What's happening to social mobility, and why do we care?
What's happening to social mobility, and why do we care?
There is enduring policy interest in what is happening to social mobility. I summarise recent headline findings that policy-makers have referred to regarding trends and cross-national differences. I ask how we should interpret these findings, referring to measurement "basics" (concepts and measures, and data issues), and suggest topics about which we need to know more.
well good evening everybody thank you for being present in this rather technical conference but it is appropriate in these series of uh conference to have come to professor jenkins because he's a scholar who really makes a leap forward in research stephen jenkins is a professor of at london school of economics he took a phd in the uk and he now teaches at london school now professor jenkins studies uh mainly relate to a very important and interesting aspect that is social mobility and disequalities that are two interconnected aspects translate into unfair conditions which is morally unacceptable but there is also a factor related to a lack of economic growth and this is a problem which only apparently is marginal because the lack of growth in europe in the last 10 20 years above all in southern europe including our country which is particularly hit by the lack of social mobility well this lack of growth coincides with an increase in disequality something which has a very important implications which add up to the already big problem of lack of growth and professor jenkins in his studies investigated all these aspects in order to hopefully be able to find a solution lack of growth is a number one problem in europe and even in terms of greece there is a lot of talk about the public data well all economists of the eurogroup and the ministers believe that although you may restructure and relieve the debt if you don't find a mechanism in order to kick off growth well that date will start to grow again and this applies also to italy you know that has a strong manufacturing industry which is the one which keeps us um above the water line which guarantees uh some growth for the future there are countries where the manufacturing industry is not so strong as in italy now it's important to remove as far as possible disequalities and we should have a certain level of social mobility now professor jane gaines you prepared a lecture and you will share with us your study and then i see that there are many young people here there will be questions so professor jenkins you have the floor hello it's great to be here thank you very much for inviting me thank you for all coming along it's pretty scary having so many of you out there i've been given two important instructions the first is that i have to speak into the microphone luckily i like gelati but if i do not speak into the microphone start waving your hands the second thing is that i am enjoying to speak slowly particularly for the translators but it's probably a good thing for you as well probably a good thing for me but that aside so again there are two reasons why you might start waving your hands the third reason may be when you hear my arguments but will come to you have your chance at the end to ask questions so my general topic is to say what is happening to social mobility and why do we care and i'm going to talk about this topic in four areas i'm going to build an argument in four steps the first one is about why we actually care and here i'm going to rehearse the enduring policy interest in social mobility and i'm going to remind you that virtually all the people who say that social mobility is a good thing refer to the principle of equality of opportunity less often but quite often they also refer to more social mobility being associated with more economic efficiency so there are good reasons it appears for being interested in social mobility then i want to take you second to headline findings and i'm referring to them as headline findings because they are headlines i haven't found headlines many headlines for italy but i'm go and i'm going to refer mostly to the us and to the uk but the lesson is that there are numbers out there that are being used to support the arguments that are referring to why we care those are the evidence and it's about differences across country it's about trends over time but then i want to thirdly step back and ask how you should assess those findings should we take them as gospel okay maybe i i said headline findings we have a journalist here do we trust or we read in the news that is essentially the questions i'm going to be asking so maybe this is the academic in me coming out what i'm going to concentrate on this talk is not giving you the answers but providing you with a set of questions with some background so section four is going to be bringing those things uh all together after having talked about some of the problems that refer to measurement issues concepts and so on so questions so the first thing area i'm going to talk about first block is why we care and the interest here is very much uh in a range of countries i've i haven't found uh many for italy but then i don't speak italian so and i'm going to focus on uk us and international examples and just convince you or remind you of the things that you already know about the interest in in social mobility that phrase i want to start with the uk it's where i currently live and what i'm going to do here is illustrate that regardless of whether or not you're on the left or the right everybody agrees that social mobility is a good thing this is a snapshot from a document that was produced by the policy unit of tony blair's new labour government by a guy called aldridge and in this document it's about social mobility it states that answers the question why social mobility is important and here are the headings you might expect it's important because equality the opportunity is important and the idea is that if you don't have social mobility you don't have an equality of opportunity moreover we want to make the best use of that everybody's talents and if we don't have social mobility it is thought we don't have that and moreover there's this idea about social mobility enhancing social cohesion and social inclusion so that's the new labor government we step forward and here we have an all parliamentary group so this is the left and the right uh all together in one particular group and you'll note that there are truths about social mobility um and there it is in two big buttons why mobility matters fairness and efficiency in red or orange and red and they mentioned equality of opportunity they mention optimal deployment of resources and in fact they even put four percentage points of gdp growth a big number as being associated with the potential loss associated with not having quality of opportunity we step forward further in history not quite till today this is the coalition government we're moving further to the right but the uh coalition government actually also stood up for equality of opportunity and social mobility as it says here in the at the extract the true test of fairness is the distribution of opportunities that's why improving social mobility is the principal goal of the coalition government's social policy so this is the left and the right nick clegg is no longer in the government he lost they lost badly but the government has not thrown away social mobility yet there is a social mobility in child poverty commission and what it set up this was set up by the the government is to look at measures to improve social policies to social mobility so in the uk this is big stuff it is also very big in the usa interestingly this is not coming from the social policy area this guy alan krueger is the chairman of the council of economic advisers and in 2012 he talked about the relationship between income mobility and income inequality that is the dispersion in incomes at any particular time uh but he was saying that and you might be willing to trade off more