Children under 15 who commit serious offences
This report focuses on offences committed by children under the age of 15. There is a substantial need for knowledge in this area, not least in light of indications that younger children’s involvement in serious crime has increased in recent years.
Related content
About the publication
- Author
- Other information
- © Brottsförebyggande rådet 2025
- urn:nbn:se:bra-1267
- Report 2025:11
Summary
Number of suspected offences has increased, but the proportion subject to LUL investigations is stable
The study shows that the number of suspected offences involving children under the age of 15 has doubled over the past decade. The increase is mainly explained by a doubling of the number of suspected cases of assault, theft, unlawful threat and (non-sexual) molestation. Boys accounted for around three-quarters of all such offences during most of the period studied, but the proportion of girls has increased slightly since 2021.
The number of suspected offences subject to an LUL investigation has also increased, from just under 1,500 in 2015 to just over 3,500 in 2023. This is probably due to the increase in the number of suspected offences involving children under the age of 15. However, the proportion of suspected offences investigated has not increased. Of all such offences involving children under the age of 15, approximately 10 per cent have been subject to an LUL investigation over the entire period. Boys accounted for just over 80 per cent of the suspected offences that were subject to an LUL investigation throughout the period. Just under 60 per cent of the suspected offences investigated concerned 14-year-olds, and just under 30 per cent concerned 13-year-olds. The share of suspected offences accounted for by these age groups has not changed significantly during the period examined.
Longer investigation times and regional differences
The most common reason for initiating an LUL investigation is that the offence carries a minimum sentence of one year in prison (40 per cent of all suspected offences investigated), and the proportion of investigations initiated for this reason has increased slightly during the period 2014–2023. Only one tenth of LUL investigations were initiated at the request of the social services, and this proportion has not changed significantly during the period examined.
During the same period, LUL investigations have taken an increasingly long time, from a median investigation time of 47 days to one of 110 days. One possible explanation may be an increase in the proportion of offences that are of types that tend to be more difficult to investigate.
The analysis shows clear regional differences in the application of the LUL legislation. There are differences between the regions in the proportion of suspected offences that are subject to LUL investigations and in the grounds for initiating an investigation. For example, the proportion of LUL investigations requested by the social services varies considerably between regions. Investigation times also vary between regions but have become longer in all regions.
The crimes most commonly subject to LUL investigations are violent offences and sexual offences among individuals in the same age group
During the years 2022–2023, assault was the offence most commonly subject to LUL investigation for both boys and girls. For boys, this was followed by offences involving robbery and rape, while for girls, theft and drug-related offences were the second and third most common offence types. The individuals who had committed the offences subject to LUL investigations had most often committed them with acquaintances or friends of the same sex and of the same age or slightly older. The victims were also usually of the same sex and age, and in the majority of cases, the suspects and victims already knew each other. Many of the violent offences, and offences such as robbery and serious threats, had been motivated by conflicts or perceived insults between children. In the majority of LUL investigations concerning rape, the suspect was a lone boy who was usually acquainted with the victim (almost 9 of 10 victims were girls). In about a quarter of rape cases, the victim reported that violence had been used, and in just over one in ten cases, the victim had been in a particularly vulnerable situation.
Brå’s analysis shows that many of the other crimes investigated under the LUL Act, despite being classified as serious offences such as arson or child pornography, had been committed out of recklessness or ignorance of what constitutes a crime or what the consequences may be. However, 30 per cent of the crimes alleged to have been committed by boys and 12 per cent of those alleged to have been committed by girls were assessed to have been particularly serious, based on the degree of violence used or the seriousness of the planned violence, the serious of injuries to the victim, the use of weapons, and the degree of ruthlessness shown towards victims who were unable to defend themselves. This category includes, for example, murders and attempted murders, which are very rare in the material, and serious cases of assault, robbery, rape, and weapons offences.
The children’s life situation is characterised by a concentration of social and psychiatric problems
The results of the study show that the children subject to LUL investigations typically grow up in disadvantaged socio-economic conditions compared with the population as a whole in the same age group. Compared with the population aged 0–14 as a whole, the children subject to LUL investigations were twice as likely to live with a single parent, twice as likely to live in rented housing, and twice as likely to live in an area facing socio-economic challenges. These children also more often lived in households that were in receipt of financial assistance and more often had two foreign-born parents.
