Why School Choice is Bad for Social Mobility

I am quite a good example of  social mobility. My parents only completed obligatory school – that was four years in rural Portugal in the 1950s – and started working at 12. They moved to France and then Switzerland in the 1970s, and worked their whole lives in unskilled jobs. In spite of that, I went to university, obtained a doctorate and have now a permanent academic position in a highly-ranked university.

However, this upward mobility was only made possible by certain conditions. I grew up in Switzerland, where the quality of education was probably much better than what it would have been if my parents had stayed in Portugal. I grew up in a relatively wealthy village with a library just down the street, that I would visit at least twice a week to refill the comics and books that I didn’t have at home. Most importantly, I grew up in a country where everybody goes to state schools, and where there is no choice in selecting your school. Social reproduction is obviously present, but there is no stigma attached to state schools, as is often the case in Britain. Actually, it may be the contrary. The only people I know who would go to private schools would typically be children of wealthy parents who didn’t make it in the state system. If their grades weren’t good enough to go to the higher tier of the state system, their parents would place them in a private school with classes with only 5 pupils and tutorials to hold their hand until university. Same for university: I just went to my local university without even thinking about going elsewhere. All Swiss universities are pretty good. Even if you don’t go to university, which is the case of about 70% of each age cohort, you can still make a very good life thanks to vocational education.

I have friends in the UK who have children, and I am amazed by how complicated it is early on to select the right school for your children. If you don’t put them in the right primary school, they then might not get into the right secondary school, and then not get good enough A-levels and then they won’t go to the right university, and then they’ll basically fail their life. Even if you’re a good student, you won’t be drilled in the same way to go to a good university in a bad state school, and people seem to rely to a much larger extent on the shortcuts provided by school or university rankings.

Sometimes I wonder what would have happened if my parents had moved to the UK instead of Switzerland, without knowing which schools are best, without being able to afford to send me to a private school or live in the catchment area of a good state school, without writing English well enough to lobby for me to go to a better school, besides not being able to help me with my homework. I probably wouldn’t have done as well because my parents wouldn’t have been able to make the right choices for me, or wouldn’t have had the money to make these choices. This is not because my parents didn’t want the best for me, but their conception of hard work has more to do with getting up early and working hard, as they have done their whole life, rather than educational achievement that they haven’t experienced themselves. In the UK 42.5 per cent of all children whose mothers were highly educated but not British were taught in disadvantaged schools. With the children of poorly educated mothers, the figure rises to 80 per cent – the worst record in the OECD. Hence, being educated but an immigrant doesn’t even offset the segregation created by school choice.

School choice in the UK places so much of a burden on parents that it is naturally bound to reproduce social inequalities. Even if the money part is left aside, it assumes that all parents have the same cultural capital to make perfectly informed education choices for their children and help them “fit in” in the middle class environment of quality schools. Hence, a government advisor – who “was raised in a £150,000 semi-detached house (…), went to Aylesbury Grammar School and then on to Cambridge University” apparently declared that working-class children

need to become more comfortable with middle-class social setting such as restaurants, theatres and offices if they are to succeed.

Working class families are urged to watch plays, visit museums and try middle class food, restaurants and shops to broaden their child’s life experience.

This is simply amazing: as if a middle class habitus could be acquired with a few trips to the national gallery. Even assuming that children need to conform to middle-class values to perform better, working-class children are very unlikely to acquire these values if all rich kids stay amongst themselves in private schools. Now it’s no wonder that Britain has one of the lowest levels of social mobility in the OECD. For instance, in no other OECD country is the income of your parents so correlated to your own income. Only 3% of the richest 20% have mothers with no qualifications. The social segregation allowed by school choice is also one of the reasons why this huge country is run by such a small group of upper-class people with so similar backgrounds, via Eton, because that is where social networks are built, and working-class kids are excluded from them. Now the government says that it wants to rewards people who “work hard and get on”, but people who do – like my parents – without education credentials have seen their real wages fall and the major parties abandon them.

This may be related precisely to so much school choice that upper class parents can more aptly exploit, either by placing their children in private schools or knowing better how to place them inthe good state schools, while children of working class parents are left amongst themselves in poor quality state schools. A strong quality public education sector seems to be a much better way to counter these tendencies, while increasing school choice seems bound to perpetuate them.


Ending school segregation is the key to social mobility

UK school the “most socially segregated in the world”

Do good universities teach better, or do they just select better students?