inequality if there was more mobility and he associates these things uh he's pointing out that as inequality has increased the evidence suggests that year to year or generation to generation economic mobility has decreased a problem uh and we'll come back to him because he christened something called the great gatsby curve and i'm going to show you a picture of that curve later on we must move further forward and again from what you might think of as the social policy arena this is the american equivalent of the president of the bank of tatalia the chairman of the board of governors of the federal reserve macro economist no less they are bothered about inequality and its relationship to mobility talking about diminished intergenerational mobility and talking about in these circumstances society faces difficult questions of how to best to fairly and justly promote equality of opportunity those words again equality of opportunity social mobility they're joined together it's not just janet it took us barack obama has made speeches about these sorts of things he's talked he in the in this in this talk he links together equality of opportunity outcomes and so on the same sorts of things that alan kruger was talking about and note that he talks about the combined trends of increased inequality and decreasing mobility pose no less a fundamental threat to the american dream so keep this in mind remember america is the land of a quality of opportunity the dream remember that when i show you some pictures later on okay very quickly uh it's not just in the us and the uk you can find european examples the eu has a document that refers to fostering social mobility as a contribution to social cohesion the oecd this is interesting this is a document about growth it is called going for growth but what's chapter five about chapter five is about intergenerational social mobility why because it talks about more mobility is fostering economic growth similarly the world bank is uh interested in this and uh there is a well-known report uh by three authors talking about these issues and i'm a little nervous because one of the authors is sitting in the front row and they the world bank uh has also been extending these ideas to other areas outside uh the oecd latin america the caribbean and so on so this is big stuff interest around the world so to summarize so far we do care about social mobility and the way in which it is normally stated is that we think we think that more mobility means more equality of opportunity and we think that more mobility will foster economic efficiency but that then of course raises the question what's the problem how much social mobility is there in fact okay before answering that question giving you some headline findings a little diversion because i have to explain how people measure it and i think what you may be intrigued by is that everybody is doing it in the same way but you may be rather interested or or surprised in the one particular measure that they have focused on so first of all you have to choose what you're thinking about mobility of now i'm going to be talking about social mobility but what that actually means typically is income mobility well actually it's often not income mobility but earnings mobility that is among the mobility of people their labor earnings among people who have a job and so it's not the wide this is not taking into account social security benefits it's not pooling husbands and wives incomes and so on it's talking about men employed men women are out of the picture most of the time thirdly it's about it's typically income is measured not using lifetime income because we don't measure lifetime income typically the problem a really serious problem is that we have snapshots from people's lives rather than movies of their income so we have to make inferences from snapshots and economists have spent a lot of time trying to work out uh how to do this and particular problems perhaps spent too much time on it there are two particular problems one is about to do with measurement errors that come from having only short observations within within people's life cycles so-called transitory variation and another issue is life cycle bias this is the idea that essentially if you want to compare the life chances of parents and children you want to measure them at comparable stages of the life cycle suppose you have a father at the age of around about the age of 40 it's no good measuring uh the sun's life chances at the age of 15 or indeed 20 because things happen thereafter there is systematic relationship so there are there is these issues so that's in fact it's not income mobility or social mobility more generally it's earnings it's men and it's this particular measure annual earnings but then how is mobility summarized let me tell you about something called beta okay i'm going to try and summarize some econometrics in a picture that may be a vain hope but i'm going to try so on the right hand side ladies and gentlemen we have a picture and it's summarizing on the horizontal axis father's income so this is the parent's generation and in fact because we're interested in proportional differences it's measured using a transformation called the logarithm but essentially the the further you go along to the right the higher father's income is and where the vertical line intersects uh is going to tell tell us about mean income also in the vertical direction we have sun's income again measured relative to the mean so what that means is that we have we have example data i can't don't know if you can see very well but there are lots of grey dots there and that is a combination if we go out and have a survey of observations on father's incomes uh father's earnings and sons earnings and you can see there are splodges all over the place so somehow we've got to summarize all these different observations so how do we do it so economists like regression linear regression i don't want to try and explain that but the idea is relatively simple and it is to think about what is the expected differences for people who have got fathers of of different with different circumstances just imagine that if every sun's earnings was the same as their father's earnings and expectations that would mean that the everything relative to the mean that every body would lie on this 45 degree line okay so let's put another way if you were to try and fit a line through all those dots the slope of that line would be equal to one so that's one extreme where if a father's income is twice the mean and that's where i've got a point here then the sums income would be expected to be twice the mean another extreme would be where there would be absolutely no relationship between father's son's income and father's income there that in that particular case we take we take this line down and we'll be lying on this horizontal axis that is the other extreme the slope of the line is zero beta is zero now in between when we fit the line to the reality out there we have lines of different slope so beta is the slope of this line and the issue is does it lie between here and does it lie between there sounds fairly complicated it's really nice economists love it it's one number summarizes everything and it has these nice two extremes so another way of thinking about it is in terms of transition matrices what does that mean uh sounds horrible it's very simple idea we line everybody up in the father's generation from poorest to richest and split that population into equal size groups let's say five groups so called quintile groups then we do the sun it's the same for the sun's generation line them up from poorest to richest split them into five groups and then we can cross tabulate origins against destinations and see the extent to which people's origins tally with their destinations to what extent you know if the chance what are the chances of people who start out at the bottom for example ending out in different places well that sounds rather abstract but let's look at italy here's an example i did