The unreliable Oxbridge signal

Oxbridge seems to recruit extraordinarily heavily from a small group of southern private schools (presumed to include Eton and Westminster) according to a new analysis by the estimable Sutton Trust.

So what? Why do we care? The answers probably seem obvious. An Oxbridge education appears to unlock the upper echelons of the British establishment.

As work by the Sutton Trust has previously shown, a startlingly high proportion of judges, senior civil servants, business executives, cabinet members and journalists (disclosure: I went to Oxford) were educated at those two quadrangle-festooned institutions of higher education.

If Oxbridge is drawing disproportionately from fee-paying private schools, these are likely to be children from relatively wealthy (if not super wealthy) families. Those children of the rich thus appear get a sizeable additional advantage in their adult life (at least if one considers entering parliament, law, the civil service or the higher reaches of the media to be a desirable destination).

This is poisonous for British social mobility, a monstrous affront to any reasonable conception of fairness.

That's why we should care. But let's take a step back. Imagine we were designing a system for selecting capable and ambitious graduates for advancement in those kinds of professions.

What would that system look like? Once you ask this question, the central importance of Oxbridge as a recruiting pool starts to looks absurd, no matter who goes there. The numbers studying at Oxbridge have risen since higher education steadily opened up to the masses after the Second World War. But the growth in new places has not kept pace with the explosion of total student numbers over the past half century.

Perhaps they should double the number of Oxbridge colleges and build on the green belt around the two picturesque cities to accommodate the secular rise in demand. But, in the real world, the annual Oxbridge undergraduate intake is destined to remain tightly limited by geographical constraints.

The two universities currently accept around 7,000 new undergraduates each year. That's now less than 1.5 per cent of national total undergraduate acceptances in 2017 (533,890). Since Oxbridge receive around six applications for every acceptance, admission is a zero sum game: one person's place is a place not won by five others.

Why are Oxbridge students so dominant in the upper reaches of certain professions? Many will say it's because so many smart children tend to go there. And that may well be true to a certain extent. But it's not the whole story. And it misses a vital economic mechanism: signalling.

Oxbridge transmits a "signal" to employers and gatekeepers of a capacity on the part of a candidate for exceptional perseverance and capability. (For some it will be signal of social class too but let's park that for the purposes of this argument.) Many employers faced with a hiring choice between two equally strong candidates for fast-track training or some other opportunity will opt for the Oxbridge one because of that signal.

Yet the signal may not be accurate. Indeed, it may be becoming less accurate over time given the rise in the numbers attending higher education. To put it another way, what reason is there to believe that those five ambitious children who didn't get in to Oxbridge have less potential to be a top lawyer, accountant, journalist, business executive or an MP than the one who did? But note the incentives the signal creates. It gives smart and ambitious children an added reason to apply to Oxbridge because they correctly discern that their chances of future advancement are higher there. This battle for places among bright kids reinforces the signal of quality, creating a vicious circle.

None of this is to argue that efforts to reform Oxbridge admissions are irrelevant or unimportant. They certainly aren't. But perhaps the bigger challenge is to, somehow, erode the prestige of these two institutions in employers' eyes, to interfere with the signal.

We need to encourage a culture where employers and gatekeepers, whether in the public or private sector, do not look at these two midsized universities as some kind of gold standard. We need recruiters to wake up (more fully - many, in fairness, have) to the reality that there are also vast numbers of extremely able and talented people elsewhere too.

In other words, do not simply concentrate on making access to the elite recruitment pool fairer but try to make that pool much bigger too. Bear in mind too that, in such a transformed environment, bright students would not have such an overwhelming incentive to apply to Oxbridge and to engage in that zero-sum competition for places. The vicious circle could become a virtuous one.

The problem with "elitism" is not so much that it privileges ability (leaving aside concerns about the social implications of a true meritocracy) but that it is so arbitrarily exclusive.

