Transcript: RFTRW S08E04

Professor Alex Bryson

• More information about this podcast

Transcript

00:00:02 Female voiceover 

You're listening to an IOE Podcast, from the UCL Institute of Education. Powered by UCL Minds. 

00:00:22 Male voiceover 

This is Research for the Real World. Conversations with researchers about the paths they've taken to to shape our everyday lives. 

00:00:43 Dr Sam Sims 

This is Research for the Real World, I'm Sam Sims. I'm a lecturer at the UCL Institute of Education, and on today's episode I'm talking to Alex Bryson, Professor of Quantitative Social Science at UCL's Social Research Institute. Alex researches labour economics, employment relations, and conducts evaluations of government policy. 

And he's affiliated with the Institute of Labour Economics, or IZA in Germany and the National Institute for Economic and Social Research in London in the UK, among other institutions. He's also co-editor of Industrial Relations and the Journal of Participation and Employee Ownership. Another important thing to know about Alex is that he's an excellent teacher and I know this because his students have told me. 

We're going to talk to Alex today about the way in which women and ethnic minorities experience discrimination in the labour market and the role of unions and governments in mitigating this. Alex, welcome to the podcast. 

00:01:42 Professor Alex Bryson 

Thank you for having me Sam. 

00:01:44 Dr Sam Sims 

One of your main interests is around discrimination in the labour market. What do we mean by that term and tell us how you came to be interested in that subject? 

00:01:53 Professor Alex Bryson 

I've been interested in labour markets and workers' experiences and how people enter the labour market. Since I was at college at University in Bristol many years ago when I studied politics and sociology and then I moved in to do a Masters in industrial relations. So always thinking about who's hiring whom and why? What's the role of workers? Human capital. That is, investments they make in their education, their work experience. Because we assume that whether they are hired by a firm will depend very much on whether they're the best person for the job. Given the match between the worker in the firm and that means you know being best for the job means having the right experience and skills and wages should also reflect those experiences and those skills essentially, whether people do well in the labour market depends what they bring to the employer. 

And there's a firm-specific aspect to that, and there's a generalizable aspect that you can move across the labour market and move up the labour market over time. But what happens when actually people don't get their just desserts? What happens if they're not hired for that job even though they were the best candidate? Or what happens if in addition to being hired you are not paid what you think you're worth is, or you're being paid less than seemingly equivalent workers, and that's where discrimination could be one possible reason for that, and it can be discrimination by an employer who may not like you. 

That's called taste-based, discrimination. And indeed if they do take you on their dislike might mean they are prepared to pay other workers more than they are to pay you, so that's taste-based discrimination and it might not even be the employer who's making that decision. It could be your co-workers who don't like you, so co-workers can discriminate against you in the way that you're being treated, and so can customers. 

The customer might ask for a white or black waiter at their table, for example, and that can have implications for you as a worker. So that's one form of discrimination. The question is, how do we actually observe it and identify it in data? How do we know whether it's driving your probability of being hired or how much you're earning? 

00:04:34 Dr Sam Sims 

You know, so you meet somebody who's a bit of a sceptic and they think, how can we untangle labour market discrimination from all these other reasons why people might get paid different amounts? Perhaps there are just innocent explanations. You know what kind of evidence would you point them to on the extent of discrimination in 2020? 

00:04:55 Professor Alex Bryson 

Yes, well, I think it's easier to identify it in whether or not you're hired and I'll come on to wages, which is a little bit more difficult. 

So in hiring, the big question is you want to know whether or not who you are in terms of gender and race, for example, affects your probability of being hired. Would think of that is discrimination because it shouldn't be linked to a worth in the labour market. It's not about your education, your qualifications, it's about your race or whether you're a man or a woman. But how do we actually identify that when, as you rightly say, Sam so many other things might be going on. The way that analysts do this, normally they may be sociologists. There may be economists is that they undertake an experiment in that experiment, normally means effectively randomly assigning gender or randomly assigning race. How does that happen or what we use is something called correspondence studies. So we send out a bunch of CVs to employees in the real world and those CVs are identical except for in one respect where we've randomly assigned to be a man or a woman, for example. 

Then we collect the number of callbacks they get from the employers, and we determined that because everything else is identical other than, let's say, in that case gender, we can attribute the difference in the number of callbacks from employers to the gender difference. 

