Valentin Bolotnyy on the Gender Pay Gap

[0:00:00 Speaker 0] mm hmm. Welcome to Policy and McCombs. A data focused conversation on trade offs. I’m Carlos car value from the Salem Center for Policy at the University of texas at Austin. Yeah. So today is March 29 2021 and our guests economist Valentin Bulletin from Stanford University’s Hoover Institution Valentine. Welcome to policy McCombs. Thanks very much for having me on. So before we start going on on details of the two papers that we’re going to talk about today, uh, tell us a little about a little bit about your research portfolio and generally what you’re interested in is an economist. Yeah, thanks again for having me on. Um, I think of myself as a public and labor economist interested in a broad range of topics. Those two fields allow one to be, um, pretty spread pretty wide in terms of research interests. So I’ve worked on the gender earnings gap and trying to understand where that comes from. I’ll be presenting on that paper later this afternoon. Um, I’ve worked on graduate student mental health and getting more and more into research questions around mental health and how they’re connected to our other social um uh issues from gun violence too. Um unemployment and the opioid epidemic, for example. Um and I’ve also uh you know, partnered with government agencies on on other issues, for example, why is it that infrastructure costs are so high, especially in the US. Um and I think that is also a fruitful direction, not just for researchers like me, but for for students, uh partnering with public sector entities um to work on questions that have to deal with public service provision and improvement of public service provision, but also at the same time increase our understanding of how societies um and uh social interactions function and helping students also um get great projects out of the work uh, towards their degrees. So, um, that’s that’s generally where I stand research twice. Okay, Before we go into the papers, you mentioned something that I want to follow up with. Why why is it that infrastructure projects cost so much in the U. S. What have you found so far? Well, yeah, it’s a really it was a load of big questions. Yeah. And and and it’s a great question with many different parts to it, but I think um one of the things that we’re finding is the devil’s in the details where uh not spending enough on maintenance uh not and when we do it’s usually in batches as opposed to a continuous flow. Um So what we have is a lot of projects that are, you know, bridges and roads and uh you know, water infrastructure that is really behind uh in in uh maintenance and and and it’s quality which then results in having to do really drastic measures, even more drastic than maybe the entity that’s managing them, realizes. Um So that’s one piece, the other piece, there’s been some really good work on this, on the role of citizen voice, you know, design right projects, but then interest really kind of uh halfway midway through the project ended up hijacking the project and essentially getting their way getting uh the aspects of the project that they really wanted initially, but didn’t get um they end up getting getting forcing those in through political economy channels and and that, you know, is a delay that is a cost over on all of those things start start colliding. So, those I think are our main the main issues. Um there’s also, you know, kind of a lack of uh competition issues and uh that that’s uh also part of the part of the part of the story. All right. So let’s move on to the topics of today. So, this afternoon, and as you mentioned, you’re gonna be presenting on why do do women earn less than men evidence from bus and train operators? Um, So I I spent a lot of time I teach statistics to, to to our students generally. And and I always speak as the big gap is a great example of something that we start with the big number. Right? And then we start thinking about where is the number coming from? This number that tells the whole story, right? It’s a great teaching tool when there is a is a tool that that oftentimes the devils is said is in the details. Um, and you have the super interesting paper that gets evidence from a very specific situation that can isolate something. So let’s go through it. Give us the big picture of it. That’s right. So we focus in on one particular workplace. Um but we think that what we find in this particular workplace has general Izabal lessons for the rest of the United States for what’s happening as you’re saying um in the big picture. And the two main forces that we find driving the gender earnings gap is first lack of flexibility in the workplace, kind of rigidity and scheduling. Um and secondly, very related uh feature our social norms are, you know, still very much the fact that women have higher demands on their time outside of work than men do. Um And and those two things interact in our workplace to generate the earnings gap that we see. Um and we think that those are issues that are affecting the country as a whole. So what does that workplace? Let’s let’s go through it through it carefully. Yeah, this this workplace is the massachusetts Bay Transportation Authority, the public transit entity that serves the greater boston area. And we’re looking at the