite people in their own field and get advice and sponsorship from them. You need to be able to network effectively in specific kinds of spaces that you may not be used to being in. This relationship building, this seeking out of advice and mentorship, I think can be much easier for people that have been raised in environments where they’ve been doing this kind of thing from an early age in gauging in more elite spaces from an early age. That’s something we’ve heard a lot.
Soumaya Keynes
Can I ask about the idea that this look at academia is relevant for the broader economy? Because I suppose in one sense you have this incredible data, you’re able to look at outcomes quite precisely. And so, you know, that looks like, Oh, this is a really good test! But in another sense, actually, within academia, there’s huge amounts of subjectivity on what is good research. So many places for implicit bias to creep in. So isn’t it possible that really academia is some kind of upper bound for the effects of class on your outcomes? It could be an area that’s really, really vulnerable to those kinds of biases.
Anna Stansbury
Yeah. So this is one of the other things that we’ve heard from talking to people and thinking about how it works in academia is these judgments that enable you to get the prestigious grant, get your paper published in a great journal, get tenure at a top institution. These judgments are based, obviously, they’re trying to be based on your research, but they’re inherently subjective because someone’s trying to see not just if you produce decent research, but if you are brilliant. If you have that touch of, you know, spark that makes you the academic genius that people want to have and promote and see. Those kinds of judgments are hugely subject to bias. And I think there’s a lot of extent to which the way you speak, the way you dress, the way you act, those kinds of things that can be affected by your social class background can be used by people subconsciously or consciously as markers of genius or brilliance rather than what they are, which is just markers of, you know, where you went to school and how you were brought up. So I think in academia, this is probably one of the other mechanisms by which we’re seeing this glass ceiling, this class gap emerge.
You said, is academia an upper bound? Now, I don’t know because I think academia does have this space for these value judgments, these subjective judgments to be made. But we do have very, very detailed, transparent, quite objective measures of research, quantity and quality as well. You can see, you know, you can read people’s papers, you can see what journals they’ve been published in, how many citations they have and all this kind of stuff. Compare that to something like professional services or law, where you’re working in teams, you might have less objective measures of one individual specific outputs. And these are settings where being able to do all this kind of social networking, form relationships with elite people is probably even more important than it is in academia, because you’ve got to impress elite clients if you’re a lawyer or a consultant or something like that, as well as the partners at your firm. So my sense is I would guess that academia is, if anything, maybe a lower bound on these effects relative to some other professions. Ideally, we would be able to have this data and so we wouldn’t have to guess.
Soumaya Keynes
Amen. Economists call for more data. That’s the first time that’s ever happened.
Anna Stansbury
Sorry to be a stereotype.
Soumaya Keynes
It’s OK. We’re all there. OK, but now I want to ask about the, so what? I mean, what do you think should be done as a result of these findings?
Anna Stansbury
So it’s quite a boring answer. But I think some of the most obvious concrete policy changes are incorporate social class background in the diversity initiatives that we already had for race and gender. It’s not rocket science to do things like track your applicant pool, your pool of employees or PhD students and how well they do to have mentorship initiatives that also incorporate someone coming from a less-advantaged family background, as well as incorporating someone who’s a minority gender or minority racial group. There are a lot of things that we know work relatively well. Don’t do everything. They’re not silver bullets, but work relatively well to advance sort of people from less advantaged groups in careers like academia that are just not really being done for class backgrounds. So there’s a lot of low-hanging fruit.
Soumaya Keynes
Do you see any movement in this space? Are there any people doing this kind of thing?
Anna Stansbury
Yeah, there’s some going on. So in academia specifically, some of the professional associations now have, you know, groups and committees set up for first-gen college grads, people from low-income backgrounds in the same way that they have committees for women or minority racial groups. I’m doing some work to look at other private sector firms. I knew very, very few private companies currently collect data on report on or target class background in their DEI report. So all the big companies have DEI reports where they’re tracking gender and race. Very, very few track class. But over the last few years, the number that tracks class has gone from basically zero to, you know, few, but not zero. And so I think there are companies that are now trying to incorporate this into their hiring, into their diversity practices. So there’s change, but we’re still on the early end of it, I think.
Soumaya Keynes
Yeah. One of the things I was actually quite impressed by when I moved to the FT, was that I was asked about this kind of information, so . . .
Anna Stansbury
Well, I should say in the UK, companies have been more proactive about this. In the UK there’s been more action over the last 10 years or so for companies to track this and monitor this and do more about it in diversity.
Soumaya Keynes
Just thinking about one final comparison. Do you have a hunch of whether this class gap would be higher in the US or in the UK?
Anna Stansbury
My initial guess would be higher in the UK because class is more directly visible. So all the mechanisms we’ve been talking about. About, you know, is someone perceived to have the polish or the brilliance required to do this job will get the opportunity. That stuff is all magnified, I think. And there’s a great book by two sociologists, Dan Morrison and Sam Friedman called The Class Ceiling, which looks at the UK and UK occupations, which I highly recommend to read for any listener. But the one caveat I would say is that in the US, inequality on basically all dimensions is higher, income inequality, even within professions, the inequality of really status of being, you know, a manager at a really, really super big, super successful company versus a medium-sized company or being tenure track at a top institution versus a middling institution. All of those gaps are magnified. And so having a small advantage in the US might actually translate to a bigger disparity later on relative to the UK.
Soumaya Keynes
OK. Something to think about Americans. Take it away. Ponder. And that is where we will finish this week. Anna, thank you so much for joining me.
Anna Stansbury
Thank you so much. It was great to talk to you.
[MUSIC PLAYING]
Soumaya Keynes
That is all for this week. You have been listening to The Economics Show with Soumaya Keynes. If you enjoyed the show, I would love it if you could rate and review us wherever you listen.
This episode was produced by Edith Rousselot with original music from Breen Turner. It is edited by Bryant Urstadt. Our executive producer is Manuela Saragosa. Cheryl Brumley is the FT’s global head of audio, I’m Soumaya Keynes, thanks for listening.
You must be logged in to post a comment Login