CentrePiece Winter 2008/9

In the third of CEP’s ‘big ideas’ series, Jo Blanden traces the evolution of CEP research on social mobility and its interaction with policy debate. Big ide Intergenerational m

n 23 June 2008, the issue of social mobility and the Liberal previous issue of CentrePiece showed, Prime Minister gave a Democrats supporting their own CEP played an important role in describing flagship speech to school independent Social Mobility Commission. the evolution of cross-sectional wage leaders in which he said The rise of social mobility up the policy and income inequality during the 1980s that ‘raising social agenda has coincided with a series of high and 1990s. mobilityO in our country is a national profile studies from CEP researchers. As At the same time, related projects laid crusade in which everyone can join and Stephen Machin’s ‘big ideas’ article in the the foundations for an enduring research play their part’. In January 2009, his strand on intergenerational mobility. In government published a White Paper 1997, Lorraine Dearden, Stephen Machin on social mobility. and Howard Reed followed up influential The opposition parties share the desire work in the United States by estimating for more mobility, with Conservative the extent to which sons’ and daughters’ leader David Cameron pledging in earnings at age 33 are associated with December 2006 to take ‘the banner of their father’s earnings for a cohort born sensible, centre-right reform’ to the The rise of social in 1958. mobility up the policy agenda has coincided with a series of high profile studies from CEP researchers

6 CentrePiece Winter 2008/9

equals downward mobility; if some used the Family Expenditure Survey to children move up, others must go down. conduct an extensive analysis of the time- One of the problems with the patterns of child poverty in the UK. transition matrix is that it is unable to take Unsurprisingly, given our knowledge of account of the extent of movements what happened to the wage distribution within groups. If those moving from the over this period, the child poverty rate had first to second quartile are just tipping risen sharply in the 1980s. over the boundary between the two The second part of the study showed groups, there is less mobility than if they how strongly family background are moving into the middle or top of their influenced children’s development and new group. later outcomes. This reinforced the To overcome this limitation, message from the intergenerational economists also adopt a regression mobility analysis that experiencing low approach, which takes account of all the income in childhood could have a mobility between generations. This profound impact on later achievement. produces the ‘intergenerational elasticity’: The Gregg et al (1999) analysis had a a result of 0.3 would say that on average powerful influence on future policy- a 10% difference in income between two makers. At the time, the New Labour as sets of parents would be passed on as a government was finding its new policy 3% difference in income between their agenda, rejecting the ‘Old Labour’ values children. This statistical approach is also of equality of outcome in favour of a new based on an entirely relative conception of focus on equality of opportunity. mobility; the amount of upward and The message from CEP research at the obility downward mobility balance. turn of the century was that The research by Dearden et al (1997) intergenerational persistence in the UK presented a picture of limited mobility in was substantial and that high rates of the UK, with results similar to those for child poverty painted a bleak picture for The approach taken by economists the United States, although the study did the future of British children. It seemed to measuring intergenerational income or not make an explicit comparison. At natural to put these two facts together to earnings mobility is relative. The most around the same time, CEP researchers ask a new research question: how had the straightforward description of mobility Paul Gregg, Susan Harkness and Stephen influence of parental background changed uses a ‘transition matrix’, which divides Machin undertook a two-stranded project as the rates of child poverty increased? the income distribution of the parents’ for the Joseph Rowntree Foundation. With the release of reliable earnings generation into equal-sized groups The first component of their study data from the 1970 British birth cohort in (usually fifths or quarters) and shows the 2000, a comparison of intergenerational proportion of the next generation that mobility over time in the UK became moves into a higher income group, the possible, comparing the 1970 cohort with proportion that goes down and the their counterparts born in 1958. Research proportion that stays the same. on these two cohorts measured the Movement away from the starting association between the income of point is seen as mobility. Notice that in parents (when their children were aged using this approach, upward mobility CEP’s 1999 analysis 16) and the earnings of children in their had a powerful influence at a time when the government was rejecting ‘Old Labour’ values of equality of outcome in favour of a new focus on equality of opportunity