So you can randomly assign gender. You can randomly assign race. You could randomly assign sexual orientation. And so the assumption is that as a result of that experiment, you've captured the feature that determines your probability of a callback. Of course, it's not quite the real world in the sense that you're not going forward and being hired for the actual job. It's a real job in the real world, but all we calibrating it on is callback probabilities. 

So that's pretty clear example, there's another very famous one in the top economics journal, undertaken by Claudia Goldin, who's a professor of economics at Harvard. In their case, they did a field experiment, so something that happens in the real world and in period one people are recruited to an orchestra to play a string instrument if I remember correctly. Possibly a viola or violin, and in period one the people hiring musicians who observe the gender of those who are putting themselves forward for recruitment. But in period two there's a curtain between those who are hiring and those who were playing. 

And those who are hiring do not observe the gender. Often musicians and they show in period two the gender differential that appeared before the curtain went down disappears, so that's quite suggestive evidence it from a real world field experiment. 

00:08:10 Dr Sam Sims 

So the idea here Alex, correct me if this is wrong is that we're going to send the same CVs out to these employers with the same qualifications, the same employment history, and so on, but we're just going to change the name to suggest that it's a man or a woman, for example. Or to suggest that it's an English person or an Asian person and then measure the results. Is that right? And what kind of results do we see? 

00:08:38 Professor Alex Bryson 

That's exactly right, and let me tell you something very horrible that we discover from this or a man called Anthony Heath who's at Nuffield College, Oxford, has recently undertaken a meta analysis of these studies, these correspondent studies 

00:08:54 Professor Alex Bryson 

They were looking at the hiring probabilities by ethnicity. So meta analysis brings together a whole bunch of surveys of a correspondent studies undertaken over a long period of time. 

00:09:08 Professor Alex Bryson 

And the meta analysis identifies what the differential is for each study, and then pull some more together to see whether that's, for example, changed overtime. What do they discover? 

Despite the introduction of legislation in the UK precluding a home discrimination in the sense that it's actually illegal, they find no change since the early studies in the late 19th century. 

There are still very, very substantial differences in the callback rates. 

For individuals who are observational equivalent, except for with respect to their race. So that study which appeared in the British Journal of Sociology in 2019 is a clear indication that not only hiring discrimination appears to continue today, but it hasn't really diminished very much despite legislative change in the UK over that period, because you will recall that there's been a race relations act that's supposed to prevent this. 

00:10:18 Dr Sam Sims 

And so that's just the first stage of the hiring process. You know, just the paper sift. What kind of evidence do we have about once people get into the interview, for example? 

00:10:28 Professor Alex Bryson 

Well, there's less evidence on that. I mean, there's a lot of psychology literature about the signals that hirers live off when determining whether or not to recruit an individual. 

But most of that literature is about the degree of noise that's associated with hiring. So a very famous psychologist who actually won the Nobel Prize in Economics, Danny Kahneman argued on the basis of this. It was normally best to hire on the basis of paper rather than face to face interviews, but there's not much literature on that, although there is quite a lot of literature about what happens to you once you've been hired, particularly with respect to wages, but also in terms of things like promotions. 

Treatment by customers and that sort of thing. One thing that we should draw attention to actually is that it's not always discrimination against women vis a vis men for example. So in occupations which are dominated by women often the hiring discrimination goes in the other direction where men find it difficult to enter. 

00:11:37 Dr Sam Sims 

And can you give some examples of that Alex? 

00:11:40 Professor Alex Bryson 

Um, well, I think there are some examples in terms of. Well, actually I don't know specific studies of actual occupations, but I have been made aware of it by my colleague Thomas Breda at Paris School of Economics who's been actually looking education-related scenarios. It wouldn't surprise me for example. 

Although I can't recall any actual study that this might occur in situations like primary schools, where there tends to be domination by women in teaching, but you know, obviously that's an empirical matter, but certainly I'm aware of the fact that there are studies, and I think some of them were in education, which suggests that the discrimination goes in the other direction. 

00:12:25 Dr Sam Sims 

And so, how far does discrimination go to explain the gaps in wages between men and women, or for other minority groups? 

00:12:35 Professor Alex Bryson 

Wages is a much, much more difficult thing to capture. Essentially, once you've been hired, the assumption in the law there is a requirement for the employer to pay people equal wages for work of equal value. 

00:12:54 Professor Alex Bryson 

Now it's contentious as to what constitutes equal value, but also under the law, let's say it's where women versus a man. A woman must find a man doing a similar job or a job of equal value in the same workplace. That's sometimes difficult in itself. Why? Well, if there's hiring discrimination going on, you might not even get a similar person from the opposite sex in the same workplace. But let's assume that you do. 

