Social mobility: the nasty arithmetic

Brown’s desire to increase social mobility seems to have led to two errors.
One is expressed by Stephen Pollard and Wat Tyler - that the decline in social mobility is due to the abolition of the grammar schools.
I doubt it. Yes, grammar schools do seem to educate people better. But they can’t be a great force for upward mobility simply because very few poor students get into them. This paper explains.
And remember - social mobility wasn’t very high even when grammar schools were ubiquitous.
The big evidence for declining social mobility is this paper, which compares the fates of those born in 1958 to those born in 1970. But it shows that, even among the 1958 cohort, sons from the richest quartile of families were 1.68 times as likely to get into the top quartile themselves as sons from the bottom quartile.
The second, bigger error is Brown’s claim that “parents must want their children to do better than they did themselves.”
Not if we want social mobility, they mustn’t. Social mobility is conventionally defined as the chances of making it from the bottom quartile to top quartile. Upward mobility by the poor therefore requires that the rich be downwardly mobile. Only 25% of people can be in the top quartile. If we want more poor kids to get into that 25%, we must have fewer rich ones in it.
And this highlights the big problem. Any serious effort to greatly increase social mobility will be deeply illiberal, as it requires the state to intervene in family life to stop rich parents passing their advantage to their children - for example by banning private education.
What are the alternatives?
One is to increase spending on educating children from poor families, to improve their chances relative to the rich.
But this would be humungously expensive. This classic US study found that equalizing opportunity between blacks and whites in the US requires up to 10 times as much to be spent on educating blacks as whites.
The reason for this is simple; the link between education spending and adult earnings is modest, so big rises in spending have only weak effects.
The back of a beer mat corroborates this. It costs seven times as much to educate a child at Eton as at a state secondary school. But to genuinely equalize opportunity between a poor kid and an Etonian requires more to be spent on the former, to compensate for the disadvantages of worse social ties or a less encouraging home environment.
Achieving equality of opportunity through education seems, therefore, impractical.
Which leaves another possibility - greater equality of outcome. This paper shows that nations with greater income equality have greater social mobility.
Again, this is basic maths. The nearer is the top quartile to the bottom quartile, the easier it is to move from one to the other.
So, I reckon Norm got it right. Equality of opportunity without greater equality of resources is just impossible.

Comments

[reason]

But this is not nasty arithmetic. You are just saying we need again a more progressive tax/social insurance system. Simple clear and well argued, then good we should do it. Otherwise sooner or later we will get a new Paris or St.Petersburg.

[dearieme]

"sons from the richest quartile of families were 1.68 times as likely to get into the top quartile themselves as sons from the bottom quartile": only 68% higher - what a remarkable egalitarian achievement! And you are wrong about Grammar Schools; you overlook the fact that their existence implies the existence of Secondary Moderns. That matters because it was a Sec Mod that rescued a cousin of mine and pulled off a promotion through several socio-economic bands. It's the advantage, you see, of specialisation in education: whichever sort of school you end up in, if it suits you, does you more good than a one-size-fits-all Comprehensive.

[Will Davies]

Surely what Brown really means is something that can't be captured using arithmetic, namely that any child of exceptional ability - and ambition - will be able to attain economic and political power. People refer to this as 'the American dream' as if it is a romantic ideal doesn't exist. It DOES exist in the US, it's just that you have to view it in an especially romantic light to find it appealing.

This model is not really social mobility, in fact it is a panacea for social mobility. It requires nurturing a sense that the poor deserve to be poor and the rich deserve to be rich, because of the small minority of people who move between the two. It's a form of post hoc aristocracy - this minority sits at the top, so they must be the best people to do so.

If it's poverty per se that offends Brown, then he should focus on poverty. But I suspect that what actually offends him is the idea that a very small minority of poor children 'deserve' better.

[Robert]

Here's an article from Philip Collins which makes a similar point:

http://www.newstatesman.com/200201140017
"More downward mobility, please: When dull middle-class children sink, and forsake avocado for tinned peas, we shall have a truly fair society"

[Bob B]

"It costs seven times as much to educate a child at Eton as at a state secondary school"

Two local maintained (non-fee paying) grammar schools for boys within walking distance of where I sit achieved better A-level results than Eton in last summer's exams with nothing like the costs or teacher-pupil ratios of Eton:
http://news.bbc.co.uk/1/hi/education/7180103.stm

A bus ride away and a local girls grammar school did better and other local grammar schools weren't that far behind. The instructive insight is that the London Borough of Sutton is top of the Local Education Authority league table for England based on AVERAGE attainment in local schools in the GCSE exams - the presence of selective local grammar schools raises the average attainment:
http://news.bbc.co.uk/1/hi/education/7180228.stm

Other curious factors is that on the official stats for London boroughs, the local distribution of income is very much like the average for London and Sutton is not a big spender on education. In fact, the borough is rather a low spender compared with other London boroughs with schools that didn't do as well.

On the evidence, it really is nonsense to claim that good A-level results can only be achieved at high cost by non-maintained schools like Eton.