employees that we’re looking at our bus and train operators. Um The reason the t is such a good place to study the gender earnings gap is that it’s actually a very controlled work environment. Everything about the way that work is structured is written out in a collective bargaining agreement between the union that uh backs up the workers and the public secretary entity, the transit agency. And what that collective bargaining agreement says is that everything should be done by seniority, the routes that people get to drive. That’s those are choices that are offered to operators by seniority, the hours that they get to work similarly by seniority. Um, wages cannot be adjusted by due to performance or other metrics. They’re just adjusted automatically given a schedule that is in the collective bargaining agreement that, you know, is irrespective of performance or any other kind of managerial potential managerial discretion. So what we have here is a place where we can cross off a lot of the usual hypothesis that get discussed when it comes to gender earnings gap. Managerial discretion is not part of the issue. Occupational sorting is also not part of the issue because we’re looking at one workplace tasks, type of job in particular is one type of job. Exactly, yeah. So the tasks we know here, the tasks are literally identical. Everyone’s driving the same kinds of buses and the same kind of streetcars. Um And uh you know, there’s no need, for example, for workers to negotiate or bargain for wages or promotions or anything like that. And it’s a component identify another research right. Of the of the associated with the pay gap, the sort of aggressiveness negotiation. Exactly. Exactly. Um and and so here, even though we we don’t have these issues at play, we’re still finding a gender earnings gap that is about two thirds of the national gender earnings gap, about an example, about 11%. Um And that’s a dollar. Exactly. Exactly. Um And so we we dig into why why that’s the case and we find that even when we look within the same seniority, so men and women who are facing exactly the same kind of choice sets within this workplace, we’re seeing women take more unpaid time off from work, especially women with dependence. And we’re seeing women accept fewer overtime opportunities than men. And over time here is paid at time and a half. So there’s an extra wage boost to these hours that women are foregoing. And overtime care is offered by by seniority as well just as everything else in this workplace. So the routes the routes lead to no difference in pay is just the time and when you work is the amount of time you work and when you work is what actually leads to the difference in pay. Exactly. So uh the and really when you work is also not part of the equation, so if you’re working a night shift where you’re working uh you know three hours in the morning and five hours late at night, for example, you’re still being paid the same amount. Um Over time is really as a way of filling in shifts that are going unfilled. So what one kind of perverse thing about what’s happening here is that women are more likely to be taking leave and that opens up slots that are then paid at overtime rates that men are mostly filled. Um So that that’s that’s what we’re finding going on here. All right. So, so let’s be uh double some specifics here. Um So you mentioned you mentioned the taking taking over time. So the decision to to take over time versus not. Right. Uh So that you in the paper, you studied this carefully and you try to look at the probability of going in and and and and taking over time. Uh And that probability is dramatically affected by gender, but not only gender cross with a few catch races. So go through that a little bit. That’s right. So we’re finding that the people who are earning the least in our setting, people who are most likely to not accept an overtime opportunity when it is offered to them. People who are most likely to be taking unpaid leave uh In our setting, they are unmarried women with dependence. Um And the people who are most likely to accept overtime opportunities and the least likely to take leave are married men with dependence. Um So what we’re finding is that through a number of other steps that women are uh they have more demands on their time. Especially we think through um childcare and other caretaking responsibilities. Um And men especially uh married men. They are uh they are caretaking in a sense through working more hours. And that’s where the social norms that are out there and outside of the workplace are interacting with the construction of of work inside the workplace. Um There is no way for operators to trade shifts with within the Npt. So um once you pick your schedule by seniority for the next three months, you can’t really move that around. Um And what we’re seeing, we think is that women uh turn to unpaid leave as a way of generating flexibility in the workplace that otherwise doesn’t offer, it doesn’t really have it, whereas men have less of a need to do that and in fact are more willing to work these these extra hours. Um So that’s that’s that’s what we’re finding. So, okay, specifically talking about the pay leave. So I hope you understand here there’s a little bit more. So they have f what is it? F. M. L. A. Right? The