The study shows further that the children subject to LUL investigations were also at greater risk of being victims of crime themselves. One-third of the boys and half the girls had themselves been victims of crime in the year prior to committing the offences that resulted in the LUL investigation, which is seven and ten times more common than among children of the same age in the general population. Furthermore, almost half of the boys and half of the girls had been previously diagnosed with a psychiatric disorder. Compared with the population of the same age, a greater number of the children subject to LUL investigations had been placed in out-of-home care prior to committing the LUL offence. Almost one-fifth of the boys and one-tenth of the girls had been subject to a LUL investigation prior to the current LUL offence.
An analysis of social services documentation (child welfare files) showed that the children subject to LUL investigations also have a number of other established risk factors for early-onset delinquency. Almost half of the girls had lived with a mother who lacked sufficient parenting skills, and one-third with a father who lacked sufficient parenting skills, at the time of the investigated crime. Among boys, the corresponding figures were one-third and one-fifth. However, it was more common for the child to live with the mother and for the father to be absent from the child’s life, which may explain why mothers were more often assessed to lack sufficient parenting skills – the fathers were not present for the assessment.
The social services documentation also showed that many of the children subject to LUL investigations had criminal associates and extensive school absenteeism, as well as recurring or serious conflicts with other pupils or serious conflicts with teachers. According to information from the social services, just over a quarter of the children had received special-needs teaching.
One-third of the boys and almost half of the girls had been assessed by the social services as having a worryingly high level of aggression, and just under one-third of the children had impulse control issues. A quarter of the boys and almost one-fifth of the girls had been assessed to have difficulties concentrating and paying attention.
The children were already known to the social services
Nearly nine out of ten children were already known to social services in their current municipality before committing the offence that led to the LUL investigation, and for approximately half of both girls and boys a notification of concern had been filed with the social services before the age of 12. It was mainly schools and the police who had reported concerns prior to the analysed crime, and the reasons behind these reports were mainly that the children had already exhibited deviant behaviour. This was the case for half of the boys and two-thirds of the girls. For half of both boys and girls, there was at least one previous notification of concern relating to a previous offence (which had not necessarily been the subject of an LUL investigation). For about half of both boys and girls, the previous notifications of concern also related to parent-related risk factors.
A majority of the children had previously been the subject of interventions
Two-thirds of the boys and four-fifths of the girls had been the subject of at least one intervention or measure from the social services prior to the LUL offence. The most common interventions prior to the LUL offence were some form of family counselling, family therapy, or individual counselling focused exclusively on the guardians.
Subsequent to the offences that had resulted in LUL investigations, the social services decided to implement measures in relation to the same proportion of boys as had been the case prior to the offence, but for only just over half of the girls. In addition, a small proportion of both boys and girls already had an ongoing intervention. Even subsequent to the LUL investigation, the most common intervention was family counselling or family therapy, and the second most common intervention for both boys and girls was a placement decision under the Care of Young Persons (Special Provisions) Act (known in Sweden as the LVU Act).
One quarter of the boys assessed to have been involved in a criminal network
Of the children included in Brå’s sample of children subject to an LUL investigation in 2023, just over one quarter of the boys and one in twenty of the girls were assessed to be involved in a criminal network.
Boys involved in a criminal network are often investigated for aggravated assault and robbery
The material includes too small a number of girls involved in criminal networks for analysis, but the analysis of boys showed that the offences resulting in an LUL investigation most often related to assault (often aggravated), robbery, weapons offences, or obstructing the course of justice. All of these offences were more common among the boys assessed to be involved in a criminal network than among those with no such involvement. The analyses also showed that the suspected offences among boys involved in a criminal network were more likely to be characterised by more serious violence and the presence of weapons, compared with the offences among boys not involved in a criminal network. It was also more common for boys involved in a criminal network to have committed crimes together with people who were at least three years older than themselves.
Consistently poorer socio-economic situation than other boys investigated for serious offences
The boys involved in a criminal network had grown up in a worse socio-economic situation than the other boys subject to LUL investigations. Network-affiliated boys were almost twice as likely to have grown up in rented housing and to have grown up in areas facing socio-economic challenges. They also more often came from families that had been in receipt of financial assistance. They were almost three times as likely to have been born in Sweden to foreign-born parents as the other boys who had been the subject of LUL investigations.
A more pronounced risk profile even in other areas of life
The boys involved in a criminal network were about as likely to have a psychiatric diagnosis as other boys subjected to an LUL investigation (usually hyperactivity disorders), but they tended to have come to the attention of the social services somewhat later. One distinctive feature of the risk profile of network-affiliated boys was that they more often had criminal associates (particularly older individuals, which is probably related to their involvement in a criminal network), but they also had a higher level of risk factors in the areas of emotions and behaviour, school, and leisure time factors. Compared with other boys subjected to LUL investigations, boys who were involved in a criminal network were four times as likely to have been assessed by the social services as having drug problems and more than twice as likely to have been assessed as having had conflicts with teachers.
