Civitas on Social Mobility

Civitas Institute for the Study of Civil Society

18 August 2010
Policy Briefing

Will Nick Clegg escape the Social Mobility Myths?
By Peter Saunders

Deputy Prime Minister Nick Clegg made a major speech today in which he identified ‘promoting social mobility’ as ‘at the top of our social agenda’. Tony Blair made a similar commitment in 1997, just as John Major did in 1992. It is something modern leaders feel compelled to talk about. But Clegg brought a fresh take to it by emphasising the importance of parenting. Coupled with the Coalition’s initiatives on tax reform and early years intervention, this should help focus attention on the real social mobility problem – the self-reproducing, welfare-dependent underclass.


Like many politicians and commentators, Clegg thinks we have a serious social mobility problem in Britain: ‘It really, really gets to me that even though … we are a relatively affluent country, children are pretty well condemned by the circumstances of their birth.’ Unusually, though, he was prepared in his speech to identify poor parenting as part of the problem. Parents, he said, are ‘on the frontline’ and must interest their children in education.

He’s right. Unfortunately, though, he has appointed former Health Secretary, Alan Milburn, as social mobility special adviser to the Government. And Milburn seems convinced the problem is institutional (in his letter accepting this latest appointment, he saw his task as ‘assessing the progress each set of institutions is making to [increasing] opportunities’).

A closed-shop society?

Milburn chaired a report on recruitment into the professions in 2009 (Unleashing Aspiration) in which he claimed: ‘Birth, not worth, has become more and more a determinant of people’s life chances.’ He even described Britain as a ‘closed-shop society’. He thinks institutions are blocking people.

Milburn will now report annually on how social mobility is being improved across the public sector, including the NHS and universities. This should set alarm bells ringing for anyone who knows anything about this issue, for it signals more destructive and ineffective social engineering could be on the way.

Opportunity is already extensive

Clegg is right to want every child to enjoy the opportunity to exploit their talents to the full. Nobody wants to see bright and hard-working people blocked through no fault of their own. As he said, ‘Fairness means everyone having the chance to do well, irrespective of their beginnings. Fairness means that no one is held back by the circumstances of their birth. Fairness demands that what counts is not the school you went to or the jobs your parents did, but your ability and your ambition.’

But what he may not understand is that, for most people living in Britain today, this is already the situation.

Britain is not a perfect meritocracy, of course. Like in every other country, children benefit if they are born to supportive parents who care about their education and make sacrifices to help their kids excel. And not everyone has parents like that. Nevertheless, the evidence from social mobility research is that, if you are bright and hard-working, and your parents have a job (no matter what it is), you will almost certainly succeed in modern Britain:

* Dividing the population into three classes (professionals, managers and administrators at the top, manual workers at the bottom, and others in-between), more than half the population of Britain ends up in a different social class from the one they were born into.
* Of children in the top quarter of the ability distribution, only 1 in 20 ends up in a semi- or unskilled working class job, while two-thirds get professional-type careers, irrespective of the class of their parents.
* Movement is extensive, up and down these classes. More people born to working class parents are upwardly mobile by the time they reach 30 than stay in the working class. Downward mobility is also common: more than one-third of middle class children fail to stay there.

Given these figures, it is outrageous that Milburn last year described Britain as a ‘closed-shop society’ where birth counts more than worth.

Why middle class children out-perform working class children

It is true that children born into middle class homes tend (on average) to out-perform children born into working class homes. A child of manual worker parents is about three times less likely to achieve a professional/managerial position than a middle class child.

But it is a mistake to assume (as several recent government reports, including Milburn’s, have assumed) that this means there are unfair advantages or blockages at work. What this explanation neglects to consider is the distribution of talent.

When employers take on new employees, they try to recruit the most talented and able people. This creates a talent gradient across the occupational classes – people in the top jobs tend to be brighter on average than those at the bottom. These people usually find partners of a similar ability level (demographers call this ‘assortive mating’). And between them, they tend to have children whose ability to some extent reflects their own. The result is that, in each generation, a disproportionate number of middle class children is born with the high ability needed to get the best jobs. Hence that 3:1 ratio.