find for italy about these sorts of things so oops sorry so here are five groups this is from the poorest fifth true to the richest fifth of suns in italy at this particular time here is the same for the fathers so this is the origin generation this is the destination and you might ask then what are the chances of people starting out at the bottom reaching different different destinations from the poorest fifth to the richest fifth and the issue is uh if everybody was stuck in the same place that is totally deterministic position this is like the beta equals one position then this proportion here the fraction of those who start out in this group ending up where do they end up if that was equal to one that means they're all stuck there indeed you can imagine if everybody was all these entries these row proportions were equal to one then that's the no mobility reference point on the other hand if where you ended up was independent of your origins all these entries here would be 20 0.20 the reality is somewhere in between so what we see for example for italy is that of those starting in the poorest fifth you can see around almost a third of them stay in the in in the poorest fifth and only just over ten percent make it up to the richest fifth if we go to the other extreme you can see around of those who started off in the poor in the richest fifth ten percent stay there but uh around forty percent stay in the richest fists now that number is the authors from the paper i took it from remark on this as being uh this high persistence in the upper tail of the distribution is distinctive about italy and we'll come back to that because i've got more information about italy okay so now you know how economists measure mobility in fact they don't often use those transition matrices they're often more sociologists who use those but there are some about beta is it so you can do econometrics now so watch these numbers okay so this is the great gatsby curve if you know the book by f squat f scott fitzgerald you'll know why it's called the great gatsby curve on the horizontal axis we measure income inequality the more we move in the right hand direction the more that inequality in a given the higher is inequality in a given country the thing i've just been talking about beta is on the vertical direction and this is the the problem that the uh alan kruger uh janet yellen barack obama were talking about this curve which shows a positive relationship between inequality on the one hand and immobility of earnings on the vertical axis and you might notice here that uh if we think in terms of the differences in beta we have some of the usual suspects down here that is the nordic countries are countries with low with low immobility high mobility this is in terms of beta then we move up to countries like the one where i was born but look what's up at the top and i've put a box around italy so italy is a low mobility country according to this measure and by the way up there with italy is the united states remember the land of opportunity okay so this is a historically been a very surprising result um people this is why a lot of people are interested in it and the uk where i live is also up there as well so keep this in mind so the big headline the things that people talk about are based on beta and particular measures what about trends over time has it always been like this well we don't have a lot of information why because it's actually quite hard to get this information remember we need information about parents and about children but we have a little bit of information i'm not going to take you through these caterpillar plots but they're essentially estimates from different researchers about the trend over time horizontal axis measuring the the when suns are born along the horizontal axis and these are measures of beta and the thing that you take away from those pictures is that the lines in the middle of each of these pictures is flat so social mobility is not increasing or decreasing which may be surprising but that's how it is the best estimates for the usa that's men if you look at women apart from this old group right right back here you can see again the line is flat the line is flat those are headline news if you go to the uk the findings about social mobility are big headlines this is old research done almost 10 years ago but it's still the headline and i'll come back to why that is later okay these guys wrote their own headlines in one of their papers but you can read that right it's big big writing social mobility is low and it's falling and just to illustrate this i've taken some numbers because i understand that you know about beta now if we're focusing on suns you can see for example the estimate of beta one of their estimates has gone from point two up to point three over time so more immobility beta has gone up and other econometric estimates different ways of playing with the data show the same result so we do care about social mobility one two social mobility as commonly measured is not pervasive we are not at the extremes remember beta equals zero definitely not there in fact there's relatively no low mobility in some countries in particular the usa the uk and italy and mobility may be falling which is definitely not what we want so the next thing we should ask ourselves is how do we assess these headline findings and i'm going out on a limb here um so being a bit provocative against all this literature so basically uh i how should we assess the findings i think we should salute because clearly a lot of work has gone into it but at the same time we need to be a little wary and show some caution and i want to illustrate that argument by taking us back to thinking about how we measure mobility and how do we measure mobility that means going back to the concepts what we actually mean by equality of opportunity and how we measure equality of opportunity and also about data data sounds boring but data are important those headlines have come from data so first of all let's think about mobility concepts and the summary mobility measures so in fact there are multiple mobility concepts and i would distinguish between at least three so called positional or relative mobility first secondly absolute mobility and a third concept which is associated with dynastic inequality reduction but i'm not going to talk about that i've mentioned it but most focus is the difference between positional and absolute mobility so what is positional mobility if we think about it or relative mobility or as sociologists have referred to it often exchange mobility this is a very pure kind of mobility where we're only interested in changes in people's ranks in society it's not the income associated with that with that position but the rank and the nice thing about pure positional mobility is that it controls for the fact that there have been huge changes over time in the income distribution between parents generations and sons generations typically we're talking about 30 to 40 years and we know that the income distribution in many countries around the world for example the average has gone up but so too has inequality in many of these countries there's an issue about how you take account of this the great thing about relative mobility positional mobility is that it's pure mobility and it's also what many politicians sign up to as the sort of mobility we should be bothered about one uncomfortable thing however for politicians is that we're if we're and remember politicians like mobility the problem is that everybody's uh position depends on everybody else's so if somebody moves down from the top one slot somebody else has to go up to take their place but in another way not everybody can be upwardly mobile and i bet you cannot find a politician who would argue in favor of downward mobility for anybody okay so they subscribe to mobility of this kind but awkwardly forget some of these problems sometimes it's also worth while talking about this because the reference point that comes out from this way of looking at