We do know under the law that there have been a number of very successful class action cases taken by women who've argued that their jobs are of equal value to men in the same workplace, but that has not been recognised by the employer, and these have led to some very important legal cases. For instance, in the retail sector, oftentimes you'll find that it's men in the warehouse being paid more than women who are stacking shelves in the store, and these cases have often successfully challenged very large retailers and led to very big payouts. 

So that is an indication that this sort of thing is going on in the world, but if you're an analyst with micro data, a big data set where you've got men and women and you've got their wages, and you've got all sorts of things about their characteristics, their education, their work experience. Guess what the number one thing is that you don't normally have in a data set which is actually fundamental to what economists would assume determines your pay? 

The answer is your individual productivity. Your output per hour or output per minute. It might have a quality component, it might have a quantity component, but we very, very rarely as analysts have data on differences across individuals in terms of what the actual productivity is. Without observing that key feature of heterogeneity across individuals, differences across individuals. 

It's very difficult to establish whether the employer is truly discriminating against them on the basis of the wage that's being paid. So oftentimes analysts make assumptions in economics is called the residual component in earnings. That is, the earnings component that you can't account for by observed differences between men and women. 

It's probably associated with discrimination. 

00:15:46 Dr Sam Sims 

Right. That's kind of everything that's unexplained... 

00:15:49 Professor Alex Bryson 

Everything that's unexplained, but of course it's not necessarily a credible assumption, you see. Because men and women can differ across a whole range of dimensions other than their productivity that can lead to earnings differences and this has spawned a very large literature and I'll give you a couple of examples. One is the preferences of men versus women for risk taking. 

And risk taking sometimes receives an earnings premium in certain jobs. Another is the extent to which men and women are prepared to compete with one another. Third is the bargaining ability of men versus women. So there's a whole literature on whether seemingly equivalent men and women are prepared to go to the employer and ask for a wage hike. 

And there is some evidence to suggest that men have a higher propensity to do that than women. And if the employer concedes that wage hike it could come across as discrimination on the part of the employer. It would in fact, it may have its origins in men asking and women not asking. In fact, the most recent literature on this suggests that women do ask but they don't get. 

And that particular paper, I think Amanda Goodall at City University was involved in that. Yes, she was with Andrew Oswald at the University of Warwick. It suggests that yes, women do ask, they don't receive, that brings it back to the employer, which again, is potentially suggestive evidence of one mechanism by which discrimination may occur. But there's another really key one in the literature that we haven't even talked about yet, Sam. I've talked a lot about taste-based discrimination, but the other sources is called statistical discrimination. 

And one thing as I've said, analysts frequently do not have information on how productive the man is versus the woman. Neither does the employer most times, especially if they haven't hired them yet. 

So what do they often do? They lack information on just how productive you are and they make an assumption based upon often stereotypical views of how productive you are likely to be based on your observable traits, whether you're a man or a woman, whether you're black or white, which zipcode or postcode you live in. 

And by doing so, they make an assumption that your productivity is linked to that of your group. Now take that a stage further, imagine a scenario which an employer is interested in hiring a worker, and that employer hopes that worker will stay with them for some time, but the woman's of child-rearing age. 

And the employer makes various assumptions about child caring responsibilities in the household and says, well, I'm going to assume that down the line this woman might be more interested in discontinuous employment. 

That is to say, flexible working hours and opportunities to leave and then return. Maybe I want a simple life. I'm going to stick with a man. Now that decision is made purely on the basis of stereotypical views about the preferences of men and women and what their productivity might be going forward. 

Arguably that some form of statistical discrimination, of course that statistical discrimination might be true based on averages. But it also could be based upon prejudice based stereotypes. That's actually the cutting edge of the literature at the moment. 

00:19:47 Dr Sam Sims 

And that's that clearly sounds unfair, Alex, because you know the individuals being you know, are assessed on the basis of other people. Is that illegal under UK law? 

00:19:58 Professor Alex Bryson 

Well, I mean, it's certainly testable, but the idea is that you should not be discriminating in terms of hiring and pay based purely on well, as there's actually a number of metrics under the Equal Pay Act and its revision in 2010, and they include gender. Now it in a sense, it almost doesn't matter what the basis for that discrimination is, whether it's statistical or taste-based. If there is a non-justifiable distinction in terms of hiring probabilities or wages around gender or race or religion, then that's illegal. Now there are only one or two key exceptions to that under the law. 

