Youth custody: Educational influences and labour market consequences

This project will provide quantitative evidence on which young people are imprisoned, the school experiences that increase the risk of imprisonment and the impact prison has on subsequent labour market outcomes. The youth custody population is disproportionately made up of the most disadvantaged young people. Prison is likely to compound disadvantage, making the transition to adulthood challenging. Between 2008 and 2017, the number of under-18s in custody in England fell by 70%, but this has not been equally distributed among young people: for example, minority ethnic representation among young offenders increased from 25% to 41%. Many young people in custody show extreme disadvantage in literacy, numeracy, school attendance, exclusions, care and mental health. Furthermore, once young people come into contact with the justice system they are likely to do so again and, in the longer term, struggle in the labour market.

This project will set out the key pathways into and out of custody, estimate how aspects of schooling (participation, peers, qualifications and exclusion) affect the probability of being imprisoned and, beyond school, estimate impact of prison on subsequent labour market outcomes. In so doing, it will inform policy aimed at helping at-risk young people achieve positive transitions.

The research will use the Longitudinal Educational Outcomes (LEO) dataset, which links education and tax records to observe nearly all young people in England from school to early adulthood. It provides information on background characteristics (e.g. gender, ethnic group, location, disadvantaged status), school experience (e.g. attainment, absences and exclusions, PRU/AP referral, special educational needs) and detailed month-by-month activity status post-16, including months where an individual is in custody. The size of the LEO dataset makes it possible to observe subgroups within the population that are too small to consider using survey data.

The first strand of analysis will quantify the extent of custody and, for young people who experience it, how it fits within their broader experiences. Cluster analysis will be used to identify a typology of transitions, capturing the full detail of individuals’ experiences over time. Econometric choice models will be used to look at the extent to which characteristics and circumstances predict group membership, creating useful insights for policymakers interested in early identification of at-risk young people.

The second strand will investigate the impacts school experiences have on the probability of youth custody. This has four elements:

• Participation: estimate the effect of the September 2013 raising of the participation age by comparing outcomes of cohorts completing year 11 in 2012, 2013 and 2014.
• Peers: through a peers-of-peers instrumental variables (IV) strategy, estimate the effect of school year composition.
• Qualifications: IV regression to examine the influence of academic qualifications, comparing pupils who narrowly achieve a ‘good grade’ in core subjects at GCSE to those who just fall short of the threshold.
• Exclusion: IV regression to estimate the impact of exclusion, and how this varies according to the nature of exclusion, the placement type for permanently excluded pupils, and the timing of permanent exclusion.

The third strand of analysis will examine the impacts of prison on subsequent post-16 outcomes, with a focus on labour market outcomes. Using a multi-state model, the research will capture the effects not just of whether an individual experiences custody, but also the number and length of custodial spells.

The findings from this research will provide an evidence base for the development of effective policy and practice to help at-risk young people. As the research addresses custody, schools policy and practice and labour market outcomes, it will be of interest to the Ministry of Justice, the Youth Justice Board, the DfE and DWP. Discussion papers and briefing papers will be produced for each strand of the research, and all findings will be synthesised with new insights in the main public report.