Boys involved in a criminal network had already been the subject of more extensive interventions and measures prior to the LUL offence
The interventions and measures that had been decided upon prior to the LUL offence indicate that the social services perceived the boys involved in criminal networks to be living with a complex constellation of difficulties. Compared with other boys subjected to LUL investigations, it was more common for boys involved in a criminal network to have been the subject of decisions involving coordinated measures from several different actors (coordinated individual plan (Samordnad Individuell Plan, SIP), social intervention groups (Social InsatsGrupper, SIG), and the school, social services, police, and leisure collaboration (Skola, Socialtjänst, Polis och Fritid, SSPF)) even prior to the LUL offence. It was also more common for them to have been offered substance abuse-related support and to have been subject to LVU placements prior to the LUL offence. They had also received interventions in the form of non-institutional programmes and contact persons, contact families, and specially qualified contact persons to a greater extent than other boys investigated under the LUL Act.
Compulsory placement is the most common measure following an LUL offence
The single most common measure taken for boys involved in a criminal network subsequent to an LUL offence was a placement under the Care of Young Persons (Special Provisions) Act (LVU), which was three times more common than among the other boys subjected to LUL investigations. Other relatively common measures among boys involved in a criminal network were family counselling or family therapy, individual counselling, and other types of non-institutional programmes. Being subject to these measures was about as common as among boys who were not involved in a criminal network.
Interviewed children’s experiences of social services support and encounters with the justice system
Of the 17 young people interviewed, almost all had experience of involvement in a criminal network. A majority described their early experiences of support from the social services as negative. This was described as in many ways having been linked to their own low motivation to change their criminal behaviour at an early stage of their involvement in offending. Their stories included receiving several interventions and measures over the years that they felt had not helped them. Several said that interventions that had been presented as being voluntary felt like compulsory measures, since decisions were made under the threat of extended interventions or an LVU placement. When interventions had been received during periods of low motivation, several said that they had the opposite of the desired effect. Several described perceiving interventions that they had received at a later stage of their criminal development as more positive, which was related to the fact that they were then more motivated to desist from their criminal activity, and they also understood the purpose of the support measures better as they became a little older.
When asked to describe the support measures they had lacked, most of the young people mentioned that they needed stable and relatively long-term relationships with adult support persons who really cared about them and understood their situation. They also wished they could have talked to an adult support person about their experiences and challenges, without this necessarily resulting in new notifications of concern and social services interventions.
Success factors and challenges according to the social services, the police, and schools
Through interviews and a workshop, representatives of the social services, police, and schools, and also other key agencies working with children who engage in crime, have highlighted a number of challenges and success factors that they experience in their work with these children. One challenge is the need for early, high-intensity interventions before the children become too deeply involved in crime. They stated that this can sometimes be difficult as a result of several factors: it is difficult to know at an early stage which children need interventions, or what type of interventions, guardians do not always recognise problems and are not always open to receiving support, guardians do not always have the capacity to embrace the type of support offered, and sometimes the interventions that are deemed necessary are not available or possible. This may, for example, involve the need for practical support for guardians or interventions that are supported by the evidence base for children with more severe problems. They also believe that interventions need to involve the child and adolescent psychiatric sector to a greater extent in order to meet the needs of children with neuropsychiatric diagnoses.
One of the success factors highlighted in the interview study and workshop was the use of structured risk and needs assessments in preliminary assessments or assessments conducted under the Social Services Act, since among other things, this would lead to more consistent assessments. Most of the interviewees from the social services believe that LUL investigations can be a success factor in motivating guardians to accept interventions, but if such motivation is not considered necessary, they find it difficult to see the benefit of LUL investigations. Interviewees from the social services also pointed out that it would be desirable to be able to vary interventions to a greater extent in order to adapt them to the individual circumstances of the child or guardian(s), to be more flexible in deciding on longer-term interventions, and to improve access to the interventions that the National Board of Health and Welfare recommends that the social services use.
Interviewees from the social services, police, and schools felt that an effective and structured cooperation between the actors who have a central role to play in relation to children with criminal behavioural problems constitutes a clear success factor, even though both evaluations and individual follow-ups need to be developed to a greater extent. All interviewees said that the child and adolescent psychiatric sector was the actor whose involvement was most sorely missed with regard to children who commit serious offences. The presence of this sector in collaborative measures is important as a result of the high prevalence of psychiatric diagnoses among the children in question. The interviewees felt that the social services and schools in particular face an overwhelming task when these children do not receive the psychiatric or psychological support they need to arrest their deviant development. They argued that the child and psychiatric sector needs to be given the resources they require to meet the needs of these children.