One is when your race or gender is a pertinent consideration with regard to the job demands. So you can imagine a scenario where it might be perfectly reasonable for an employer to say I wish to hire only women if we are fitting clothes for women in a clothes shop, yeah? 

That's reasonable, but there are one or two cases in which positive discrimination is allowed under the law in the UK. If I remember correctly, in Northern Ireland is still possible for employers to discriminate positively in favour of Catholics over Protestants in certain circumstances, but that's in order to rectify discriminatory practices over a long period of time. It's very rare indeed under UK law that one can positively discriminate in favour of one group over another. 

00:21:46 Dr Sam Sims 

And what about other minorities? Alex, what do we know about how they fare in the labour market? 

00:21:51 Professor Alex Bryson 

This is very much a moving situation at the moment because first of all sexual orientation is somewhat more difficult to identify in data. Often it's self declared. You might think it's same sex marriage or partnership. It might be based on sexual behaviour. The employer may or may not know about your sexual orientation so it's tricky and it can be fluid, so it's a bit harder to pin down in something that's a bit more obvious in observable terms like sex or ethnicity, but what we do know there's a great paper by a labour economist in the Netherlands, Eric Plug, who's looked at this but shows first of all that there's a lot of what's called workers sorting across employers according to perceived worker tolerance by employers of their sexual orientation. So there's a lot of segmentation in the labour market according to sexual orientation. 

00:22:53 Dr Sam Sims 

So just to check, I understand this now. So this means that people would apply to certain jobs, essentially sort of self-sort into certain jobs. 

00:23:01 Professor Alex Bryson 

Exactly right, and of course this self-sorting is a very important thing to understand when we're looking at any forms of discrimination. The point here is that as an individual worker, the amenity that you derive from a job is to be seen as a as a whole package, it's not just the wage. 

It's not just safety conditions, but it's how you're going to be treated by co-workers and an employer, so that's part of the amenity from a job, and the literature indicates that especially on sexual orientation some workers are who are not heterosexual, are prepared to take a wage hit in return for being at an employer is perceived to be more tolerant of them. They take a compensating wage differential essentially in order to avoid discrimination. 

That's quite important. The other thing we know from the literature again, going back to the correspondence studies, but it's certainly the case that callback rates tend to be lower for people who are not heterosexual. 

But in a very recent study, very interesting cultural differences across areas. So, for example, there's a very fine paper for Germany, which shows this very substantial discrimination in hiring rates against homosexuals in Munich, which is in the south of Germany and a Catholic area, but not in Berlin. 

So one needs to be careful about generalising about how employers are going to behave. Some of it's about particular cultures and locales, and things can change over time. 

00:24:57 Dr Sam Sims 

Yeah, and particularly insidious, of course when it ends up in people sort of sorting into jobs to avoid, for example, workplace bullying and essentially having to take, you know, a hit in terms of pay or following their career goals in order to kind of avoid that workplace discrimination. 

00:25:15 Professor Alex Bryson 

So the big thing that's happened in the world in most of the advanced economies in the world recently is that women go into the labour market much better qualified than men. You then got to ask yourself an obvious question. 

Why is there still a big gender wage gap? 

And so we've done some work on this in recent times. We've got an Economic and Social Research Council grant looking particularly at this at the Institute of Education. I'm the primary investigator on it, and we show that over time the gap is closing, but it's closing very slowly. 

Whereas we should, perhaps in many instances, seeing women being paid possibly a premium that at least reflects their higher qualifications than men. One of the reasons that we don't see this is that women appear to be prepared to accept a compensating wage differential for more flexible schedules offered by employers in order for them to balance work and life responsibilities in a way that even now in 2021, men appear less inclined to do so. Women essentially have a greater preference for flexible work schedules. 

And this can often come with a lower wage, not necessarily lower starting wage. But through lost promotions, things like that. You see women's earnings rising more slowly over time and then in particular occupations like the law. 

Where there's a high premium on presenteeism because customers want to see you face to face as their lawyer. Women find that they are either precluded from the top jobs, and in the law that will be a partner in a firm, or are receiving a wage penalty that comes as a result of them taking more time off work than the man would do, and the work in this area clearly indicates that the size of that wage gap between men and women will partly be a function of the costliness that the employer perceives to be associated with being absent from work or having a discontinuous employment history. 

00:27:52 Dr Sam Sims 

Right, so here we've got female workers choosing to receive lower wages... 