It's a sad irony that the intense media focus on Oxbridge admissions may serve to emphasise the institutions' signalling power. If everyone thinks it's terribly important how these universities select students, doesn't that underline that they are also terribly important institutions, that "the best" are to be found there? University-blind admission processes, like the one introduced by Deloitte a few years ago, are welcome experiments and we should hope for more of them. If we paid less attention to where others went to university the world would be a less tedious place. It could also be more just.

How to solve the Oxbridge problem with poor people and minorities

"People make rational decisions and respond to financial incentives. Most economists use those assumptions as rules of thumb to predict behaviour. But though they are often useful, they don't always work.

Parliament's Treasury Committee last week complained about the stark lack of women at senior levels of the Bank of England. This was attributed by the new deputy Bank of England governor, Dave Ramsden, to the fact economics is a heavily male-dominated course at universities, making the pool of potential female appointees to senior economics roles smaller.

It was also a week in which the Labour MP David Lammy drew attention to data he had collected confirming the under-representation of black and working-class students with top A-level results at Oxford and Cambridge.

From a financial incentives perspective such under-representation is puzzling. Economics graduates earn considerably more on average than most other graduates, with a median salary of around £32,000 after five years. That's around £10,000 more than the equivalent for science or English students. And the average salaries of those who graduate from elite universities such as Oxford and Cambridge are also relatively high. Research by Boston Consulting Group suggests Oxbridge graduates earn an average of £1.8m over their lifetime, compared with £1.6m for those who attend other respected universities.

Why? It's possible those who choose economics and Oxbridge are more financially motivated and capable than the average student and that this explains why they go on to earn more; that the courses themselves don't really add much value. But it's also highly likely that those degrees act as an important signal to employers, which is advantageous to the graduate when they enter the jobs market. Such a signallingbenefit would make it economically rational for far more bright women and black and working-class students with sufficiently good A-level results to apply for those courses and those institutions.

Mr Lammy fears discrimination is behind the relative lack of black and working-class students at Oxbridge.

It would be foolish to dismiss this as a factor, particularly the unthinking form of discrimination whereby interviewers might be unconsciously tempted to select people in their own image. Yet it's unlikely to be the whole story. Oxbridge colleges and economics departments in recent years have stepped up outreach programmes. Many academics at these institutions are genuinely concerned with widening access and increasing ethnic and gender diversity.

There is plainly also a demand side-issue. Too few qualified black students apply to Oxbridge. And evidence shows too few well-qualified young women choose to study economics. To explain this, the insights of behavioural economics are likely to be more useful than the classical theory of rational choice, particularly the propensity to herd behaviour. If those in your school or social peer group are very unlikely to apply to elite universities, it can have a powerful discouraging affect - outweighing or even cancelling out a rational estimate of financial benefit.

Gender stereotypes about domains of ability may also be relevant in terms of deterring women from economics. A study last year showed women were more likely to underestimate their ability in subjects deemed to be more "male", such as economics. Cultural factors are also relevant. A young US scholar Alice Wu recently presented evidence of a misogynistic culture on a popular anonymous chat site for economists. A well-grounded fear of future discrimination is likely to put people off applying for the course, even if their grades at school mean they are qualified on paper.

So does this mean elite universities and economics departments can shrug off responsibility for under-representation of various groups such as women, ethnic minorities and people from a disadvantaged background? Can they put the blame on pervasive gender stereotypes in society and hardwired psychological biases? On the contrary, it shows they need to work even harder to offset them. They need to spend more resources on outreach, to collaborate more effectively and intimately with schools and sixth-form colleges to encourage applications from under-represented groups. And, in the most difficult challenge, they need to reform their cultures if those cultures are putting off potential applications and reinforcing damaging stereotypes.

If under-representation was a consequence of primitive discrimination by a social caste determined to keep hold of its powers and privileges - what Mr Lammy calls "social apartheid"- it would be a relatively straightforward problem to solve. But the problems are complex, multi-faceted and structural. And so must the solutions be.