Brå’s assessment
Based on the report’s findings and previous research in the field, Brå highlights key areas for development in the work to prevent serious crime among children under the age of 15. The study fills certain gaps in knowledge regarding this target group, but at the same time confirms existing knowledge about risk factors among children and challenges and success factors in preventive work.
The fact that many of these children have risk factors in many areas of life means that efforts are required from several actors at the universal, selective, and indicated levels. In interviews, social workers described how they are burdened with an unreasonable amount of responsibility. Brå emphasises the importance of a broad cooperation between actors such as the social services, schools, preschools, and healthcare. Crime prevention work also needs to address structural factors such as economic exclusion and socially disadvantaged areas. Effective interventions should be based on existing cooperation structures and the strengthening of the capacity of the welfare system.
Early detection, early intervention, and better assessments
The early identification of risk factors is crucial for early intervention, and Brå’s study shows that the social services were already aware of 9 out of 10 of the children subject to LUL investigations prior to their commission of the serious offences that resulted in these investigations. However, what constitutes early detection and early intervention depends on the individual child’s risk profile. Almost half of the children studied had been referred to the social services before the age of 12, primarily as a result of challenges experienced by their guardians (and they were reported for their own deviant behaviour only later), while the other half of the children were referred to the social services for the first time at the age of 12 or later, primarily as a result of their own deviant behaviour (without previous referrals to the social services). With this in mind, Brå emphasises the importance of children’s health centres and preschools paying greater attention to challenges that characterise the family situation of many children in order to be able to help them when necessary.
In addition, it appears that the social services only use structured risk and needs assessment methods to a very limited extent in connection with preliminary assessments and child welfare investigations. The more widespread application of such methods may be necessary to achieve evidence-based and consistent practices across the country’s municipalities, and also to increase the chances of adequate interventions being implemented earlier and with greater effect.
Interventions that better match the receptivity of children and guardians, and increased availability of evidence-based interventions
Brå notes that the interventions most commonly used by the social services are individual counselling or family therapy, while the child’s problems often involve difficulties in several areas of life. Based on the overall knowledge base, there may be a need for both greater flexibility, with individual social workers being given the scope to use their own professional judgement to decide on interventions based on the individual needs of the child and family, and the increased use of the manual-based methods that have been shown to be effective for this group of children and which are recommended by the National Board of Health and Welfare. These methods should be available regardless of which municipality a child lives in, something that currently varies greatly.
The children interviewed described a need for unconditional support and supportive relationships with people they can confide in without triggering the duty to report them to the social services. Brå’s assessment is that there is a need to review what opportunities exist to meet this need. However, it is important to find a balance between the (sometimes therapeutic) needs of children to be able to talk to someone and society’s responsibility to protect them, which is tied to the duty to report. Several professionals also raised the need for additional ways of offering more long-term support to children with multiple risk factors. One such option involves the use of mentors or contact persons, but evaluations of how different forms of mentoring are currently being used for children under the age of 15 may be needed.
The receptiveness of guardians affects both whether social services’ interventions are accepted and the effect of these interventions. Many may have a distrust of the social services, their own mental health problems, poor language skills, or a limited ability to participate in interventions during office hours. The study shows that there is a need for greater flexibility in order to meet the varying receptiveness of guardians. For example, there may be a need for more practical forms of assistance for guardians based on the child’s needs.
School is a key protective factor that needs to be strengthened
Brå’s study confirms findings from previous research which show that various forms of school failure are common among children who engage in crime. These include extensive absenteeism, poor grades, and conflicts at school. Schools play an important role both as actors in the early identification of children at risk of developing deviant behaviour and as a platform for preventive interventions. Brå emphasises the importance of strengthening schools’ ability to take early action by providing support to children who are experiencing challenges at school and collaborating with other actors, for example through school social welfare teams and professional collaboration models that include schools.
Developed operational collaboration
The study shows that the social services continue to request LUL investigations to a limited extent. Interviews revealed that these investigations are particularly useful to the social services when guardians are unwilling to recognise their own child’s need for support in relation to criminal behaviour, and that the investigations often take longer to complete than the time that the social services have available to decide on interventions. At the same time, those who work in close cooperations between the social services and the police believe that this cooperation often enables the social services to receive the information they need from the police even before LUL investigations are complete, and that such collaborations serve to improve the quality of the police’s notifications of concern. To increase the efficiency of LUL investigations and improve the information sharing between the social services and the police that is a prerequisite for this, it is Brå’s view that close cooperations between the police and the social services should be developed where they do not already exist. In this regard, interviewees spoke of positive experiences of co-locating representatives of the social services and the police.