00:28:01 Professor Alex Bryson 

I wouldn't put it that strongly, Sam. I mean nobody wants a lower wage, do they? 

00:28:06 Dr Sam Sims 

Okay, right, and presumably in a world where flexible working was the norm, it wouldn't be necessary for them to incur any sort of earning penalty in order to access, you know, more flexible work schedules and so on, so this sort of leads to the question of what can be done about this in policy terms? 

00:28:26 Professor Alex Bryson 

Yes, it's not just policy terms actually. Indeed one of the great studies on this by Larry Catz and Claudia Goldin distinguishes between lawyers and pharmacists. Now, pharmacists are fairly highly paid as well. It's an occupation where there's very high pay, but technological changes meant that the face to face engagement between worker and customer is not necessary in quite the same way anymore. It's less costly for the employer to offer a flexible wage schedule and as a consequence of the lowering of the cost over flexible schedules, women see very little pay discrimination in that occupation, so employers, technology can come along and change the landscape in a way that can equalise pay between men and women by virtue of technological changes that change the costliness of employers accommodating those preferences. 

It is probably also a role for government in terms of what they can expect from employers. The minimum standards that they can expect employers to adhere to in terms of the package of opportunities they offer to women. The current policies are problematic in the sense that if we go back to the introduction of the Equal Pay Act in the UK in 1975, that led to a stepchange closure in the gender wage gap. 

And the main reason it happened, it happened actually before the legislation came in because trade unions between 1970 and 1975 bargained in the knowledge - and this is all down to Barbara Castle, who was the minister at the time responsible. They knew that this legislation was coming, and they persuaded employers to move before the legislation came in. 

Trade unions ensured that women closed the gender wage gap by 10 percentage points. It was a massive reduction, however since then, guess what the rate of convergence has been in the gender wage gap between men and women since then? A half a percentage point per year. 

And is currently around 18 to 20 percentage points on average, so you're talking about 40 years away before equalisation occurs. That's actually the study we've just done that's about to be published in the Oxford Review of Economic Policy. And so you would have to ask yourself the question, isn't it reasonable to have another push on policy, and if so, what? What policy levers should we be using? The government has recently introduced one which is only currently being evaluated by economists, but is quite potentially quite valuable. It's the requirement for employers to report publicly what the gap is between the men and the women they employ. 

So it's a reporting requirement. And the reason it's powerful is the reputational damage. Being seen to be an employer who pays women less than men. Imagine a scenario because you have to report it year on year. 

Imagine the scenario if you slip down that league as an employer and then it's reported in the newspapers. The reputational damage to you, I would argue is equivalent to the sort of damage that an employer might suffer if they are discovered to be engaged in some other deplorable employment practice. 

00:32:32 Dr Sam Sims 

So it seems like one way this has made it into the press recently is with respect to the BBC, where there seems to have been a lot of press coverage about the different presenters being paid very different amounts and that ending up with, for example, women being paid a lot less than men for what appeared to be very similar jobs. And do we have any evidence from other countries? 

00:32:54 Professor Alex Bryson 

Yes, we do. In fact, about a year ago, together with a colleague of mine, John Forth, ran a conference in Germany where there are a number of papers presented on this very issue. Most of them from Europe and they confirmed that there does seem to be an impact of these policies on the way the employers are responding so very much like the BBC scenario. In fact, what tends to happen is in fact, in that BBC case, of course, the BBC was taken to court by a very well known journalist and so they dragged their feet somewhat in that case, but essentially what normally happens is employers will undertake internal audits, especially if they discover and sometimes they don't know that there's a big gap between the average pay of the men and the women. And sometimes it's quite inexplicable at one level. So for example, there might be big occupational differences within a particular employer. 

Although one might be concerned if those occupational differences reflect differential promotion possibilities in that workplace or in that firm. But sometimes there can be reasonably innocuous reasons for the gap because they are being asked to report a raw gap on the average. 

But oftentimes there are no good reasons, although there appears to be cause for concern. And some of that concern is often about the transparency of the pay structure. And so that can often lead to job evaluation schemes. And this is where the likes of trade unions come in, because they often push for these things from employers to ensure, and it's actually in the best interests of employees, because it's probably the best protection against legal cases is to have undertaken a thorough going job evaluation assessment of the actual value of the jobs undertaken in that workplace. That should lead you to pay the right money for the job. 

But if employers haven't done that, then they can potentially be open to claims that the differences might potentially at least in part, reflect discriminatory practices. 