family medical leave. Um And and they have the ability to use that. But even though the strategic use of that relative to them paid leave, it plays a role here. Right. Right. So, so the unpaid leave here is largely FMLA. Um So there’s no, okay, so there’s no there’s no okay, so there’s no paid leave associated with this with this job, right? There is there is some limited amount of sick leave, but that actually that the number of hours of sick leave that you have that are paid um come into play more and more later on in your career, they also increase with seniority. So what you have here is especially the junior folks, right. Especially the younger people who might be starting families are in that process. Um They are the ones who have fewer days of paid sick leave, uh and they are the ones who are stuck with the most inconvenient schedules, weekends, holidays, other times when childcare might not be available. Um And what they end up turning to is uh family Medical leave Act, which is, you know, there’s no there’s no paid leave for caring for Children or caring for older adults, older relatives, for example. Um In the US we just have really the Family Medical Leave Act, which is protected unpaid leave. So you can take unpaid leave from, from work, but not fear having to lose your job in the process. And that’s what we’re seeing women use at much higher rates than men and that and that contributes to the gender or anything. Right. So I know that was not the focus of the paper were just wondering about the about that that that that process right? Of demanded by social norms, demanding more flexibility in scheduling has a particular effect on the junior folks. As you mentioned here, they have less flexibility in the choice at right. Um, so that do you see in the data a clear sort of dropout rate for women in that job? In the beginning that’s higher than men, which then would lead to a different aspect of the there’s some components seniority that I’m assuming that you’re going to have more senior folks in the force that are going to be male versus female. That is that does that happen there as well? That that is happening? Yes, the women are slightly more likely to leave um earlier than um than the pension eligible uh age than men were finding. Also when we look at 2016 when the T decided that, you know, it was going to kind of clamp down on leave taking um and F. M. L. A. Was used in the workplace um as a friday to monday Leave Act, that’s what it was called among the operators. So the t. Was taking steps to clamp down on that kind of leave taking. And what we found is uh women taking as a result a lot more unexcusable e not coming into work at all without giving notice um uh to to their employer. And so that that’s um you know we’re finding that actually clamping down on the availability of of leave and on the flexibility within the workplace, but that is a big contributor to into the gender earnings. So uh interesting because there’s a there’s a sentence you have here is that when they clamp down, think there was a 2016 change in policy in 2017 change in policy, you saw a reduction on the pay gap From 12 to 9%. But then you you you stayed at the same time there. Well there was a reduction well being for the female workers as a result as well. Exactly explain that a little bit. Exactly. So uh what what we’re seeing is that F. M. L. A. After this policy change after it becomes harder to take leave through the family medical leave act, people do in fact take less of it. Um But uh there’s a rise in unexcused leave, that’s not quite one for one. Uh So FMLA falls by more uh then unexcused leave rises. I apologize for the noise behind me. Um But um in the process, uh, you know, it’s costly, unexcused leave is especially costly to women if FMLA is excused and they don’t have to fear about losing your job, then excuse me. They can actually face layoffs or suspensions. Um So they’re turning to costly early, which is uh worsening their welfare. Um And uh and uh their their earnings gap is shrinking right there. They’re forced to come into work more and they are actually working slightly more hours. Um But welfare is decreasing and uh the tea is also not gaining from the perspective of service provision because managers get less of the heads up with an excuse leave than with F. M. L. A. And it’s harder to schedule routes harder to make sure that they’re they go Phil. Um So as a result, what we find is that service actually didn’t improved. Uh and female welfare decreased even though we had a shrinking of the gender earnings camp. So that we think is quite important when when workplaces think about what they’re aim is what they’re going for. It’s actually possible to shrink the gender earnings gap and make it seem like you’re making progress on gender equity. But actually in the process make women were soft. Mhm. Comes to mind a policy that we have universities that um we give everybody in the tenure track uh line of work equal leave associated with a child. And typical I think the typical you get actually you entertain your clock right? And as a result of a child, I think the research recent research that shows that actually improves increase the productivity of males in terms of the type of C means they built by the time you go up for tenure, but