Nobody likes to talk about this – least of all, politicians. When they see middle class kids outperforming working class kids, they prefer to blame ‘unfair social conditions’. But the principal explanation is differences in average ability levels.

In research recently published by Civitas, half of the explained variance in the occupational destinations achieved by the 1958 birth cohort was due to just one variable – how well they scored on an IQ test when they were aged 11. This is a much better predictor of their eventual fate than the class they were born into, the type of school they attended, or any other social factor.

The fallacy of social engineering

Because politicians don’t like talking about ability differences, they keep trying to tweak educational and occupational selection procedures so more working class children will clear the assessment hurdles. They have been doing this for 50 years: abolishing grammar schools, scrapping streaming, turning Polytechnics into universities, expanding higher education places, dumbing down A-levels, attacking the private schools. Given the chance, Milburn will pursue more of the same (e.g. by penalising top universities if they do not take more entrants from lower social class backgrounds).

But none of these radical reforms has changed anything: ever since the war, middle class kids have been out-performing working class kids in that same ratio of around 3:1. All that more social engineering will achieve is a further diminution of educational standards.

Focus on the welfare underclass

There is one thing governments could and should be doing, though: intensively targeting children growing up in households where nobody has a job and parents are neglectful or absent. It is here, one suspects, that the real problem lies, and it is the one gleam of hope to be taken from Clegg’s speech. As he said: ‘Parents are in the frontline when it comes to creating a fairer society, in the way that they raise their children.’

It is often forgotten that research on social mobility excludes children in jobless households. When economists study income mobility, they exclude people with no earned income. When sociologists study occupational mobility, they exclude people with no occupation. The real social mobility problem is almost certainly concentrated in the welfare underclass of this country – but they are not being picked up in mobility statistics.

Instead of harassing Oxford and Cambridge to change their selection criteria, or fiddling with taxes and benefits to flatten the income distribution (something the 2009 Milburn Report was very keen on), politicians should devote their energies to improving parenting for young children growing up in welfare ghettoes. For everyone else, the opportunities are already there if they have the ability and motivation to take advantage of them. Milburn must be told there is no case for more social engineering.

Policy implications

There are three key policy implications that follow from all of this:

1. Help parents in workless households to assist their pre-school and older children.

2. Require people capable of work but currently on welfare to take a job.

3. Improve schools and provide supplementary schools so that they make up for unsupportive parents.

And one thing to avoid: don’t waste time on income redistribution or more quota-based social engineering. Not only are such initiatives ineffective, they are also counter-productive. When politicians repeatedly blame social conditions for outcomes, they breed fatalism by discouraging us from making the effort to overcome obstacles ourselves.

For more information contact:

Peter Saunders on: 07900 412420

Civitas on: 020 7799 6677

Notes for Editors

i. Civitas is an independent think tank. It receives no state funding either directly or indirectly and has no links to any political party.

ii. Peter Saunders is Professor Emeritus of Sociology at the University of Sussex and Professorial Research Fellow at Civitas. For further information about the author see:

iii. To buy Social Mobility Myths by Peter Saunders, click here. A press summary of the report can be read here.

This e-mail was sent to because you are subscribed to at least one of our mailing lists. You can easily remove yourself from our mailing list by visiting this link. One click will confirm your removal:

11 responses to “Civitas on Social Mobility

  1. chris southern

    The biggest problem is that all of this has been known for a very long time.
    Politicians like to blame everything but the real problems as admitting they have not only been wrong but have screwed millions of peoples lives up would lose them votes (and votes make prizes.)

    Why are intelligent but not academically minded kids forced to spend two years getting bored stupid when they could be starting apprenteships and spending some time in education on the neccesary subjects that are relevant to the apprenteship (many people pick up theory far quicker when ties in with the physical skills they are learning, it’s how I get kids to understand music theory in no time at all.)

  2. I find this bold assertion that members of the professional class are genetically superior, written by members of that same professional class, quite amusing.