mobility this particular concept is that you have complete mobility when the destination positions are independent of parental origins but this is what's linked to equality of opportunity normally okay now let's contrast relative mobility with absolute mobility so here we're actually concerned with individual income growth has my real income level got higher than my father's or in fact is it lower or is my chance of becoming a millionaire now higher or higher than for other people so here is a situation where mobility is not necessarily defined relative to everybody else's it's possibly for everybody to be upward low mobile and of course the fact that average incomes have increased over time becomes really important it's part of the measure well okay so that may be uh just academic um playing around but the important thing to keep in mind is that uh which i summarize the in detail in this slide is if you believe in positional mobility then you should not believe in beta but remember beta is what everybody uses the problem is that beta mixes these two concepts up okay and there are ways of trying to control for it but they don't do a great job and uh so in fact if you're really interested in position of mobility i would argue that you shouldn't use measures not beta other ones that take account of or go to the concept directly for rank-based measures transition matrices and so on but we haven't done that a lot why does everybody do it some reasons are good some reasons not so good uh perhaps the bad one is that everybody else has done it before and so academics are fairly conventional they want to compare their results with other people or maybe that's respectable but anyway there are various other technical reasons some addre to do with addressing some of the problems that i referred to earlier that are that are important but the point is everybody still focuses on beta at least amongst economists is this an important point or not i would argue it is an important point and i would like to illustrate it by contrasting some new results that have recently come up out which compare estimates of mobility based on beta with ones that are based on positional mobility these are just three countries canada sweden and the usa and perhaps the interesting contrast is with sweden and the usa because you remember that sweden is one of the good guys in terms of beta that is relatively high social mobility usa was the bad guy with low mobility that's a new beta estimate and you can see that beta immobility is much higher in the usa than in sweden in fact this recent research has shown that if you move to a pure rank-based measure the rank correlations then it turns out that sweden in the usa have got the same amount of relative mobility so is there a great gatsby curve when you look at it this way maybe not okay so there are different mobility principles the choice of principles really matters here's some further evidence from the usa i won't talk through the details but just look at the picture and believe me on the vertical axis we have a measure of mobility not beta but a rank uh based measure uh equivalent and on the horizontal uh axis is time and basically what you can see on that picture is that that line is flat so interestingly in the usa when these guys have um raj chatty and colleagues have done these estimates the estimates of rank based mobility are constant as well um so we have these questions about the principle of how we think about mobility another way of thinking about mobility is that we were interested in mobility because we think beta measures equality of opportunity and the transition matrix this idea when beta is zero or when the entries in the transition matrix are all equal to 0.20 20 percent then this notion of origin independence however john romer an american economist has sort of pointed out that the emperor hasn't got many clothes on that for example the degree of origin independence is a direct and good measure of inequality of opportunity only if two rather special conditions apply the first condition is essentially that all parental advantage is summarized by income or in these these studies earnings the second one is that we're basically somewhere we're not allowing for the fact that outcomes remember this is incomes for children sums in particular we're looking at our reference point is entirely in terms of parental income family background we're not allowing for the fact that some of the differences in the race of life depend not just on where we start but also the amount of effort that people put in so really what we should perhaps do if we're interested in measuring equality of opportunity is that we should try and measure it directly so that leads to a whole new area of literature by john roma that has led to lots of work including by uh peter peragini here about equality of opportunity or its inequality of opportunity so this literature makes a distinction between circumstances which are the things that people cannot affect and so cannot be held responsible for in a moral sense uh or social justice sense virtually most things associated with family background i've been talking about income but it might not necessarily only be that it could include things like uh skin color ethnic minority group perhaps where people live and so on but there is also the issue of effort that is uh things that people can affect or choose and so um they may be morally responsible for and one key example would be might be here work effort in the labor market people's earnings may differ because even if they have the same talent some people may work harder than others and of course outcomes reflect basically circumstances efforts and policies and john's argument is that equality of opportunity occurs if and only if the difference is in the outcomes of interests we're talking about children's incomes here reflect differences in effort alone okay so the idea is that circumstances are irrelevant so he's trying to control for extra stuff and uh i i've gotten the other bullet points how people have gone about trying to explain those sorts of things i won't go through the details what i'm going to show you however is a picture just to try and convince you that this approach leads to something that is different from the mainstream and headline approach so this is drawing on uh a world uh world bank report in in including some of veto's work there are two charts here i'm afraid they're rather indistinct but you may well see better on your handouts the one on the vertical axis at the top rather is a measure of inequality of opportunity defined in the particular way that they do in the study they call it iowa and on the horizontal axis is our friend beta now if we were measuring the same things these observations for lots and lots of different countries would all lie on a straight line ladies and gentlemen do those lots all right on a straight line the answer is of course no there is a lot of dispersion around those lines so bottom line we have two different concepts here and so the choice of how you think about your normative concepts actually matters the other thing to take account is the second picture remember the most famous graph in social mobility studies at the moment is the great gatsby curve that i told you about before this is the attempt uh well my i'm calling it an attempt i'm not sure the authors called it that this is um uh measuring income inequality on the horizontal axis this is measuring inequality of opportunity in a roman sense on the vertical axis and again here we have all the country observations there do they all lie close to a straight line in the same way as the great gatsby curve answer no so once we start muddying the water asking these questions about going behind the the headlines it's not so clear so we might say again picking up this theme about which mobility concept we might ask ourselves whether or not we should look at absolute mobility more and of course there are various ways of doing it and let me remind you that none of them is measured