decreases the one for females trying to help right at the same time, you create this imbalance as a result, right? Trying to be general neutral. Hell. But then you you make it you make it uh the consequence again, because the flexibility and the demands at home are different between those social norms. Right? Exactly. Exactly. I think that’s a great example. And uh this is one of the conclusions of our papers that um employers should think about more than just um looking within the same occupation, within the same job and trying to make sure that they’re equalizing wages between men and women. There are other factors that might seem tangential to the gender earnings gap, and you might seem completely fair like seniority and are setting where it is completely gender neutral. Um but the way that it interacts with the rest of social norms and what’s happening at home for these individuals, that might be actually what’s what’s driving the gender earnings gap uh and and and welfare in our study. So flexibility seems to be like this huge, huge important component of it. Right? So, so um one of the recommendations you have a discussion you have is that work workplaces figure out ways to be more flexible that will generate will be like you might still end up with the with the pay gap in the sense that that, you know, choices are such that might lead to a difference, but the overall well being might be actually better as a result, right? The way people devoted their time elsewhere, uh, cost flexibility is crucial as the cost of flexibility goes down and uh, this workplaces embrace, uh, flexibility more. That should uh, that should make welfare that should increase welfare and likely decrease the gender earnings gap by allowing women especially to work more hours. Right? And I just want to connect a little bit to some of the work that has been done before on. I think the NBA work that might remember trend and co authors have done on that. Right? I think it shows something similar as well there, that it partly connected to the to the notion that folks in the general level in particular by lacking flexibility, you tend to drop out at higher rates, right? So you might start at the same pay level and as things progress, especially at that age where there’s more demands at home because it’s a time of childbearing and so on. The probability of Vincent changes dramatically as a result of it. And even though for the ones that stick around to get some might not be there later on, but there’s a huge change in trajectory, right, as a result of elective flexibility so that that those two things connect very well. Yeah. And, and you know, that’s a setting that’s looking at MBS, right? And in our setting, we’re looking at bus and train operators who have a high school education. Uh, and you have the same rigidity and inflexibility in the workplace, um, generating the same kinds of kinds of patterns, the gender earnings camp. So let me ask you a difficult question in some ways. Think about, we observe also on the surface pay gaps not only engender, but we also observed that in race. One thing that you are identifying here, it’s very clear is this effect of unmarried women with dependence in particular, have this bigger this bigger gap right now that correlates very strongly with race, unmarried with the Bennetts. Uh So are you familiar at all with that that any research that looks at that uh in the in the I guess I guess the gender. So then you have a sort of double layer issue going on for one group of people that end up being like showing up in some statistics, that’s very, you know, uh is punished by the labour market in a way that it has to do severely with those social norms or just the the realities of of of of their their their household situation. Yeah, I think we can extend what we’ve been talking about to to understand uh these these differences as well. Um If we think about what are the kinds of demands that might be facing women more than men. Um It’s it’s caring for Children in caring for elderly parents and relatives, uh also in this covid world that we’re in um even teaching right uh is an additional piece to caretaker. Um And what we’re finding through through the surface that are coming in is that women are much more likely to be taking on these responsibilities. We’ve seen a drop in labor force participation, especially among women. Um And that is uh you know, I think, cool that also to kind of keep thinking about what’s happened happening recently, um the uh there folks that have been able to work remotely have been able to um transition smoothly into the covid kind of economy. Uh and those are usually folks uh earning higher incomes and you know, with with higher levels of education and uh the folks who are uh in the service sector, working at lower paid jobs, um they’re the ones who had to choose between staying in the workplace or coming out completely and and focusing on caretaking. Um and the social norms are interacting there as well if you have more demands as a result of just having lower income and having, you know, we’re also, we know that lower income folks and people of color have been harder hit by the pandemic by the health impacts. So now if you’re also dealing with that piece, taking care of sick relatives and potentially