  3. As you know, I am semi-persuaded there is something in it. Consider:

    1. Intelligent people *tend* to have intelligent children.
    2. Stupid people *tend* to have stupid children.
    3. In any society not purely based on caste, clever people will tend to rise into the higher classses.

    All this being so, an open society will tend after about a century to have similar mobility to a caste society.

    I’ve noticed this in America, where white and black working class people tend to be astonishingly thick, and most reliable manual workers are recent immigrants. I’ve also noticed how in Slovakia, where there has been little social mobility, there are some highly intelligent manual workers.

    The last time I made this point, Ian B went into a hissy fit and accused me of approving the most awful things. Back then, I shut up. However, I would say now that facts are facts. If there is any truth at all in the propositions given above, no good comes of denying them.

    Equally, the possible truth of these propositions doesn’t validate the sort of state action the eugenics people wanted. Governments are no better at planning population than they are at planning finance or heavy industry.

    And, on a possibly optimistic note, scientific progress seems to be taking us into a world where parents of whatever quality will be able to produce clever and beautiful children. Given a world of big states and big corporations, this will take us into the sort of world described in Gattaca. But that is another issue.

  4. And, on a possibly optimistic note, scientific progress seems to be taking us into a world where parents of whatever quality will be able to produce clever and beautiful children. Given a world of big states and big corporations, this will take us into the sort of world described in Gattaca.

    Your meaning is unclear, Sean. Is the ‘optimistic note’ the fact that science will allow the production of ‘clever and beautiful children’, or that we are going to end up in a Gattaca-esque world (or both)?

  5. Perhaps there is a note of irony and truth? Technology, after all, has saved us from starvation and dying of TB/Malaria/Polio/Smallpox . . . . .
    However. It might be an idea to back of trying to control things. I know at least one highly intelligent person whose life was somewhat trashed by the presumed best efforts of his parents and the education system.
    Sometimes our wisdom is a bit lacking.

  6. Sean, Pythaogoras, Archimedes, Aristotle and Plato were born of people who, just a few generations before, were simple herdsmen without written language, let alone mathematics or philosophy. The entire history of civilisation is only a few thousand years removed from a human population who, in its entirety, did nothing more than hunting and gathering, where there was little intellectual activity and progress was painfully slow. We have little understanding of how human intelligence works, or why it evolved at all. But it seems to be the case that it evolves without direct pressure to engage in intellectual activity. The ability to do math evolved long before math was invented.

    Every ruling class likes to believe it is special; indeed every class likes to believe it is special in some way, and will develop a justifying ideology. My “hissy fit” at your lapse into eugenics thinking was that you were echoing the ideology that got us into this mess. It is a short step from presuming superior ability in the ruling class to declaring that that ruling class has not just a right to rule, but a duty to rule.

    Anyway, there is a lot wrong with simplistic cod-Darwinism, which would take a long article to explain. But one thing we can use as an example is something called “Regression to the mean”. Statistical outliers tend to have progeny who are cloers to the average. Very tall people will have shorter children, and very short ones will have taller children. The same is true of intelligence. In order to preserve the unusual tallness or shortness, you need a reproductive selective process; that is you need to stop the statistically more average ones breeding. This doesn’t happen. The result is that unusual traits are not preserved. There is no means biologically for this genetic class distinction to preserve itself.

    This Darwinian argument is seductive and simple and continually reasserts itself in the hegemonic discourse. But it is wrong.

  7. “Sean, Pythaogoras, Archimedes, Aristotle and Plato ”

    I think that first comma ought to be a – or a ; or something :oD

  8. You are confusing education with intelligence. The ancestors of Pythagoras were not mathematicians, but they do not seem to have been stupid.

    As for objections to genetic determinism, there is a lot of persuasive evidence – the Bach and Mozart families in music, for example; also certain English artiscrocratic familaies. I don’t believe this is all the effect of upbringing. Look, children of white people are white. Children of fat people have a predisposition to be fat. The Hapsburgs mostly had that funny jaw. My daughter has been cursed with my big ears. We know that physical characteristics are passed on from one generation to the next – even if there are variations over time. Why should it be at all strange if intellectual and even moral characteristics can be passed on in the same way?