directly by beta which is what people use absolute mobility may be particularly important these days as because we know this is a period where average income is not uh continuing to grow as eugenio sidner's introduced introductory remarks economic growth is particularly important and we know we know now we're in an age of the great recession followed by the austerity age we're not currently in the era of what american economists referred to as the great moderation that followed the post-war gross growth spurt so income levels come into play i think very importantly and i think it may chime in with a lot of people's views about what counts as mobility or what they're worried about when parents and children are worried about i'll just give you one example of this and there is or two examples this is an american example the pew trusts a research organization surveys people about income mobility and their attitudes to economic mobility and uh amongst other things they ask them about absolute mobility and it turns out that uh in this particular time what this long quotation is summarized rather shortly it briefly says is that americans now a much larger proportion of americans are worried that their children will not be as well off as they themselves were in an absolute sense we're talking about real incomes and this comes back to some of the issues that you were talking about at the beginning because if you don't have economic growth or in fact if you're at the bottom and we know that people in the bottom in particular are not doing particularly well their incomes are either flat or falling over time then you have this social mobility problem there is also similar problems in the in the uk uk coming up um i've got a reference here to tony blair uh tony blair was famously accosted in a television interview by a very aggressive interviewer called jeremy paxman who was trying to get him to say whether or not it was acceptable for inequality to go up tony the consummate politician of course dodged the question but what he was his answer was to say that what he was interested in is increasing the incomes of everybody he couldn't care if david beckham's income increased a lot uh the issue was also was essentially whether everybody's income increased a little uh or it must so uh absolute mobility is potentially an issue there are also issues about whether or not we should talk about uh earnings versus income i said that that that was uh an issue what surely we should think about wider concept than just earnings so perhaps we should think about that we should think about all individuals not just men also bringing women so you ask yourself are the data not out there i would say there are data out there we should do something but i should also acknowledge that once you start bringing women into the picture it becomes more difficult and perhaps economists to their discredit have shied away from these sorts of difficulties if if somebody is not working in the labor market so an unemployed man not just a woman who is staying at home and looking after the children their earnings are zero and that's a problem for the way in which mobility is measured in these particular ways and people have shied away from addressing that problem i think it's uh that's a potential problem perhaps we should spend some time although it's messy and we should also investigate that problem does it make a difference the answer is yes we have old evidence about this a study i worked on uh when i was doing my doctorate showed that beta measures were there was more mobility in terms of income there is for earnings raj chatty has also recently shown this with his colleagues for the usa there are also issues of data we might you think move to new new types of data and people are doing this increasingly usa that's good advice but right not so good in countries that don't have those sort of data or don't have available sorts of data what we can do is of course study these the uh the the case studies that are already out there and investigate them in more detail and ask whether or not the headline about the reliability of the headline findings so let me take uh a uk case study and go back to that big headline that i talked about before remember social mobility is low and may be falling this is based on just two surveys folks so when you think about trends you might think trends you build up inferences from many data points this is just two points which is potentially problematic you've got to stick clearly and you're joining up these two points but i think you don't want to put too much uh rest too much weight on those conclusions without further checking moreover if you dig a bit further in these studies you'll find that the measures that are used for parental status and sun status are not exactly the same there are questions about the quality of the data and i detail them in the bullet points my point here is not to say that the researchers who did the study are bad researchers i just think we need to be more honest and open about the fact that we don't know everything and or act as if we know everything in this area so are the results robust i think they're potentially sensitive here is one later study that has reinvestigated it using different data slightly different methods and essentially what they have here is a row of estimates of beta they've gone from two to many and essentially this is evidence that's consistent with the earlier study okay maybe i was being too harsh it's by my former colleagues at the university of essex essentially showing this is beta you can see that it's been constant constant this is essentially what blandon and colleagues were showing and then this beta rising mobility this is the same period that blandon and others were talking about so if you take this particular approach you appear to get evidence that is consistent okay they were economists what happens if you wheel on the sociologists okay to be fair remember they're talking about something different they're talking about social class mobility but you might expect that they're because social class is measured in terms of occupational hierarchies and we know that earnings are closely related to occupational position you might expect that there would be some form of consistency in terms of the results and john goldthorpe and others has essentially argued and used lots of data that has got multiple data points is basically argued that there is not evidence for the blandon and others claim that mobility is falling in the uk they're not saying it's rising they're simply saying that the the case that it's falling has not yet been made convincingly and they try and explore the reasons for why their results differ and the there are detailed technical arguments about the way in which earnings differ within occupations and the extent to which that dispersion of earnings has changed over time that is between generations and they construct a story to explain why their results differ from blandon and others and implicitly argue their results are more reliable of course the original researchers do not take that lying down they fight back and in a recent article in the journal of the royal statistical society they dismiss the sociologists argument i can't say who's a winner yet because they're still fighting um but what i want to leave you with is this issue that we need to be uh cognizant of the potential uh result that you know results are not always clear-cut they're not always robust and i have a second italian example here it's rather old data actually produced by two two sociologists and uh about mobility in italy and this is essentially a measure of association um between uh occupational position of fathers and sons it's a bit like uh kappa their measure is a bit like beta and it's going down down down but very slightly and these measures of our statistical confidence in the results so essentially the point estimates suggest that there has been a slight increase in social or occupational class mobility for men but we'd have no evidence