Monday 18 January 2010

Introducing the social mobility fact-check service

We will hear a lot about social mobility in the next few months, as Gordon Brown's speech on Saturday to the Fabians and David Cameron's response on education today suggested. More broadly, it is probably a good thing that worrying about 'low social mobility' offers a polite and socially acceptable way for almost everybody to express a concern about how much class structures British society, even if some seem at times unaware that this is what they are saying.

But more nonsense is talked about the facts of social mobility than perhaps any other public issue. So Next Left today begins a modest "social mobility fact-checking" service, aimed at politicians, campaigners and public commentators, and would welcome other bloggers and commentators joining a push to name and shame the public discourse into a more accurate discussion of mobility in British society.


As this could prove an equal opportunity exercise, let us begin with LibDem leader Nick Clegg and right-wing blogger Guido Fawkes, who both got the social mobility facts wrong yesterday.

(1) "Social mobility has fallen under Labour", said Nick Clegg in his Andrew Marr interview, in passing but hardly for the first time. This is one of the most commonly repeated mythical political claims.

It isn't true.

As the most influential academic study of this question reports:

We cannot find any evidence that the sharp drop in mobility observed for children growing up in the 1970s and 1980s has continued. But nor can we find evidence that mobility has improved.'

(2) Paul Staineswrote on Guido Fawkes yesterday that council house sales were "the single greatest boon to social mobility since free education for all". You might date that to 1944, or to a large extent 1870, so its quite a claim. But I can not see what evidence Staines could possibly have for this rather more extravagant piece of social mobility mythology.

This would seem to both deny that the post-war 'room at the top' wave of social mobility took place, while also imagining a century-best surge in social mobility in the very period when it declined most sharply. (In referring to 'social mobility', Staines may have meant something different: perhaps that council house sales were a boon to a broader dispersal wealth and assets. That is also commonly believed and often asserted. Since this period was one of an increased concentration of wealth, as an increasing number on the right including Ferdinand Mount and Red Tory Phillip Blond now also stress, the positive effect of council house sales was outweighed by stronger drivers towards greater asset and wealth inequalities: that is itself one of the contributors to stratification and immobility over time).

So what are the key facts on social mobility?

(Future fact-checking ripostes can be much shorter if we gather some key sources in one place to start with).

The most cited research on social mobility has been carried out by Jo Blanden, Paul Gregg and Stephen Machin, Centre for Economic Perormance, LSE and Bristol University, with the Sutton Trust doing much to popularise this.

Their finding that social mobility was lower for those born in 1970 than those born in 1958 is the point which informs most public commentary about 'falling social mobility'. Because that research was published with new Labour in power (their main research papers on the 1958/1970 cohorts were published in 2002 and 2005), this helped to create a 'social mobility has declined under Labour meme', despite the researchers consistently pointing out that it is a misreading of their findings, and that the research could not be said to measure the impact of post-1997 policy.

It is very strange to blame the New Labour government's for the occupational outcomes of those who were 27 in 1997.

Since we are measuring intergenerational mobility, it takes some time to have robust results. As Anthony Giddens has explained:

It is absurd to use these findings to argue that social mobility has gone down since Labour came to power, nor did the study claim any such thing. It takes a minimum of 30 years to measure how socially mobile someone is, because we are comparing the jobs people are in today with those of their parents. It is commonly agreed that what happens in childhood is crucial to a person's job chances, hence if social mobility has declined it is the result of influences dating from the 70s and 80s, when, in fact, the Tories were in power.


Any impact of SureStart, improved primary education scores, inner city academies or the modest income redistribution towards the lower 50% of the income range, would reveal itself fully over time.

However, there are proxy measures which can enable a preliminary prediction about social mobility impacts. These rely largely on the observation that early scores in educational tests have been a strong predictor of later educational and employment outcomes.

So the LSE/Bristol team December 2007 research paper went on to report on the children of the 1958 and 1970 cohort, who were (on average) born around 1985 ('58 cohort), now in their 20s, and 1999 ('70 cohort), using "the relationship between children's educational outcomes at different ages and parental income we can predict likely patterns of mobility for cohorts who have not yet reached adulthood", allowing the researchers to begin to examine the impact which more recent policy could have.

