Suntory and Toyota International Centres for Economics and Related Disciplines (STICERD) LSE RSS Contact Us YouTube Twitter

A recent report by HEFCE, following 130,000 students and controlling for background, finds that state school graduates outperform private school graduates at university. Still, while private school graduates are only 7% of the UK graduate population, they constitute 71% of senior judges, 62% of senior armed forces officers, 55% of top civil servants, and 43% of newspaper columnists. Moreover, private school male graduates are up to 10 per cent more likely to land top jobs than state school graduates with the same grades from the same university [1]. It seems that private school kids get all the good jobs but the few state kids that make it do much better. This means that state school graduates must face a higher bar. This cannot be an efficient way to allocate talent in the economy.

Our recent research, 'The Coevolution of Segregation, Polarised Beliefs and Discrimination: The Case of Private vs. State Education', is set to provide an explanation to the above. Specifically, we tie this misallocation of talent to the phenomena of segregation and the prevalence of stereotypes. School choices in the UK are persistent and show high levels of segregation. Evans and Tilley (2011) find that 43 per cent of the privately educated who have children have sent them to private schools, nearly five times the rate for parents who went to state schools. Among married individuals, 41 per cent of the privately educated are married to a privately-educated person, compared with six per cent of those who are state educated. Of those who are married with children, 65 per cent of couples who both went to a private school have sent a child to a fee-paying school, compared with only six per cent of state school couples.

This persistence of segregation in schools also translates to polarised beliefs in many dimensions, such as systematic differences in political preferences and social attitudes among graduates of different schools [2]. In a 2013 YouGov survey, in response to: "From what you know, do state secondary schools generally give talented children a good education and allow them to achieve their full potential?", 58 per cent of state school parents replied yes compared to only 48 per cent of private school parents. 21 per cent strongly agree to: "Pupils at private schools are generally brighter than pupils at state schools", with responses exhibiting big differences across different political affiliations and socioeconomic status. These different attitudes have substantial effects on the UK labour market as we saw above.

Negative stereotypes about different education paths on one hand, and segregation in schools, may fuel each other. This can happen in many other social contexts beyond schooling choice, for example in the cases of gender, race, religion etc. It is a bit of the "chicken or the egg"; stereotypes and biased beliefs affect where we live and who we interact with and in turn, echo-chamber effects imply that who we talk to affects our beliefs and attitudes.

In our study financed by the ERC, we set to better understand the co-evolution of segregation and negative and possibly wrong attitudes towards others, and its economic impact. Our focus is on the beliefs about the labour market productivity of graduates from state schools (or more generally, about the difference between state and private school graduates). We analyse a dynamic theoretical model of school choice, socialisation in schools and the labour market. We characterise the conditions under which persistent segregation and polarised beliefs arise, leading to the prevalence of possibly wrong stereotypes and labour market discrimination.

We consider a model in which at every period, individuals attend school and their beliefs (about productivity outcomes of different schools) are shaped by their parents' beliefs and by their school peers. Individuals then enter the labour market, some as employers and some as employees. Employers decide whether to hire an assigned employee, given their beliefs which may depend on the employee's school. Labour market experience entails some learning about the true productivity. Finally, individuals become parents and transmit their beliefs to their offspring. Parents also decide on their offspring's education path, state or private, given their own beliefs and the schooling decisions of other parents. We assume that this process is repeated ad infinitum and track the long-term beliefs and behaviour of individuals in society.

We show in what environments segregation in schools and discrimination in the labour market persist in the long term. In that case, wrong beliefs can persist through segregation in education, while segregation in education, through parental school choice, is in itself motivated by wrong beliefs and labour market discrimination. In these cases, there is persistent discrimination in the labour market against state school graduates (over and above actual productivity differences). Parents who send their children to a private school believe that the difference between the schools (in terms of the productivity of its graduates) is higher than what it really is. Parents who send their children to a state school realize that there is discrimination, believe it is not justified, but are priced out of the private school. Finally, those who went to the private (state) school will also send their children to the private (state) school. Thus, the "old boys" network is endogenously formed.

We show that history matters; to sustain discrimination, those in the private school have to start from a relatively low opinion about state school graduates. We also show that the higher is the intensity of socialization in schools the easier it is to sustain discrimination. This arises because the echo chamber effect becomes more dominant. Finally, discrimination is easier to sustain the less individuals learn about others from their labour market experience.

The model provides a new framework to think about policy to limit discrimination in the labour market. Specifically, we provide another rationale for school integration beyond those provided in the literature (that focuses mainly on academic achievement): in our framework successful integration will change wrong stereotypes and the beliefs of individuals on one another, inducing less discrimination in the long-run.

The model also sheds light on some common practices in school admissions. As we show, when private schools admit pupils only according to their willingness to pay, such schools will not necessarily be able to survive in the long-run. Alternative selection criteria, based on beliefs, values or culture, will allow such schools to increase their long term survival. This accords with a common practice of many private schools and universities in the UK and in the US which hold interviews, often with parents as well, or have a legacy criterion for admission.

One should bear in mind though that the phenomenon we describe below is not restricted to the UK or to secondary education. The US schooling and university markets share similar features to that of the UK. Other prominent examples are segregation via religious or ethnic affiliations, which can both be affected by negative beliefs on others, as well as affect negative beliefs on others. This is the focus of our current and future research.


[1] These facts are taken from a recent study by the Social Mobility and Poverty Commission which tracked 20,000 students.

[2] Also in Evans and Tilley (2011).