uh comparable evidence for income or earnings or occupation so again all the issues that i was raising for britain i think you should probably also be asking in italy as well we may well agree that there are that the social mobility is relatively low in italy but there are questions about its trends um there are also arguments about uh the relationships between germany and the usa but i'll skip that in the interests of time so i'll just i've now got to my fourth uh step so i've been building up my argument by saying that there is a lot of interest in social mobility with appeals made to the principles of equality of opportunity and economic efficiency but i've stirred the mud and i want you you know these are potentially questions that we can answer discuss amongst ourselves what really are the links between mobility and the underlying moral or normative principles that we're really interested in and what is the interest in particular the relationship in particular for example between social mobility equality of opportunity and the more recent measures of equality of opportunity that have been argued for by john romer my second point was that there are lots of headline findings out there about uh mobility about differences across countries and about mobility levels and they're often taken as gospel certainly in britain and the usa uh in in at least in the public arena by politicians but they deserve scrutiny and i've given various reasons they're to do with how we summarize mobility what is the appropriate mobility concept who should be included in the analysis indeed how should we summarize life chances in terms of income earnings social class or what what are the relationships between these measures and of course if we want to derive robust conclusions we need good data sets uh and uh that that's important as well so overall i would say that we've made a lot of progress in the study of social mobility over the last few decades we know a lot more about some countries than others and you can see that in my focus on the us and the uk i found a couple of italian examples but there are you know there are relatively few and not many about other countries so there's still much to learn as uh danish uh philosopher peter tynes said problems worthy of attack prove their worth by hitting back thank you for discussing thank you so much for your presentation and thank you so much for working on these issues central issues indeed and the work of professors like jenkins actually consoles us because they really explore in-depth such dramatic and urgent issues the fact that in italy there is an issue of social mobility is evident to everybody it is the land of relations also in america the fact that the uh american dream is a failure well this is evident we saw the images on tv of ferguson and other riots well a bit more surprising is the uk and this perhaps comes from the old issue of lords aristocracy and the differentiation into different backgrounds so it's really current uh problems and we don't have countries such as russia and qatar in these uh diagrams where probably there are issues of social inequality now what i'd like to ask uh jenkins before uh turning to the audience is the following question well we have seen in your speech that many scholars many economists work on the problem and even very popular people or jenkins quoted obama blair the chairperson of the federal reserve who as a woman certainly experienced some forms of discrimination so you see people who have the power who should have the power to solve these problems perhaps there is a fear to limit privileges or touch privileges why do you think people also powerful powerful people cannot face these problems and solve these problems and in europe do you think that europe could take some initiative in order to reduce disequalities and increase social mobility uh thank you very much for the the questions uh essentially the first question is about uh politicians uh and why yes what how they operate in relation to social mobility the first thing of course is that no politician would ever say that they are against social mobility okay we all agree that it's a good thing and of course that gets them a long way as i've tried to explain social mobility is a rather vague concept in terms of making it concrete for less scrupulous politicians that of course is very convenient it does they don't have to do very much but that that's perhaps a slur on politicians there for example the uh british new labor government that came in with tony blair was seriously intending to try and do something about social mobility and that means help they redirected a lot of attention to helping families uh families with children in particular and that you know is if if the problem is in the home then help families and they they did a lot but here here comes to the second problem for politicians and uh assessing what's going on is that we're talking about social mobility is something that takes place across generations a generation is a long time again politicians work to short-term timetables and that often uh i mean clearly the benefits of making investments in children and families take a long time to come off there are early signals that we may look at for example the extent to which uh family background difference is in achievement early in primary school or in secondary school or entry to the labor market may provide us with early evidence about what's going to happen but it's not not the evidence the problem is so again this long-term nature of the problem stops people doing it and i guess there is also a third problem and that is priorities and we all agree that social mobility is a noble aim but so too is increasing economic health at a particular time or return to growth and right at the moment after the financial crisis the great recession uh governments are very concerned about fixing growth getting the economy back on track rather than perhaps long-term uh solutions so three reasons what about the possible role of europe because everybody says that europe should be a place where problems are solved and not created and yet it seems that the opposite occurs and that only problems are created in europe so europe perhaps should do more for social mobility what do you think about it do you think it's difficult yes it's definitely difficult but um maybe i'm more optimistic than you are um in the sense or well remember i live in a country that is about to have a vote on whether or not to stay in uh the answer is yes of course we've got to stay in uh but uh the issue remember that uh if you politicians are in in terms of not just uh concrete policies and doing things they're in the business of changing minds and one of the important things that we know about the millennium goals is that it helped concentrate the mind on poverty around the world whether or not it's led to a concrete actions and other story but it's helped contra contrast the mind focus the mind similarly tony blair and gordon brown in new labor stood up near the beginning of their first government and said we're going to abolish child poverty get rid of it totally stupid politicians to make such a big claim but on the other hand it changed the nature of the game this is aspirational politics that really set out the goals now we come back to europe which is what you asked about remember that there are europe 2020 objectives now it's true that social mobility is not of the short short list of objectives for euro 2020. poverty how poverty reduction is however one of them but i'm perhaps more optimistic that uh one thing that could be done at the european level is to change the nature of debates and these nature of the debates are going to become increasingly urgent as we move further into this austerity age this means that we should start talking about what europe can do not only in defensive terms but also in proactive terms and in constructive terms we open the floor to questions now mr jenkins you made a very convincing case about the difficulties in measuring social