Based on the multi-problem profile exhibited by many children in the study, Brå further emphasises the importance of continuing to develop operational collaborations that involve all the key actors from which children need assistance in the country’s municipalities, for example, through SSPF and SIG.
A strengthened role for the child and adolescent psychiatric sector
The study has shown that many of the children subject to LUL investigations have psychiatric diagnoses and that, based on the findings of previous studies, even more are likely to receive such diagnoses at a later date. Regardless of whether a child’s challenges manifest themselves at a very young age or closer to adolescence, the study’s results indicate that the the child and adolescent psychiatric sector is not currently dimensioned to address today’s complex problems. The social services, schools, and the police, who encounter these children and their care needs on a daily basis, all call for a stronger role for the child and adolescent psychiatric sector in terms of both cooperation and concrete measures.
A gender equality perspective needs to be integrated into crime prevention work involving children who engage in crime
The LUL investigations show that children’s attitudes towards ideals of masculinity are clearly linked to criminal behaviour. This applies to many of the offences in the data set, for example both sexual offences and violent offences, but is most evident in the crimes committed by children involved in criminal networks. Brå’s view is therefore that it is important to continue to strengthen and develop universal interventions targeting destructive masculinity ideals. A gender equality perspective needs to be integrated into the work with children who engage in crime – not as an add-on, but as an integral part of the development of prevention methods. Programmes and education for both children and professionals should also generally include reflections on gender norms, and these interventions also need to be evaluated.
Improving the quality of care during and after placements under the LVU Act
The study clearly shows that the children who become involved in criminal networks have complex support needs and that society’s interventions are not tailored to this target group. When previous interventions have not helped, the only remaining option is to break the link with the criminal network and facilitate motivational work in connection with, for example, LVU placements. However, the quality of the structures surrounding these placements varies, and the social services described a frustration about risks such as peer contagion effects and a lack of support upon returning home following a placement. The children interviewed also described that they had often committed their most serious offences during periods when they had absconded from a placement. Brå sees a general need to develop better assessment criteria for the social services in relation to placements, to improve the quality of care in placements, and to develop support for those returning home after a placement.
Knowledge development and follow-up
Finally, Brå notes the lack of a systematic approach in the way municipalities’ interventions are documented and followed up. A more uniform use of terminology, and standardised, operationalised goal formulations are needed in order to be able to follow up and evaluate the effects of the interventions on a broad front. This is crucial for building knowledge and improving this work over time.
About the study
The Swedish Government has tasked Brå (the Swedish National Council for Crime Prevention) with studying crimes committed by children under the age of 15 that are investigated under Section 31 of the Young Offenders (Special Provisions) Act (1964:167) – known in Sweden as the LUL Act. These investigations are comparable to criminal investigations and may be conducted for a number of reasons. They are most commonly initiated when children are suspected of crimes that carry a minimum sentence of one year in prison. There are also other grounds for initiating an investigation, for example, if the social services want better information on which to base their decisions on interventions for a child, to investigate whether a child over the age of 15 has also participated in an offence, or to seize goods obtained through the crime. In essence, the crimes subject to LUL investigation are serious offences committed by children under the age of 15.
The government assignment instructed Brå to describe the crimes that give rise to LUL investigations and how they have developed over time, as well as the life situation of the children subject to LUL investigations. Brå was also instructed to describe the interventions implemented by the social services for the child suspects, both before and after the crime. In addition, Brå was to describe various aspects of the investigations conducted under Section 31 of the LUL Act: on what grounds they had been initiated, how long they took, and whether there were regional differences in the application of this legislation.
The study is based on several different data sources. In order to describe trends in suspected offences among children under the age of 15 and suspected offences investigated under the LUL Act, register data describing developments over the last decade were obtained from the Swedish Police Authority and the Swedish Prosecution Authority.
Analyses were also conducted on a sample of 430 children who were subject to an LUL investigation, mainly in 2023. The LUL investigations were analysed for these children, and the child welfare files from the social services were also analysed for 215 of these children. Register data from the Swedish Police Authority, Statistics Sweden, the National Board of Health and Welfare and Brå were obtained in order to describe the life situation of the children subject to LUL investigations, as well as any links to criminal networks.
A small interview study was conducted with social workers, police officers and, to some extent, school staff who work with children who have committed serious offences, as well as with 17 children and young people who have personal experience of committing serious offences before the age of 15. Finally, a workshop was conducted with representatives from several key agencies.