    As said, i am not making a case for compulsory sterilisation or murder of the unfit. I am simply stating an hypothesis for which there seems to be a lot of common sense evidence. When reasonable people talk of equality, they mean equality before the law, and a vague moral right to equal consideration and fairness. We don’t need to assert that all children are born with exactly the same potential, or that any variations in potential are randomly distributed.

    The only practical implication at the moment is that, if social mobility appears to be slowing down, it may be largley because the opportunities presented during the past hundred years have mostly now been taken.

    Oh, this is not to say that clever children will not continue, every now and again, to be born of stupid people. Nor does it deny that there are growing institutional barriers between the classes. For example, I was born in humble circumstances. I was able to pass a handful of objective examinations and go to university with a nice public subsidy. If I were 18 now, I’m sure I’d find everything much harder – examinations that don’t measure intellectual ability, the gross expense of study in the cartellised university sector, the difficulty of finding work in an economy more dominated than ever by big business and in which unpaid internships are increasingly the way into the nice jobs, etc etc.

    In a free society, there would be more social mobility than currently exists. Even so, it might be less than in the more social democratic second half of the 20th century. One way or another, the big sorting that can only happen once has already taken place.

  9. Observable data.

  10. The report, if you read it in full, is actually very interesting. The policy recommendations are mostly anecdotal and based purely on the writer’s prejudices, BUT the statistical analysis is very thorough and based on observable data, and there is also a good summary of philosophical views on “meritocracy” (view for it, and both egalitarian and libertarian views against) that is well rounded and treats all arguments seriously and with respect.

    A lot of the “meritocratic clases are just like castes” comments here don’t seem to have read the report because if they had, they’d be aware of “regression to the mean” – a well-known phenomenon in population genetics that is examined in this report. Broadly speaking, a child born to two short parents, is also likely to be shorter than average – but will most likely be taller than its parents. Similarly the child of two tall parents is likely to be taller than average, but is most likely to be shorter (i.e. will have regressed closer to the average) than its parents. Intelligence is believed to work in a similar same way. It’s worth remembering that the “musical genius” of the Bach family, or the mathematical genius of the Bernoulli family, all petered out after several generations. However, regression to the mean doesn’t imply that eventually everybody becomes the same. Random variation means that you get some people born who are exceptionally tall, and that this is more likely to happen to taller than shorter parents. And these people will have a chance of passing on the trait to their own children, although it’s most likely they too will regress back down towards the mean.

    This pheonomenon is why a “solidified”, caste-like class system with little social mobility, would be indicative of a failure of meritocracy. Eugenics doesn’t work in a clearcut way: a certain proportion of the children of the smart will be dim, and a proportion of the children of the dim will be smart. The interesting thing about this report is that it shows that the proportion of children from the lower classes going upwards would be well accounted for by an inheritable trait (intelligence is raised as one possibility, but the report also suggests that motivation may be heritable, or indeed other candidates exist) but that the proportion of children from wealthier backgrounds sliding down the scale, is too low. This suggests that bright but poor kids can work their way up, but upper middle class parents are to some extent – though not completely – able to restrict the slide down the scale of their less clever children. Interestingly, the report produces strong evidence that private vs state schooling (and indeed other aspects of educational structure) isn’t a key factor, but parenting techniques may have a large influence.

  11. At this time of year, though, precisely predicting what falls where is hard you want to do more than a matter of hours in advance, as Monday and Wednesday’s storms showed. There’s a finer line than everything you can put in the forecast a day in advance, Bryan Jackson, a meteorologist in the National Weather Service’s Baltimore/Washington forecast office, said for the divide between rain and wintry precipitation. The message we’ve been saying is snow and sleet, changing into rain. In both cases, the snow did accumulate across the Baltimore area before changing into rain. In the meantime, more treacherous weather is within forecast for Thursday.