mobility please speak in italian and you will be translated into english so that the rest of the audience can also understand my question is a very interesting uh overview of the debate about social mobility and your evidence to how difficult it is to measure it in a reliable way my question is whether theoretically speaking social mobility promotes economic efficiency or not so let's assume that we have a better and bigger social mobility would that contribute to economic growth or not yes thank you so we're returning to uh big picture uh questions big so i'll give you a big picture answer and of course the difficulty is making that concrete so how do we actually do it and the interesting thing about uh stephen aldridge's quotation that i gave you right at the very beginning is that the uh cabinet office policy unit that he belonged to had developed these estimates of how it can be done so it you know it is possible to provide estimates of the the gains that could be done from social mobility of course it's and similarly from equality of opportunity it's quite hard to do and again the results are probably uh you take them with a grain of salt but yes we could calculate these things i think the the problem is not at the level of the general arguments i think as i said before i think we possibly all agree with them the difficulty is making them concrete and coming up with uh robust answers first and secondly being very clear about the different sorts of concepts that we're interested in thank you and i would add that uh stephen uh examines the problem stun is the problem then it is up to politicians and to those who work on the field to to to uh implement the outcomes of the theoretical research thanks also to exact uh measures professor jenkins in the past days several speakers underlined the need to place education and social state at the very center of the political agenda and of course this entails a number of costs and expenses do you think that the tax commitments so to say that have been made and that derived from uh being part of the fiscal counter compact and the factor for growth do you think that they are compatible with the overall aim that we said or are they two big constraints uh yes clearly uh if governments sign up to austerity packages either by lowering spending or increasing taxes then that is a big constraint on the other hand there are always choices and so the question is not just about the levels of taxes or spending of different types of spending there is choices about the structure of that spending and so for example in the uk the labor government introduced policies uh such as an educational maintenance allowance which was to promote uh staying on at school uh and that is was removed by the by the coalition government who substituted something in its place that was i think not quite as adequate so but it's a rep there is an example of choice so again i refer to in terms of policies that promote uh equality of opportunity might devote resources to children let's say early broadly speaking early years investments which may be in education but generally preschool families and those sorts of things again that's a choice uh even if the the the amount that you're taking from taxes going on there's still a choice about whether or not you put it into pensions the health service or this so there are choices to be made uh more pessimistically i i'm not sure that those politic at least speaking about britain um those choices are going going to be made we should see the the current government for example has has just announced that or recently announced that um it's going to be uh extending the amount of uh free or subsidized child care and the the uh that on top of what already exists and which you might think of was a quality of opportunity promoting policy but it's unclear yet we don't have the details whether or not that will actually do it and there are already people saying that it's not funded particularly well so again it comes yeah money is money matters but i think choices also matter the it is a question of choices and kotarelli mr cotterelli said that yesterday here in trento he said he was the commissioner for spending review he was called mr katz and he said that it's not just a question of curbing of cutting you have to make choices because some things have to be reinforced and that is education training research university and as professor correctly said education in general uh this is what we have to head to uh you technically i have a more technical and abstract question in terms of the philosophy of knowledge if i understood correctly the message of your contribution is that social mobility is a very important topic we have to be concerned about it we have to care for it and but we also know very little about it we have contrasting approaches to that so much so that you ended your contribution uh opening up a number of questions about the measurement of social mobility itself with a person sitting next to me we were discussing about it and said that conceptual schemes contribute to creating the reality that we want to describe which is constructed based on those conceptual schemes now the core of my question is this conception this concept which is a bit confused of social mobility that we have doesn't it derive from the lack of dialogue between different disciplines say economics and sociology so there are different and separate approaches but you quoted a number of sociologists and results there have been compared but it is difficult still to elaborate a common language that we can use for communication purposes among ourselves and among different disciplines uh yes that's true um but in actually the field of social mobility is probably a lot better than many other fields in terms of its uh dialogue between uh different groups i'm privileged i'm married to a sociologist so it's uh we talk uh but the the uh the the serious point is it i mean i i you you gave a rather pessimistic uh description i think about interchange between disciplines and to some extent i would agree that there are many areas where economists and other disciplines don't talk this is probably one of the better ones actually because there are for example there are in terms of concrete evidence in terms of data there isn't a lot of there aren't a lot out there so everybody's beginning to use the same sorts of information you may be intrigued by the fact that last year in the american economic review published uh some papers by leading sociologists uh on intergenerational social mobility over a couple of centuries so there are there are dialogues i mean it's always going to be the case that people will look at these things differently i i find it rather interesting in fact that uh at least in the british context that there is now a group of sociologists and a group of economists who are having this uh dialogue okay they talk past each other a little but uh they're at least shouting at each other which is a good start and we're making progress and it really is interesting to know about the relationship between occupational social class and hierarchies and how that relates to earnings and uh income i mean i i think now i'm more i'm more positive than you are yeah so the restaurant there are also study groups and set up by the oecd and the world bank for instance and normally they are multi-disciplinary isn't it so steven they include economists psychologists sociologists mathematicians for the statistical part so really the effort goes in the direction of being multi-disciplinary now is there an additional question you convincingly said that a larger social mobility would promote or facilitate growth but it's not clear to me what is the basic intuition at the basis of this uh notion so in from the theoretical viewpoint how would you explain this correlation between social mobility and economic growth thank you okay it's a very simple idea not one that is particularly well filled out but the idea is that everybody is talented in different sorts of ways if we do not use