7 CentrePiece Winter 2008/9

early thirties. The association was found to and children’s earnings developing out of ‘are individuals in better class occupations be stronger for the later cohort growing a two-stage process. than their parents?’ Second, relative up in the 1980s than for the first cohort First, parental income relates to mobility is about the extent to which there who grew up in the 1970s. children’s characteristics, that is, those are movements between classes that are The message from these data was with better off parents have more not driven by the overall changes in the that the rise in inequality and child poverty education; and second, these class structure. had coincided with a fall in social mobility. characteristics are rewarded in the labour The sociologist John Goldthorpe (of This study was published in 2004 in a market, that is, those with a better Nuffield College, Oxford) and co-authors book edited by Canadian economist education earn more. Using the rich data have been tracing the progress of social Miles Corak. in the British cohorts, the study found that class mobility in the UK and the rest of The finding that mobility had declined the great majority of the increase in Europe for several decades, in general was well-timed, with initial versions of the intergenerational persistence could be emphasising similarity across nations and findings (presented to the Royal Economic accounted for the strengthening of the stability within them. Investigations by Society at Warwick in 2002) attracting relationship between parental income and Goldthorpe and Jackson (2007) and widespread media attention. This interest children's performance throughout the Erikson and Goldthorpe (2008) of changes peaked in 2005 when the findings were education system. in mobility using the 1958 and 1970 presented as a summary report for the The results discussed above are all cohorts demonstrate that there has been Sutton Trust. based on relative income mobility; this is no change in relative mobility when social The tendency of the media at the time certainly not the only measure of ‘social class is used as the outcome measure. was to sum up the research with the mobility’. There is a long history in the UK The difference in results between the headline ‘Social mobility in the UK is of measuring social mobility by observing sociological and economic approaches has falling’. While convenient for newspaper changes in social class within dynasties. led to a lively and productive debate, with editors, this was actually misleading, Social class has tended to be measured by the sociologists asserting that the implying that the fall observed over the fairly large groupings of occupations (say differences are due to weaknesses in the 12-year period in question continued over seven); as with the transition matrix measurement of family income in the the following 20 years. Researchers will approach this may obscure substantial cohort studies. Using a number of not be able to evaluate this properly for amounts of mobility within classes. approaches, Blanden et al (2008) another couple of decades, although a Another issue is that there is clear demonstrate that measurement error is recent follow-up study by Blanden and change in the social class structure over not the issue. Instead, they explain the Machin (2008) suggests that the degree of time due to old occupations dying out and results in terms of the large inequalities in mobility is unlikely to change between the new ones emerging. This means that family income that are found within the cohorts born in 1970 and 2000. There is social class analysis can explore two broad social class groupings used by certainly no evidence at this stage of the dimensions of social mobility. First, sociologists; in light of these, there is no situation continuing to deteriorate. absolute mobility considers the question reason to suppose that the two Hot on the heels of the finding that methodologies should find similar results. intergenerational mobility had declined The dialogue between sociologists came the search for insights into why this and economists has certainly helped to had happened. Work by Jo Blanden, sharpen the policy debate on mobility. Lindsey Macmillan and Paul Gregg sought Many political speeches have made to discover more about this, using a reference to ‘social mobility’ without a framework that considered the clear conception whether they are relationship between parental income The central referring to absolute or relative mobility, message from data on the 1958 and 1970 British birth cohorts was that the rise in inequality and child poverty had coincided with a fall in social mobility