Research commissioned by the  Department for Business, Innovation and Skills (and reported in the Times Higher Education Supplement) argues that the wage premium associated with studying at a Russell Group university is not statistically significant if you control for the social background of students and their A-levels. Graduates of Russell Group universities do earn 36% more than non-graduates and 15% more than graduates of former polytechnics. However, differences between Russell Group and non-Russell Group graduates vanish if you control for the profession of their parents and their grades in high school. Put more simply, if you are clever and come from an upper middle-class background, it doesn’t really make a difference if you go to a top or an average university. Now of course, if you are clever and upper-middle class, you are more likely to go to a better-ranked university (and top universities are more likely to want you), but going to a top university per se doesn’t seem to provide any added value. The difference is only due to a selection effect. Going to a good university is correlated with being smart (and having higher incomes throughout one’s lifetime), but it doesn’t make you smarter. Yes, let’s assume that income is an indicator of ability, just for a moment.

Somehow, this echoes old research by Bourdieu and Passeron on the role of education in society. In The Inheritors, they argued that the education system is more an instrument of selection and classification than an instrument of well, education. The educational system is better at legitimizing and reinforcing inequalities derived from socialisation than at compensating for these inequalities, or simply at teaching people things. Students from upper-middle-class backrounds have better grades on average not because they are more clever, but because school positively evaluates abilities that are more prevalent in upper-middle classes, and that are not taught at school. Similarly, employers prefer to hire graduates of higher ranked universities because higher ranked universities have selected the better students, but not necessarily because they provide better education. A degree from a good university is just a way to certify abilities that are acquired outside of university. While marking essays, I have often wondered about this: am I assessing if the student has learnt anything from the course that i have taught, or her general ability to structure an argument which has been acquired out of university? If you take this logic further, you could think that the material that is taught at university (and that students will mostly forget after exam time) is just a pretext for marking and ranking students, because you need to evaluate them on something after all. In contrast to the type of university, grades make quite a big difference for graduate earnings, however.

Now, what is interesting with the results of this study is that even if the education provided doesn’t matter, you would still expect Russel Group Universities to provide a wage premium (independently of the education provided) compared to former polytechnics because of these signalling effects. Even with equal abilities, equal social backgrounds and equal A-level results, graduates of Russell Group universities should still earn more because of the prestige of the brand, but as far as I understand, they don’t. Hence, there is something that offsets this “brand premium”.

There are different possible causes. One possible reason is that the education received at Russell group universities is actually worse than that provided at former polytechnics. It sounds counterintuitive, but there are actually a number of reasons why this could be the case. Research-intensive universities are, well, more focused on research while former polys are more focused on teaching. Time is not extensible, and attracting research money, getting grants and publishing a lot means less time to teach. Research is considered much more important than teaching for prestige and tenure. It is not uncommon for high-flying professors at prestigious universities, who may attract lots of money and prestige which count for rankings, to not teach at all, or very little. A great bulk of the teaching is carried out by graduate students on casual employment contracts, or in the US by a lumpenproletariat of adjunct faculty. In the UK, a third of students considered that their courses were bad value for money, considering what they pay and the actual contact hours they spend with their teachers. Of course, one may argue that universities who are good at research should provide also better teaching because of spillover effects, but this should be reflected in the results, and it isn’t. If great researchers do not teach, there is no reason why these spillover effects should take place. WBAEPHR6SFGY

Why is Portugal lagging behind all other European countries in educational attainment?

Average Years of Total Schooling in 27 countries, 1950-2010, Population 25 and over (Source: Barro and Lee)

I don’t have the answer. I have just been playing with Barro and Lee’s dataset on education attainment. As the name indicates, it covers education attainment in 146 countries between 1950 and 2010 for different age groups. The two charts above show the average years of schooling of the population 25 and over since 1950 in 27 countries. I know that the first graph  is difficult to read, but the point is to show Portugal’s outlier status in comparison with all other European countries. With 7.73 years of schooling on average, the Portuguese workforce was the least qualified in the EU in 2010; the average for the countries showed in the graph is 10.62. Portugal started below everybody else in 1950 (compulsory schooling under Salazar was 4 years, and many were simply not enrolled) but I cannot really understand why it is still lagging behind Spain, which started from a similar position, namely a long fascist dictatorship keen on maintaining a docile and uneducated populace.


The three graphs below show the population (25 and over) between 1950 and 2010 in Portugal, Spain, and Sweden (as a point of comparison for “advanced” countries) by highest degree attained (but not necessarily completed). Even if a higher proportion of the Spanish population had no schooling at all in 1950, this share has been reduced dramatically while it has remained at high levels in Portugal. According to the data, 11.5% of the Portuguese population had no schooling at all in 2010, and 46% had gone through primary schooling only. The corresponding numbers are 1.6 and 25% in Spain, and 1.6 and 11% in Sweden. This happens in spite of the fact that Portugal has a fairly low student-teacher ratio and has invested massively in education over the past few decades. However, as the number of PhDs has exploded, the problem of low skills at the bottom is still huge. As we know, low productivity is a big problem in Southern Europe, and this is probably strongly related to low education levels. Cuts in spending certainly won’t help to solve this problem.