Their conclusion was:

Evidence from the earlier 1958 and 1970 cohorts shows that as mobility declined in the past the relationship between intermediate outcomes and parental income strengthened. We therefore conclude that, under realistic assumptions and in the absence of any significant unanticipated changes, the decline in intergenerational mobility that occurred between 1958 and 1970 birth cohorts is unlikely to continue for cohorts born from 1970 to 2000. Mobility is therefore likely to remain at or near the relatively low level observed for the 1970 birth cohort.

In an LSE article will the downward trend continue?, explaining the research conclusions, they wrote

It appears that the decline in social mobility may well have flattened out. This may be either good news or bad news for policy: while it may have stopped the previous decline, it has failed to lead to an overall improvement in mobility.

There is more information on the 2007 study's conclusions here.

A more recent (2008) paper by Paul Gregg of Bristol University found family background less important for those who sat GCSEs in 2006 than 1970. This was reported as a sign that social mobility could be rising again. Sheffield academic Danny Dorling warned against making too much of this social mobility green shoot.


The Cabinet Office last year produced discussion paper analysing social mobility trends, which offers a fair summary of the academic evidence on occupational changes, opportunities and international comparisons.

The Cabinet Office report also sets out why the government has placed a particular emphasis on political economy strategies to increase the number of skilled and white-collar jobs, seeking to create more 'room at the top' in a globalised economy. This is intended to, at least partially, escape the meritocracy dilemma which Stuart White notes of whether or not those advocating more (upward) social mobility are prepared to advocate 'downward social mobility' too. (However, it is unlikely that any effective 'social mobility' strategy could do so entirely, particularly where educational outcomes, professional careers or other goods are seen as 'positional goods').

Some accounts of social mobility propose a more optimistic reading than the cohort studies. David Goodhart wrote a Prospect piece reporting contested debates within the academy over the meaning of social mobility and how to measure it. Goodhart's own argument was that we should be more optimistic about the levels of mobility within the 'low mobility', and more sceptical about the fact or extent of its decline.

It is true that "low social mobility" does not mean a rigid caste system, though Goodhart recognises the danger of complacency in pushing that analysis too far, and does highlight the particular UK challenge of a relatively closed professional elite at the top. With the UK ranking near the bottom of the occupational social mobility table, with only the USA doing worse, there ought to still be plenty to concern meritocrats, as well as other egalitarians.

Whether or not you agree with Goodhart's conclusion, his article does also highlight some of the nuanced academic debates about what to measure and how. David Willetts, undoubtedly the most social science evidence-based minded member of the opposition frontbench, has been among those to note that the feminisation of the workforce over the last half century makes the social mobility story more complex. Women's mobility has been higher than men's between the 1958 and 1970 cohorts, though this may be more due to changes in occupational structure of the economy more than a diminishing effect of family background.


The evidence that social mobility has been stabilised but not reversed highlights a broader difficulty for the government in defending its record.

John Denham made this point well in the equality discussion at Saturday's Fabian conference:

If you designed a social policy to have no impact, the effect of global forces would be to make opportunity more unequal. To halt or modestly reverse trends is, because of the downward escalator effect, a significant achievement in itself. we need to be careful to argue that, though I think it is clear that we do have to go further in the future

This also applies to the related but distinct issues of poverty and inequality, as Stuart White's Next Left analysis of the record set out.

On social mobility too, the counter-factual case that policy has not made a difference could prove a recipe for a further collapse in social mobility.

I felt the reality of that danger was captured well in Robert Yates' insightful Observer magazine feature from Liverpool Walton. Improving employment opportunities in Britain's poorest constituency remains very difficult; but the most striking change is the step-change in the quality of public services. The most recent educational results show inner city schools improving most.

That this may be a necessary but not sufficient condition to kickstart social mobility is a more convincing than the idea that policy has made no difference and could be reversed without a negative impact.


We will try to continue to highlight intelligent approaches to illuminating the challenges of social mobility - recommendations are welcome; as are academic offers to contribute on how public discussion could reflect research findings - as well as challenging those accounts which seem at odds with the social mobility facts.


Newmania said...

Face your own problem . Social mobility is actually a left -safe way of talking about opportunity whereas in its heart the left wants equality of outcome
If there is opportunity the argument for state control becomes weaker and the role of the left disappears and that is why it is commuited to maintaining poverty by rewarding idleness and creating an underclass

Robert said...

Lets get this right, New labour has been a walking talking living disaster.

they talk the talk sadly thats it.

Social mobility died a death now Labour is talking crap about 1000,000 new jobs for middle income groups what about the rest, we do need a New Government brown has nothing at all to offer.