[Bob B]

These figures for household income distribution in the London boroughs confirm my claim above that Sutton has virtually the same distribution as the London average - namely, 21% of Sutton households have a household income of less than £15k (compared with 22% for London); 53% have a household income less than £30k (53% for London); and 85% have a household income less than £60k (85% for London).
http://www.londoncouncils.gov.uk/londonfacts/londonstatistics/Householdincomedistributionin200607.htm

From a national perspective, I would have thought that these factors will likely have a significant impact on social mobility:

"Last year [2004], a report from the Organisation for Economic Cooperation and Development (OECD) revealed that Britain came seventh from bottom in a league table of staying-on rates for 19 countries. Only Mexico and Turkey had significantly lower rates of participation for this age group. Italy, New Zealand, Portugal and Slovakia have marginally lower rates."
http://education.guardian.co.uk/gcses/story/0,16086,1555547,00.html

"Some 26 million adults lack maths or English skill levels expected of school-leavers. . . An estimated 5.2 million adults have worse literacy than that expected of 11 year olds, while 14.9 million have numeracy skills below this level." [BBC]

[Bruce]

Isn't this the same illogic that has Brown trying to lift children out of relative proverty?

[Bob B]

Malicious rumour has that this tribute to the London Borough of Sutton could be one explanation why schooling there is so good:
http://www.youtube.com/watch?v=W7eff_O3Wwg

Some readers may recall reading about Chris Woodhead, the somewhat infamous chief inspector of schools in England from 1994 to 2000, who was given to expressing such outrageous observations as: "I am deeply shocked by the degree of waste within the independent sector," and, "The difference between schools serving very similar catchment areas is extremely wide."

It was only after my son had gone on to university that I learned Chris Woodhead had attended the same maintained grammar school just down the road.

Btw I was amazed to learn that only 6.7 per cent of school pupils in Britain attend non-maintained (fee-paying) schools and only 5 per cent attend maintained grammar schools, for all the harm these institutions supposedly inflict on social mobility.

[james higham]

Upward mobility by the poor therefore requires that the rich be downwardly mobile.

Yes but the whole model is wrong, Chris. It has nothing to do with percentiles.

[Dipper]

Bob B I'm not sure that all those impressive statistics on Sutton mean anything? Surely some selection of population is going in, as it is relatively straightforward to move in and out of the borough, so the comparisons with neighbouring boroughs are not like-for-like

[Dipper]

I can't help feeling that the use of blanket statistics cloud rather than illuminate, and there's a lot of faulty logic going on here.

Chris recently referred to a study showing children of middle class lefties who went to inner-city schools did well. Many ethnic groups have been upwardly mobile in recent times with children performing very well at inner city schools.

If a school can produce just one person who is an academic success, then the school has shown what it is capable of. Analysis should then be on the basis of why other children have not been able to copy this success, not using bulk statistics to smack teachers and schools round the head.

[Bob B]

Dipper: "I'm not sure that all those impressive statistics on Sutton mean anything?"

C'mon. Sutton has been continuously at or near the top of the Local Education Authority league table for England since the league tables were first launched in the early 1990s. It seems to me a pretty obvious and basic question as to why that is so.

By the Greenwich judgement of 1989, grammar schools were barred from giving priority in offers of places to local residents:

"It was back in 1989 that what has become known throughout the education world as 'the Greenwich judgement' established that local education authority-maintained schools may not give priority to children simply because they live in the council’s geographical area."
http://www.londoncouncils.gov.uk/doc.asp?doc=12440&cat=1581

It is evident from my bus trips to go shopping for groceries at the end of school days that a percentage of boys going to the school down the road are catching trains and buses away from the local catchment area but I'm informed that this is very definitely a minority of places. Sutton's income distribution - compared with London's - is therefore relevant to the frequent sweeping claim that the success of grammar schools is largely due to the pupils having affluent parents. In the case of Sutton, the parents are very average by London standards and there is consistent local support for the retention of the excellent grammar schools.

This is hardly surprising as the pupils in these local maintained schools are getting as good or better schooling than at Eton. That should lead us on to another obvious question: How come, given all the extra resources at the disposal of Eton and the affluence of the parents of the pupils who attend there?

[Bob B]

Apropos my above post and in case readers missed this little news gem from two years ago:

"Fifty leading independent schools, including Eton, Harrow, Winchester and Cheltenham Ladies' College, have been fined a total of £3.5 million after being found guilty of fee-fixing by the Office of Fair Trading."
http://www.telegraph.co.uk/news/uknews/1511429/50-public-schools-fined-for-fixing-their-fees.html

[Quinn]

Bob B; loathe as I am to indulge you, but on the one hand you state that "Sutton's income distribution - compared with London's - is therefore relevant", and on the other hand you say "grammar schools were barred from giving priority in offers of places to local residents". In other words, is it safe to say that Sutton's school results are bumped up by including some of the brightest pupils from the neighbouring areas? In saying that I'm assuming that those in neighbouring districts aren't clamouring to get into Sutton's Secondary Moderns.