chronic health issues coming into play, um that’s all going to given social norms that we have, that’s all going to land on the lapse of women, much more so than the men. And I think that’s a I think that can explain quite a bit of of not just what we’re seeing gender wise, but also with the racial component as well. So before we move on here, just uh social norms, I think a lot of things we tend to try to find blame and trying to find a corporate and try to find a way to go and solve it. What your thoughts on on like not thoughts, but just like what do we know about our ability to or even even transition? Like what do we know about the, how this this social norms have been changing in time and therefore might be the reason why the gap is changing. It is shrinking time, right? Is that because our companies are being more aware of it, our companies are being better behave, or is it because somehow the structure at home has been changing also as culture changes and therefore, you know, things are more equal, you know, at home starting there. So um what do we know about? Yeah, it’s I think it’s both happening in parallel. We’ve had information technology uh become quite versatile and quite widely used to the point where there’s certain professions, pharmacists, for example, are one um that essentially have no gender earnings gap. Um and that’s because of changes in the technology that workplaces used, that allow for greater flexibility in the workplace. Um But we’ve also seen social changes, the birth control pill, for example, has allowed women much more control over how they manage their professional priorities and personal priorities. We’ve seen women get more and more going into college and getting higher education degrees. Um and we’ve seen, you know, women are still doing a lot more household work than men, but but there’s been more more rebellions, certainly now than, you know, uh three or four decades ago, for example. Um So, so that’s, I would say, uh, progress. There’s been progress in terms of couple equity at the same time that we’ve had, um, progress in gender equity, the labor market. Um, There’s some evidence, for example, in Quebec, um, where they’ve uh worked with the way that they provide parental. They’ve experimented with certain approaches. And one thing that they’ve done is offered paid five weeks of leave that they’re calling the Daddy Quota leave that only men can take and and benefits that only men can receive if they take, um, if they take parental leave. And there’s preliminary research from that that shows that not only does take up of uh, the increase when you set certainly besides specifically for men, but also several years down the road. Uh They do more household chores and they spend more time taking care of the kids. So there’s a certain way I think in which incentives can be sent by public policy um to help us move in that direction of greater couple equity and um to nudge norms uh in in that in that direction. Alright. So final question on this topic, um there’s a recent paper that looks at the gender earnings gap on another public transportation device which is on Uber. Um And can you comment a little bit on the similarities of your results to their results? I think I’m looking at their their abstract here and in there, the identifying experience is a big component uh constraints over where to work. So I think something through safety in particular and men being more willing to work in unsafe locations, uh and the final one is driving speed. So apparently the choices again risk taking from from males lead to more more more runs, I suppose in the system. That’s right, that’s right. That uh setting is in some sense the exact opposite of ours. In that with Uber drivers have complete flexibility, complete control over their schedules, whereas in our setting, operators are very constrained. Um And as a result, I think that’s why in this Uber paper, what they’re finding is that uh you know, the key differences are have to do with kind of the nature of driving men driving faster than women. Um or with the fact that uh if you just look at kind of the whole pool of drivers, men have been driving for slightly longer than women, uh and there have greater experience with where to go when the rates where the rates will increase at which times. But I think when they control for that experience, that part of the story goes away for them. So that’s uh that’s those are the kind of I think the differences but that is one flexibility is 11 big difference. And that’s a good example of a space where technology uh really, you know, takes takes flexibility. Um It’s a very high level and still choices remain. Right. That’s that’s the part that I think it’s it’s it’s a very interesting as well is that there’s this notion that somehow we can make it disappear everywhere. And like well the group differences and choices and preferences. And that’s might be true because of social norms maybe choose because of risk tolerance. There are things that you know, will persist no matter what policies we have in those two examples, you have very gender neutral policies in in in the in the uh the tea and Uber and still you’re going to observe that, right. That realization I think is an important thing as well is that, you know, unless we reduce