The Professor Will See You Now
Illustration: Tracy Worrall
In an occasional series, Professor Philip Cowley offers a political science lesson for The House readers.
Here: the graduate voter. Illustration by Tracy Worrall
When Margaret Thatcher achieved her second landslide in 1987, seven in 10 voters had left education at 16 or earlier. Those with no formal qualifications made up 40 per cent of the electorate, outnumbering university graduates by more than five to one.
Fast-forward 30 years, to 2017, and the voters that Theresa May faced – with somewhat less success – looked rather different. Nearly a quarter had a degree, a group now more numerous than the unqualified.
The tipping point came at the end of the New Labour era. Before 2010, graduates were outnumbered, often massively, by those without qualifications. From 2010 onwards, graduates outnumbered those with low-level or no qualifications.
In their excellent book, Brexitland (from where I have cheerfully lifted the above data), Maria Sobolewska and Robert Ford identify the growth in education as one of the biggest changes in the composition of the electorate over the last 40 years.
When graduates made up such a tiny proportion of the electorate, no one really cared that much about their political behaviour. Early election studies paid little attention to education as a factor that might influence someone’s vote.
But as their numbers have grown – and continue to grow – they have become more important; a process that has been magnified by education becoming a more significant cleavage among the electorate, as a result of the Brexit referendum and the two Brexit-suffused elections that followed, where education was a significant divide.
A university education does have a direct impact on someone’s political attitudes, but it is much less than you might think
In 2015, the difference between the Conservatives’ lead among graduates and the rest of the population was never larger than seven percentage points. In 2019, by contrast, Labour enjoyed a small lead among graduate voters despite trailing among the rest of the population by more than 20 percentage points.
Some of this is the result of age. Following the post-1992 expansion of universities, graduates are on average still a relatively younger segment of the population. But this doesn’t explain all of it, because graduates have long been known to have distinctive attitudes – being, on average, more liberal than the rest of the electorate – a finding that dates back more than 50 years.
What is less clear is why. What causes this? For some it’s all about universities as Marxist madrassas, full of leftie lecturers, proselytising in the classroom when they should be teaching advanced algebra or 14th-century crop rotation.
As someone who has spent all his working life in and around universities, I am aware of this complaint but have always been a bit sceptical about the idea that academics have much direct impact on their students’ views. Given that they pay so little attention when I’m talking to them about departmental select committees or Early Day Motions, it has always seemed unlikely that I’d manage to shift their fundamental values all that much, even if I wanted to.
But if it’s not their time in the classroom, then what causes graduates to have different values?
Alternative explanations focus on a process of selection (that is, graduates are more liberal because liberals are more likely to go to university) or post-graduation sorting (graduates are more liberal because after university they earn on average more money and work in more liberal occupations where they hang out with more liberal people).
Disentangling these effects is incredibly difficult, because it requires data on attitudes over time or on people’s pre-university views and values. But two recent research papers, published within weeks of one another, attempted to work their way through this Gordian knot, and came to broadly similar conclusions: a university education does have a direct impact on someone’s political attitudes, but it is much less than you might think – and not always in the obvious direction.
For example, Ralph Scott utilised the 1970 British Cohort Study, which enabled him to track the attitudes of individuals throughout their life and revealed some evidence of a direct university effect. Going to university lowered levels of racial prejudice by a small amount, as well as making respondents less authoritarian. You might note that, being a study of the 1970 cohort, this is mostly people now in their 50s and who graduated more than 30 years ago, and you might, to misquote LP Hartley, believe that university then was a foreign country, where they did things differently.
But another paper, by Elizabeth Simon, using more recent data from between 1994 and 2020, finds something similar. Using household data, which allowed her to examine siblings – comparing those who go to university with those who don’t – she found that the apparent effect of a university education reduces by around 70 per cent once you control for family background. In other words, what looks like the effect of going to university is almost all explained by the type of people who choose to go.
There is still an effect, though. Going to university does result, for example, in having more liberal attitudes to gender roles. But once you control for pre-university selection, the scale of these effects are fairly small. How small? Put crudely, on a scale of zero to 100, Scott’s largest finding is that university moves someone’s views by about seven points, Simon’s by at most one point. These are the largest findings, and the latter is almost homeopathic levels of influence.
The kicker is that both these research papers – different time periods, different methods – also find that university does not always liberalise.
Simon found that the direct effect of a university education was to make people less environmentally friendly, and both pieces of research found that when it came to attitudes to the economy, university made people more right-wing.
Again, the differences are small to vanishing, but they are not quite what you’d expect from a Marxist madrassa. If my more left-wing colleagues really are proselytising, then they are not very good at it.
That’ll be 9,000 quid a year, please.
M Sobolewska and R Ford, Brexitland (2020); E Simon, Demystifying the link between higher education and liberal values: A within-sibship analysis of British individuals’ attitudes from 1994–2020, The British Journal of Sociology (2022); R Scott, Does university make you more liberal? Estimating the within-individual effects of higher education on political values, Electoral Studies (2022)
Get the inside track on what MPs and Peers are talking about. Sign up to The House's morning email for the latest insight and reaction from Parliamentarians, policy-makers and organisations.