Reducing crime starts with both school attendance and conflict resolution

There is a widely held belief that education helps to reduce youth crime by "keeping kids off the streets" (Machin and Sandi, 2025). Being at school provides young people with activities, guidance and support in the hours that might otherwise be filled with boredom, idleness or trouble. But what happens when schools are closed on an ordinary weekday - times when the rest of the world carries on as normal, and parents perhaps haven’t made provision for teenagers enjoying a day out of school?

Our new study offers one of the most detailed answers to date. Using term schedules for state-maintained secondary schools in England between 2010 and 2019, matched to police and juvenile court data, we examine the effects of short breaks from school on youth crime. While the results partly confirm the common view that education does keep kids off the streets, our findings suggest a more nuanced interpretation of the overall impact of school attendance.

Attending school prevents pupils from committing robberies but raises the likelihood of schoolyard brawls

Weekday school closures do indeed change youth crime rates, but the effect is not uniform. While serious violent offences and property crime increase, interpersonal violence decreases. It appears that attending school does not simply reduce juvenile crime overall but instead shifts the type and severity of offences that young people commit.

Schools as institutions of crime control

Secondary schools in England set their half-term breaks and teacher training days, commonly known as in-service training ("inset") days, independently. This creates what economists call a natural experiment: within one town, one secondary school could be closed on a particular date for an inset day, while a neighbouring secondary school remains in session on that day.

This creates variation in which young people from the school closed for an inset day represent the "treatment" group and those in the neighbouring school with no inset day are the "control" group in an experiment. Thanks to the random scheduling of inset days, we can observe changes in youth crime rates related specifically to our treatment group. In the following week, the pattern of school closures could be reversed, and we would observe changes in juvenile crime due to a new treatment group.

This variation allows us to build a school-by-day panel: we collected closure data for almost 3,300 schools over a 10-year period. While all school closures are scheduled in advance, we were particularly interested in half-term breaks and inset days, as parents are less likely to plan supervised activities for their adolescents during these shorter breaks. By looking at the same school on a weekday when it is closed compared to a weekday when it is open, we can isolate the crime-specific effect of school attendance.

Our dataset, which has more than 11 million school-day observations, allows us to distinguish between two common effects of school attendance. First, school has an incapacitation effect, as young people cannot commit crimes when they are physically attending school. But second, attendance could also exhibit a concentration effect, as putting a large group of teenagers together in one place can lead to frictions that could boil over into physical violence.

Juvenile property crime spikes during school closures

When school is closed on a weekday, the probability of a property crime involving the pupils at that school increases

When a school is closed on a weekday, the risk that pupils from that school will commit a property crime (compared to not committing such a crime) increases by 12%-13%. And the total number of property crimes committed by pupils at that school increases by 14%-23%, driven by indictable property crimes. Indictable property crimes, which include theft, fraud, criminal damage, arson and burglary, are considered more serious in nature and are tried in the crown court. Less severe "summary" offences include fare dodging and cycle theft.

Because the nature of these school closures means that most young people are in school and just a few are not, it might simply be easier for the police to clear crime on those particular dates. To ensure that our results reflect a genuine spike in property crime instead of increased crime clearance by the police, we analyse changes in the likelihood of a guilty verdict obtained in court. When school is not in session, juveniles from that school are 13%-14% more likely to be found guilty in court of committing a crime on that date, an effect commensurate in size with the probability of being involved in property crime.

This provides compelling evidence for school attendance having an important incapacitation effect on youth crime. By engaging adolescents in productive and structured activities, education acts as an effective large-scale crime-prevention mechanism.

The effect on violent crime is more nuanced

For violent crime, the school effect changes depending on the severity of the crime. We distinguish between indictable crimes and summary offences. Indictable violent offences include such crimes as robbery, rape, manslaughter and murder. Summary offences are considered less severe. They are heard and decided in a magistrates' court without a jury. Minor assault, threatening behaviour and intentional harassment are examples of summary violent offences.

On weekdays when schools are closed, the risk that a pupil at that school will commit a serious violent crime (compared to not committing such a crime) increases by 14% and the number of violent crimes rises by 20%-23%, an effect also reflected in court conviction data.

Remarkably, minor violent crime like minor assault, appears to decline on weekdays when school is not in session. The risk of summary violence drops by 19%-22% and the number of guilty convictions for summary offences falls by a similar margin. Much of this decline is due to "assaults against the person" - these are the types of fights, scuffles or minor altercations that are typical of school environments.

Put simply, attending school prevents pupils from committing robberies, but increases the likelihood of being involved in schoolyard brawls.

What explains this divergence?

Two mechanisms seem to explain these contrasting effects. Attending school has an incapacitation effect on serious crime: by attending classes, young people who might otherwise be on the streets during peak hours have fewer opportunities to commit serious violent or property crimes. But when school is in session, it has a concentration effect on minor crime: bringing a large group of adolescents together in sustained contact fuels micro-conflicts that can boil over into quarrels, bullying or minor assaults.

These dual effects underpin a subtle but important point: the impact of school on juvenile crime is not only a question of whether crime happens, but also about what kind of crime is prevented or displaced.

What are the key takeaways for policy?

Our research suggests some important implications for crime prevention and education policy.

First, schools are a crucial incapacitation mechanism. By providing a positive environment with structured activities, school is a powerful deterrent to property crime and serious violent offences.

Following the Covid-19 pandemic, chronic absenteeism continues to be a persistent issue for schools in England (Department for Education, 2025). Since these young people are at higher risk of committing or encountering crime, school personnel and education policymakers should reinforce the importance of regular attendance policies.

But there are important costs to the concentration effect. Adolescents packed together in a school environment can lead to frictions that result in bullying, quarrels or minor assault. These skirmishes could, in turn, affect the learning experience and wellbeing of other pupils in class. Increasingly, policymakers will need to consider providing support services within schools to support wellbeing and provide ways to resolve conflict and prevent bullying.

Staggering half-term breaks and inset days across neighbouring schools can reduce the risk of compounding opportunistic crime

We also find that scheduling matters. Policymakers in the local education authority should coordinate school closures within their jurisdiction. Ensuring half-term breaks and inset days at neighbouring schools are not scheduled simultaneously can reduce the risk of compounding opportunistic crime, though some overlap in school closures is likely in local authorities with many secondary schools, particularly in big cities like Birmingham and London.

Finally, the same dynamics at work in schools will also apply to other youth services such as summer camps, sports facilities or recreation centres. There is clearly scope for schools to develop coordinated programmes with such institutions to run enrichment programmes and recreational activities during these breaks. Designing these facilities to maximise the incapacitation effect and minimise the frictions associated with the concentration effect will require policymakers to work closely with schools and youth support services.

---

20 February 2026     Paper Number CEPCP726

Download PDF - The complex relationship between education and criminal activity

Download Press Release

This CentrePiece article is published under the centre's Crime programme, Education programme.