those, I don’t know how to change that part of risk aversion. For example, if there is a difference between the groups, I don’t know how to fix that if that leads to higher pay in some uh some settings. Right, right. I think the question is not whether um you know, we’ll ever get to a gender earnings gap of zero, But but whether 20 or 18 we’re looking at now nationwide uh anyone and something like 30 40 for people of color, women of color. Um that that’s you know, are we at an optimal now? And I think the evidence generally suggests that that we’re not. So let’s move on to mental health in particular. You you wrote a paper recently on the mental health of graduate students, graduate students and economics. Um so yeah, tell us about that. Why, why did you choose that set up? Was a set up just because were you familiar with uh you know, where you just finished grad school or or was that any any specific reason for that? Yeah, I started the project when I was a graduate student myself um at Harvard, and uh this was winner of 2015 and there was a suicide in the Mighty Economic Speech program. A graduate student was also in the third year, just as I was, and our cohorts were very close. Uh and at the time, we were also lucky to have departmental leadership and um just general interest and energy among the students to understand mental health issues among graduate students, better to understand why something like this could happen in our community uh and and to do something about it. And I ended up collaborating with another student in the in the Harvard Economics Program and with paul Barreiros, until recently, the head of the University Health Services. And the way we approach the problem is the way I think economists uh generally approach these things. We went out and collected data, analyzed it and tried to see if there were some policy suggestions that we could we come up with what we found when we did our surveys. Um is that uh they’re really high rates of depression, anxiety, um even suicidal ideation among economics PhD students, We found rates that were 2-3 times. Um what we would see in the general population of uh individuals, kind of in the same the same page range, uh, mentally 20s. Um and just last question there two or three times, and that controls for like the sort of level of stress on the job or or a similar types of jobs as well. Or? No, just generally for the cohort age cohort? Just generally for the age cohort. Okay. Generally the age cohort, um there’s a additional work that’s been done. Harvard now has a graduate student mental health initiative, that’s university wide, where they’re doing similar surveys across other graduate departments and generally what they’re finding there are even higher rates than what we found in the Economics Department. Um uh Prevalence rates of moderate and severe symptoms of depression anxiety of 20 plus. Um And and we’re finding something like 17 18%, the economics programs. So, uh that’s uh, you know, we that’s really high, that’s really high uh in uh in terms of just, just in in in absolute terms and in relative terms that those are really high numbers. Um And just just to to um again, immediately go into the selection. The selection aspect here, is there any reason to believe that graduate students, just the type of people that decide to go into PhD programs and or even people that do the things in life in order to be a graduate student at Harvard in particular. Right. Um Is there a selection that leads to people that have more tendencies of of of of uh mental health problems? Or or not? There’s no reason to believe that there is there is slight selection in that uh we see this when we asked students, um, were you diagnosed with the mental health issue prior to coming into the PhD program? And those rates are slightly higher than what we would expect, but not enough to explain the the largest numbers, you’re fine. Right, right. Exactly. And we’re also seeing that um students towards the end of their PhD program are experiencing um much higher rates of uh these symptoms then at the beginning of their PhD program, even though when we look at the cohorts, uh they’re coming in with a similar similar prevalence rates of mental health issues. So there’s there’s kind of course, this is uh this is not a causal study that we perform, we, you know, collected data and correlated things. Um but uh there’s, I would say it’s kind of a preponderance of evidence to suggest that uh this time spent in uh impeachment programs um has has has a significant negative effect on mental health. Um We dove in to some of the reasons why this might be the case and what we’re finding are things that have also been talked about in the US broadly and in fact in developed advanced economies broadly. Um but we think there’s a particular um application of those forces in the graduate programs, so increases in isolation uh and and loneliness, um lack of social support, whether it’s from um so students or from advisers or uh, you know, outside of the PhD program. Um a in particular, this is an issue where economic students, um, a kind of normalization of suffering uh very few economic species students, even those experiencing some of these really awful symptoms are actually going and receiving treatment. Um about two times as