00:35:20 Dr Sam Sims 

Right, so we've got the disclosure providing the information which can then be used by either the trade unions or by you know, through the courts in order to kind of hold employers to account and kind of incentivise them to get their house in order early. So all these different parts of the policy landscape interact in order to bring about potentially these positive effects. 

And Alex, I know you've done lots of research on discrimination in professional sports as well. What have you learned? 

00:35:54 Professor Alex Bryson 

Yes, well the great thing about sports is that we do know how productive the workers are on the field, whether they're playing baseball, basketball, football, what Americans call soccer. We know what each worker does each day they play. 

So we know how their productivity varies across time and how it varies across the workers in the team. 

00:36:24 Dr Sam Sims 

So I guess we sort of, you know, runs per innings for cricket there. 

00:36:28 Professor Alex Bryson 

Absolutely. I mean, let's take English football. You know, we've got everything, Opta statistics every single game. How far people run, how many sprints they do, shots on goal, tackles made, fouls, so on. And this has spawned a whole literature on the productivity of workers and the extent to which, as economists would assume, and as I've said before, in the absence of discrimination, but wages should reflect that productivity. 

And probably the best paper in the literature, a paper in 2002 by Goff, McCormick and Tollison in the American Economic Review looks at racial integration in professional sports, specifically baseball and basketball in the United States, and this is how it goes. 

So in a world in which as occurred in the 40s and 50s in the United States, Black males were heavily discriminated against in terms of their ability to enter professional basketball, and professional baseball. Any Black player that turns up in a team will have to be so much more productive than the white equivalent worker in order to overcome that discrimination, I was referring to earlier, yeah? 

So in the early period of discrimination, you don't see many Black players playing the games. Those that you do see, boy do they have to be good. 

Now, what does this mean? Well, if we think there's going to be a period of racial integration then those productivity differentials based on race will die out over time because in the absence of discrimination, the average productivity of black and white players should be similar. 

Does that happen? Well, it turns out that it does happen. The racial integration, as indicated by the average productivity, or actually the marginal productivity of the last player that comes into your team, whether they're Black or white doesn't matter from about the mid 1980s. 

What's particularly beautiful about that paper is they go further and say who the first firms to move away from racial discrimination? Were they (a) the worst teams who had no choice but to overcome their taste-based discrimination and employ these particularly good Black players or (b) the good employers who just do everything good and they saw an opportunity to steal a comparative advantage by waving goodbye to taste-based discrimination and making sure that they employed the best players and the answer is it was the second. 

The best teams moved first in starting to racially integrate the players, so that's the sort of thing that you can do to look at things like racial discrimination, (a) when you're really, really clever, and (b) you've got the theory and the data to do it. 

Now, it's not game over in terms of racial discrimination in sports, nowhere near. In fact you only need to talk to sports players to hear that, and certainly that there are all sorts of studies that come up with different answers as to whether or not there's racial discrimination both in hiring and payments in a range of sports. And it's not just about employer discrimination as well. 

There are papers on how referees treat you, how umpires treat you, and so on. So it's it's a very live debate. But what's really really valuable about it is that we do know something about whether this is discrimination, because we do observe the productivity of workers. The downside, of course, is to what extent can we truly extrapolate from this unusual segment of the labour market, often sprinkled with superstars who are paid far more than you and I could ever imagine. How can we extrapolate from them to the economy in the labour market as a whole. So you know, one has to bear that in mind. But crucially, you can use these data to shed light on important principles about the way the labour market operates and whether or not you see discrimination in that labour market. 

00:41:23 Dr Sam Sims 

Alex Bryson, it's been absolutely fascinating talking to you about your research. Thanks for coming on the podcast. 

00:41:29 Professor Alex Bryson 

Thank you very much, Sam. 

00:41:31 Dr Sam Sims 

You can find out more about Alex's research by following @alexanderbryson on Twitter and you can find links to his research in the show notes. 

If you have questions or topics you would like us to address in future interviews, follow the links in the show notes where you can also record your questions using voice or text. 

If you enjoyed listening to the podcast today, there are now seven series of conversations with IOE researchers, all available from wherever you get your podcasts. And as an added bonus, you can find the Research for the Real World playlist featuring tracks contributed by previous guests and our producers. Follow the links in the show notes to Spotify. I'm Sam Sims and this has been Research for the Real World, goodbye. 

00:42:20 Female voiceover 

Thanks so much for downloading and listening to this IOE podcast from the UCL Institute of Education, University College London.