everybody's talents including those from uh lower family backgrounds however measured then we're wasting talent that is the short answer that that's as far as it goes now of course when to make it more concrete and to come back to the situation that we were talking about in answer to the question before then you have to start getting much more precise and to be honest i don't know how they did their calculation but that's the direction of travel so looking you look at the gradients the differences by different social groups in terms of educational achievement and starting from a situation where you believe that say uh to all children of the same intrinsic talent you just need to nurture that talent and we know that there are differences and if you were to bring them up to the same then that will have knock-on effects we know about rates of return to education and so on and so forth that's the way that's the way in which the calculations are done also because as a journalist i can say that if there is more upward mobility well you try to increase also the earnings and so taxes will be higher and then with a better education with better educational attainment people will have better paid jobs which will create a wealth that perhaps they will set up their own businesses so it's a virtuous circle which uh aims at growth well if you think about that i don't know stevens if i'm saying something stupid but that's the reason why we say that when a country gets old it gets impoverished not because all people are bad but because there is a failure of new initiatives for example one way of interpreting what you said is that um absolute if everybody's income levels go up that's a good thing and of course that's absolute mobility the problem is uh whether or not everybody's maintaining the same relative position or not and one of the consequences that has been much described more by sociologists than by economists is that the fact that those who are several rungs up the ladder want to stay there and if education is one of the key routes uh for social social mobility getting on then uh middle and upper class parents are going to do their damnedest to keep their children in good education and maintain those positions so if i were to turn pessimist for a moment then one interpretation of what what you said is that it depends where the money is going and if it's in different parts of the distribution if if better off parents are doing well and better than the people at the bottom and would like to move those resources into you know maintaining their position well it doesn't necessarily have a good effect on social mobility and the chances of those at the bottom it's not necessarily that everybody will gain so again it comes back to these concepts as well perhaps they can recruit poor people unemployed people well anyway i have a question when we look at reality reality is one but then the viewpoints or the patterns that we use are different and you said that we must be cautious because the approaches should consider the income men women well apart from the issue of other parameters the issue of wealth meant in a broader sense i think that well the could have an influence there what can you tell us i can tell you that i left that out by well i should have mentioned at the beginning in shortening my slides the number of slides that i had i removed a slide that's well that's my excuse no wealth after all we've had thomas piketty here uh this week as well who reminds us that wealth is very important i was talking about income you have very correctly reminded us that it's not just income but also as the stock of assets that people have which is very important and moreover it's not only financial wealth but also human capital which is particularly important uh to be honest i was i'm been focusing on one particular literature partly because that's one you know we could we could be here all night and tomorrow as well that'd be fun but um to to narrow the scope of the talk focusing on that but yes uh very good point thank you for reminding us that it's not just income it is also wealth that matters and of course uh other sociology sociologists in the room would remind us that it's not you know there are issues of power and other sorts of things that are associated with income and wealth but or with occupation but it's there there is all wider measures of social position uh than the relatively narrow ones that i've been talking about indeed one of the lessons that i was trying to get off across by reminding you of what is is to remind you of what the headline measures are actually based on we all we all think of these wider concepts but in fact they're based on relatively narrow ones so your point is a very good one thank you for it thank you so much for this very clear lecture on how to read and i hope it's not offensive to make a question about the causes because today there is a lot of literature on data but then we economists are not ready to give answers so it's fashionable at the time to say that everything is caused by institution but then it's not clear what institutions are now the data that you showed demonstrate that the us and sweden on the one hand are the same results but then uk and italy have the same results but these are countries uh well where instinctively there are very different institutions if we think of the welfare state for example so i wonder if in your opinion whether the welfare state matters and what is the theory which can explain us why the welfare state is so important in building up a social mobility uh how long have i got so big question very big question but you warned it i mean yes there are clearly differences uh across these countries and it's easier to talk about country the nordic countries compared to other countries like italy or the uk or the us because they're very different welfare states and one of the things well let's take an example we've been talking about the promotion of uh family policies as one way of improving social mobility equality of opportunity and so on and maybe it is that the if depending on how you define the welfare state if you think about the nordic welfare states that it is uh rather rather broader than it is in many other countries and i think that that sounds like a rather banal statement that i think it begins to get to the heart of of of what what we're at and uh and so the moves uh that i reported for the blair government and the later brown governments about putting resources towards families helping families generally in in into the labor market the so-called early years policies are a move that might be interpreted in the nordic direction and help you know will have these longer term payoffs and at the other at the the other end the the villain of the piece countries like the usa you know simply don't have these policies in the in the same sort of uh way so the nature of at that level where you know when you're talking about welfare states in a very general sort of way i think i think it's you know there is a fairly easy answer to it the difficulty of course is as i've been saying in my talk the devil is in the details um and and they become important so uh as i was emphasizing before um choices matter and related what that also means in the policy arena is that policies matter and there are different policy regimes that countries can choose or have chosen and they will lead to different outcomes but of course it you know there are other sorts of things going on it's the the nature of the labor markets uh in particular the nature of the education system but of course that brings us back to governments again and they're things that governments and imprint who are influenced by the uh electorate uh can make uh ultimately make choices about and change directions it's a bit like an oil tanker it takes 10 kilometers to change the direction but you know things can happen in the longer term and i think um you know well let's hope it does towards us anyway thank you and goodbye thank you very much everybody foreign oh um you