many um students, uh graduate students large at Harvard, for example, are who explore experiencing these symptoms are getting. Um so that’s also we think an issue and we resumed in specifically on advising relationships. Um, that’s where uh, that has in conversations that we had with students, and also in the survey data that we collected, that was, you know, that’s where they’re kind of, the sirens were blaring. Um, and what I think are most valuable insight there is that it’s not so much quantity of a number of meetings that students have with advisors or how many advisors they have. It’s the quality of the relationship. Um, and with advising. Uh, it’s tricky because advisers are both evaluators, they’re the ones signing off on the dissertation and writing recommendations for jobs. Um, There are also sources of support, or they’re supposed to be who are guiding the individual through the program and trying to make sure that they succeed. Um, but that generates that that tension generates potential for lack of trust or, you know, not, not very high levels of trust. Um and we’re finding students telling us that they really can’t be honest with their advisors about not just personal issues and mental health issues, but even research progress and what they want to do after they graduate. Um and that that kind of tension uh you know, when you’re putting your whole life or five or six years of it into, into this work, that tension can really, we think hurt mental health and and if it’s accumulating over years, that can be particularly acute towards the end of the program, when these kind of very big decisions start becoming very salient and very important. Yes. So, so, I mean, anybody have gone through grad school will realize that a lot of what you’re saying is not surprising, right? So a lot of times when you’re finding research data on something like, oh my God, that’s surprising if you’ve gone through grad school, you know, all you’re saying makes absolute sense. You know exactly how fearful can be a conversation with an advisor sometimes if you’re having trouble and so on. So I’m sure a lot of people relate to that, but um clearly the levels is what I think is very surprising. The level of of not getting to a point of just that there’s a very stressful time in your life. But to a point where now you have actually derived mental health issues associated with it. That’s what gets really scary. Um But at the same time, I think there’s a, you mentioned a couple of things that are are I think sources of variability that might be interesting to explore here. And I wonder if you have any information on that. So first you mentioned the notion of um isolation and that is particularly true in economics. So, so I come from a fuel that’s a little bit different economics, a lot of similarities, but a little bit different where we are encouraged to work together and co author from the very early days of your PhD. So, you know, it’s not uncommon to get somebody getting out of their their their their experience, a graduate student and statistics having four or 5 papers, all of them co author whether one adviser or multiple people and so on. That’s that’s not a typical and that’s sort of similar to what the model you see in science is generally like if you’re working in the lab, if you’re working on yes, you’re gonna be responsible for project, but there’s a team involved in that. Uh So is there variation in, but you did mention that it seems that Harvard is pointing out to maybe more in other in other groups, which is would be surprising to me. Um And the other one that that I wonder if you have any information on is is how much of that is there in the time series? How much of that has an impact on the job market? Right. Again, the exit aspect of it. So are the folks getting into tenure track jobs which are the most stressful time undergraduate career is gonna be, it’s gonna be that stage. Right. Are they is that being a causal link to be successful in the professorship sort of chase or not? I think we lost you there for a second. Yeah, I think the co authorship question, yeah, yeah, I can hear you, I cannot your back but Okay, sorry about that. Okay. I think the co authorship Question is a really important one. Um I think we’re yeah, I can hear you can hear you. You shouldn’t make if you can hear me okay, as long as you can hear me, that’s yes, I can hear and see you right now. Uh Great, okay, great. The the co authorship point and incentives for collaboration, I think that that’s a really that’s a really great point. Um and one that we’re working with departments on widely um co authorship has generally been discouraged in economics. You’re absolutely right about that. And I think especially, you know, for for students in the earlier years, um when you’re trying to figure out what you want to work on what you wanna specialize in if you’re not talking to your peers about you’re losing out professionally, but you’re also losing out personally. And those are the things that we really can be, things that open gates to other social connections and other conversations. I am really good friends with a lot of my co authors. Um and those initial, uh you know the way that I became friends with them was through this environment where I felt comfortable talking to