[Bob B]

"In other words, is it safe to say that Sutton's school results are bumped up by including some of the brightest pupils from the neighbouring areas?"

In part, although the neighbouring borough Croydon has its own selective school - Whitgift - and the nearby borough of Kingston has several, such as Tiffins Girls School.

As for secondary moderns in Sutton, other local secondary schools are Comprehensives. No one is obliged to apply to the excellent grammar schools to take the entrance exams.

But all this is evading the essential points here:

- Sutton's grammar schools have nothing like the funding or teacher-pupil ratios of Eton, nor have the dozens of other grammar schools elsewhere which produce better A-level results. Sutton is a relatively
low spender on education.

- the claim that schools need equivalent funding to Eton to achieve similar results is therefore false. Norm is flat wrong.

- Chris Woodhead's observation: The difference between schools serving very similar catchment areas is extremely wide.

- the percentages of pupils attending non-maintained schools (6.7 per cent) or grammar schools (5 per cent) are, in fact, relatively small - and the non-maintained sector has a long tail of indifferently achieving schools.

- focusing the debate on non-maintained schools and grammars overlooks the negative contribution to social mobility from the high drop-out rate from all education and training in Britain compared with almost all other advanced industrialised countries.

- it also overlooks this worrying factor:

"Some 26 million adults lack maths or English skill levels expected of school-leavers. . . An estimated 5.2 million adults have worse literacy than that expected of 11 year olds, while 14.9 million have numeracy skills below this level."
http://news.bbc.co.uk/1/hi/education/4095153.stm

[Bob B]

As for the supposed harm inflicted by secondary modern schools, try this on the (illustrious) career of Sir Peter Mansfield who attended one after failing his 11+:
http://en.wikipedia.org/wiki/Peter_Mansfield

[Chris Williams]

Can we have some kind of Godwin's Law against argument from anecdote about education policy? I can list people who went though all sorts of different forms of education who did well, amd/or others who went through just the same forms and did badly. So what? It's the numbers, stupid. Bob, I'll see your Sutton and raise you Kent.

[Comprehensive (ex Sec Mod) and Oxford U, since you're wondering].

[Bob B]

"I'll see your Sutton and raise you Kent."

Yes - although Sutton's continuing ranking at or near the top of the LEA league table does raise difficult questions about the best structures of schooling in LEAs - on the evidence.

Factors we have left out of the discussion so far are the cultural and ethnic dimensions, both or which are potentially significant.

English working class cultures have a lingering antipathy to education values. George Orwell remarked on it in: The Road to Wigan Pier (1937) chp.7 and, in places, not much has changed since he wrote that. And we have this recent insight:

"Government figures show only 15% of white working class boys in England got five good GCSEs including maths and English last year. . . Poorer pupils from Indian and Chinese backgrounds fared much better - with 36% and 52% making that grade respectively."
http://news.bbc.co.uk/1/hi/education/7220683.stm

That might shed some illumination on the poor Kent results with selective schools. As it happens, the school just down the road from where I sit, with better A-level results than Eton, is anything but WASP. It's pupils are hugely ethnically mixed.

As for anecdotes, from time to time I've met various university professors who almost boasted that they had failed their 11+ exams. By several accounts, Einstein did badly in school exams, which perhaps explains why his first job was as a clerk in the Swiss patent office.

As for numbers, sadly, what tends to happen in some non-selective schools is that pupils - especially boys - with academic interests get bullied. Certainly, there are often peer group pressures to conform with group norms which is probably why Britain has one of the highest drop-out rates from education and training at 16 according to that OECD survey cited above here.

I have a problem understanding why pupils can be selected for sports academies but not for academic schools:

"GCSE performance in arts, business, languages, mathematics, technology and science schools is no different from other state secondaries, they said. They found that only schools specialising in sports showed any difference, with slightly lower GCSE results than other schools on average."
http://education.guardian.co.uk/newschools/story/0,,2164505,00.html

Btw why do we almost always debate this issue in such an insular way? What of the (selective) Gymnasium schools in Germany and the Netherlands and the Lycées in France?

[dearieme]

"argument from anecdote": but so often the "figures" available for social issues are rubbish - at least anecdote gives you something factual to test conjectures against.

[Chris Williams]

Just because some 'social research' is rubbish, there's no reason to ignore it altogether in favour of what some bloke down the pub reckons. Unless that's the only way to advance your otherwise untenable position, of course.

[AnInfantileDisorder]

My version of the Godwin's law would be the automatic response to a social mobility question being an education debate. What the useful definition of mobility (movement from bottom to top quartile) points up is the need for an argument in favour of the progressive nature of inheritance taxation. The Tories managed to frighten the middle classes about the value of their homes and try to make abolition of inheritance tax popular and then Gordon ineptly tried to steal their policy. That was a cheap tactic and interestingly, marked the start of his problems.

[TomJ (OW)]

Just for the record, there are 3 other independent selective secondary schools in Croydon: Trinity (Whitgift Middle School as was)for boys; for girls, Old Palace and Croham Hurst, though a quick google tells me these laater are about to merge.