8 CentrePiece Winter 2008/9

and mobility measured by income or social But this is not the only contribution made Further reading class. With contributions from both by CEP researchers to our understanding sociologists and economists, the recent of social mobility. Jo Blanden (2008) ‘How Much Can We Learn Cabinet Office discussion paper ‘Getting Indeed, CEP contributions predate this, from International Comparisons of Social On, Getting Ahead’ has helped to clarify with our researchers being among the first Mobility?’, invited paper written for the the meaning of social mobility and has to document the strong association Sutton Trust summit on social mobility. therefore set the scene for more between family background and later transparent policy-making in the White achievements in the context of the UK’s Jo Blanden, Alissa Goodman, Paul Gregg and Paper on social mobility. high child poverty rates. CEP contributions Stephen Machin (2004) ‘Changes in In June 2008, a number of CEP have also moved the debate past the fall Intergenerational Mobility in Britain’, in Miles researchers (both past and present) in mobility, to investigate ‘what happened Corak (ed.) Generational Income Mobility in attended the Sutton Trust and Carnegie next’, and to consider the relationship North America and Europe, Cambridge Foundation’s trans-Atlantic summit between different measures of mobility University Press. on social mobility in New York, an event and what these might mean for policy. star-studded with top academics and It is also clear that the research on Jo Blanden, Paul Gregg and Lindsey policy-makers. intergenerational mobility discussed here Macmillan (2007) ‘Accounting for Jo Blanden gave the opening relates closely to the work on cross- Intergenerational Income Persistence: contribution, which compared the levels of sectional inequality reviewed in the Non-cognitive Skills, Ability and Education’, mobility across countries using a variety of previous issue of CentrePiece. Our interest Economic Journal 117: C43-60. methodologies, emphasising what can be in intergenerational mobility is in part learned from taking a multidisciplinary encouraged by the recognition of the Jo Blanden, Paul Gregg and Lindsey approach, and demonstrating the UK’s high inequality levels and exceptional Macmillan (2008) ‘Intergenerational correlation between low levels of mobility child poverty rates. Persistence in Income and Social Class: The and high levels of income inequality. More recent work has attempted to Impact of Within-class Inequality’, mimeo. Sandra McNally discussed the contribution understand more about the link between that schools policy could make to inequality and mobility, both in terms of Jo Blanden and Stephen Machin (2008) ‘Up promoting mobility, and Stephen Machin how inequality may influence different and Down the Generational Income Ladder in appeared on the policy roundtable measures of social mobility, and more Britain: Past Changes and Future Prospects’, alongside cabinet minister Ed Miliband. profoundly whether greater inequality in a National Institute Economic Review 205: The New York summit was followed nation leads directly to less social mobility. 101-17. by a one-day conference at CEP, which presented some of the cutting-edge work Lorraine Dearden, Stephen Machin and on intergenerational mobility currently Howard Reed (1997) ‘Intergenerational being carried out in Europe. Again, Mobility in Britain’, Economic Journal 107: there was a strong policy focus to 47-64. proceedings. More information about the discussion can be found in the previous Robert Erikson and John Goldthorpe (2008) issue of CentrePiece. ‘Income and Class Mobility between In conclusion, there has been a Generations in Great Britain: The Problem of justified focus on the finding that Divergent Findings from the Data-sets of intergenerational mobility fell in the UK Birth Cohort Studies’, mimeo. between the 1958 and 1970 cohorts. CEP contributions have moved the John Goldthorpe and Michelle Jackson (2007) debate past the fall ‘Intergenerational Class Mobility in in mobility, to Contemporary Britain: Political Concerns and investigate ‘what Empirical Findings’, British Journal of happened next’ and Sociology 58: 526-46. whether greater inequality in Paul Gregg, Susan Harkness and Stephen a nation leads Machin (1999) ‘Poor Kids: Trends in Child directly to less Poverty 1968-1996’, Fiscal Studies 20(2): social mobility 163-87.

Paul Gregg and Stephen Machin (2000) ‘Childhood Disadvantage and Success and Failure in the Labour Market’, in David Jo Blanden is a lecturer in economics at the Blanchflower and Richard Freeman (eds) University of Surrey and a research associate Youth Employment and Joblessness in Advanced in CEP’s education and skills programme. Countries, University of Chicago Press.

9