them about research. That that’s the first for a lot of students, uh a kind of good icebreaker in some ways. So if we’re encouraging that kind of those kinds of conversations encouraging students to create quality social relationships with each other, I think that can go a long way, not just within the PCT program, but also professional for for many years after that. Um There are other issues I think in the lab sciences for example, and in some of these other programs that are showing higher rates of mental health issues than an economics, you might be working in a group, but if your supervisor has created an environment that isn’t very friendly, um you might have mental health issues as a result of that, as opposed to a lack of social connection from, from your peers. So this is where I think that department specific introspection is crucial to understanding what’s going on. Um And then your question about 10ure, uh I think um the one piece of evidence that I can provide is that, you know, we asked students uh huh how important are certain things to you, um from a perspective of feeling fulfilled and feeling like you’ve had a successful life and career. And the things that we listed and had students respond to word tenure tenure at high ranked academic institution. Those lists of items, and what we found is that depression, anxiety, all of these symptoms were affecting students similarly across all of these items. So if someone was saying that they really value tenure to high ranked institution that’s really important to their sense of success in life, they were just as likely to have hi uh, you know, in poor mental health scores, as those who were saying that actually, that’s not that important to them, What’s really important to them, It’s uh, you know, having a high income for exam. Um, so we think that these kinds of issues, at least in grad school, are affecting uh folks pretty um independently of what they’re um uh desirous or for next steps after grad school. Um, but but of course, there may very well be that kind of selection that you’re talking about who goes on to stay in academia. Uh could very well be. Um folks who are folks who have lower mental health scores better, better mental health scores, which is just a matter of resilience, right? That that folks that on a particular cohort might end up end up having the ability to manage that easier than or better than than others. Right? And but again, that’s a testable thing. I suppose you could look at the same survey when you go to people like one year before getting tenure. Right now, we’re looking at comparable levels of mental health. There, is there any work on that? On on on the, I would assume that there is some research done on, on faculty members um, at that stage, Yeah. You know, not not so much and not so much and not not that I’m aware of actually. Um, we try to include questions about mental health in the faculty surveys that we also administered. And we got some pushback on that. I think even with uh lots of promises of data security, uh, those kinds of things, they were to fill out mental health related surveys and provide that data to researchers. Um there’s work happening now trying to survey other departments and not not just at Harvard but work in europe that I know of and work another um and other universities, but I don’t think any of them have been able to um have been able to get faculty on board with these kinds of questions so that that’s I think whoever can do that and do that effectively I think has a great paper on their hands. So a question linking back to the previous paper, any gender effect in this data in terms of prevalence, interestingly. Uh yeah, they’re very very small difference between men and women. There usually is um in national data, women are more likely to be experiencing these moderate or severe symptoms of depression and anxiety. Um, in our data, uh, women were only slightly more likely that that surprised I’m not statistically significant surprises me a bit because we have at least anecdotally a lot of discussions about how the environment in grad school and economics in particular might not be so um accommodated, not not so welcome to women, for example. Right. There’s a lot of discussion on that currently. Uh, so I would think that that would lead to maybe more unhappiness for one group versus the other, but I guess I’m glad you’re not finding a big difference there. Yeah, that’s right. I think we’re still when we’re measuring, for example, um, uh, how easy would it be for you? How readily would you raise your hand in the seminar setting? For example, we’re finding much more hesitancy on behalf of the female students than uh, the kinds of hesitancy that male students share with us. There are, you know, sexual harassment is much more directed towards women, uh in economics departments than than towards men. There. Uh, there are these environmental issues, um but you know, perhaps they’re causing stress and other issues that are not showing up in these uh um moderate or severe symptoms of depression and anxiety. Yeah, valentin, well, thank you so much. This is super interesting stuff. I’m looking forward to your talk this afternoon. Thanks very much for having me comments. Thanks for listening to policy at McCombs. Yeah. Mhm. Mhm, mm