Trends in Adolescent Disadvantage

Policy and Outcomes for Young People under

Labour, the Coalition, and the Conservatives
(1997 to 2019)

Moira Wallace

SPDO research paper 15

August 2023

1
Acknowledgements
This project has been carried out at the Centre for the Analysis of Social Exclusion, where
Moira Wallace is a Visiting Professor of Practice.

The project has been conducted in association with the Social Policies and Distributional
Outcomes (SPDO) in a Changing Britain research programme, which examined policies,
spending and outcomes for all age groups across a range of policy areas, focusing in particular
on the period since 2015. Detailed papers on child poverty, early years, school education,
higher education, social mobility, employment, social security, homelessness and complex
needs, health, and physical security, all of which have a bearing on young people, can be
found on the SPDO website https://sticerd.lse.ac.uk/case/_new/research/spdo/

The project has received funding support from the Nuffield Foundation. Thanks go to the
Foundation and the SPDO advisory board for their support and input.

Thanks also go to many current and past members of CASE for their help in bringing this
project about, to Rana Khazbak for her research assistance in data collection and
deprivation analysis, to Rikki Dean for his assistance with an earlier related project, and to
Nóra Takács and Cheryl Conner for their assistance with the report’s production. I am
grateful to many people for inputs and comments on drafts – to many former colleagues
from the civil service, and to Jonathan Bradshaw, Tania Burchardt, Tammy Campbell,
Chris Goulden, Rana Khazbak, Ruth Lupton, David Utting, and Polly Vizard. I also thank
Aveek Bhattacharya, Alison Hadley and Kitty Stewart for comments, data, and pointers
on specific areas. I alone bear responsibility for errors or omissions in the final text.

The Nuffield Foundation is an independent charitable trust with a

mission to advance social well-being. It funds research that informs
social policy, primarily in Education, Welfare, and Justice. It also funds
student programmes that provide opportunities for young people to
develop skills in quantitative and scientific methods. The Nuffield
Foundation is the founder and co-funder of the Nuffield Council on Bioethics, the Ada
Lovelace Institute and the Nuffield Family Justice Observatory. The Foundation has funded
this project, but the views expressed are those of the authors and not necessarily the
Foundation. Visit www.nuffieldfoundation.org

This report draws on Office for National Statistics statistics which are subject to Crown
copyright and are reproduced under the Open Government Licence v.3.0.

Paper version: v2 (minor corrections), August 2023

2
Contents
List of figures and tables ............................................................................................................ 4

Summary .................................................................................................................................... 7

Chapter 1 Secondary Schools and Post-16 Participation......................................................... 26

Chapter 2 Youth Crime and Justice .......................................................................................... 67

Chapter 3 Child Poverty and Cross-cutting Youth Policies ...................................................... 83

Chapter 4 School Exclusion ...................................................................................................... 96

Chapter 5 School Absence ..................................................................................................... 120

Chapter 6 Teenage Pregnancy ............................................................................................... 144

Chapter 7 Adolescent Alcohol Use ........................................................................................ 165

Chapter 8 Adolescent Drug Use ............................................................................................. 186

Chapter 9 Conclusion ............................................................................................................. 210

Appendix 1 Other Youth Indicators ....................................................................................... 219

Bibliography ........................................................................................................................... 234

3
List of figures and tables

Secondary Schools and Post-16 Participation

Figure ED1: Trends in workforce in maintained secondary schools, England, 1997 to 2010
Figure ED2: Trends in workforce in state-funded secondary schools, England, 2010/11 to 2018/19
Figure ED3: Trends in pupil: teacher and pupil: adult ratios in England, maintained / state-funded secondary
schools
Figure ED4: Trends in percentage of state-funded secondary schools that are academies, England
Figure ED5: Trends in Special Educational Needs, England
Figure ED6: Trends in Level 2 attainment by year turned 16, England
Figure ED7: Trends in proportion of cohort attaining Level 2 by age 16, by free school meal eligibility and SEN
status
Figure ED8: Trends in Level 2 attainment at 16, comparing least and most deprived areas
Figure ED9: Trends in Level 2 achievement at age 16 by ethnicity, 2004 to 2019
Figure ED10: Trends in Level 2 attainment at age 16, white pupils in mainstream schools, by gender and free
school meal status
Figure ED11: Trends in Level 2 attainment by region, mainstream state-funded schools, selected years
Figure ED12: Trends in participation in full time education by 16- and 17- year-olds, England
Figure ED13: Trends in participation in full time education, by gender
Figure ED14: Trends in 15–19-year-olds in education; UK compared with OECD average
Figure ED15: Trends in NEET at age 16, 17 and 18, England
Figure ED16: Trends in NEET rates of 16- and 17-year-olds by gender

Youth Crime and Justice

Figure YJC1: Trends in police workforce (full-time equivalents) in England and Wales
Figure YJC2: Trends in crimes as measured by the Crime Survey for England and Wales
Figure YJC3: Trends in main volume crimes included in the Crime Survey for England and Wales
Figure YJC4: Trends in offences involving knife or sharp instrument in England and Wales.
Figure YJC5: Trends in first-time entrants to the criminal justice system in England and Wales (10–17-year-olds
receiving first reprimand, warning, caution or conviction)
Figure YJC6: Trends in self-reported offending by 10–16-year-olds in mainstream schools
Figure YJC7: Trends in age of offender as estimated by victims of criminal damage, England and Wales
Figure YJC8: Trends in age of offender as estimated by victims of violence, England and Wales
Figure YJC9: Trends in knife and offensive weapon offences committed by 10– 17-year-olds resulting in caution
or sentence

Child poverty and cross-cutting youth policies

Figure CP1: Relative poverty: proportion of children living in households on incomes below 60 per cent of the
contemporary median: 1996-97 to 2018-19
Figure CP2: Absolute child poverty: proportion of children living in households below 60% of a fixed (2010-11)
median in constant prices. 1996-97 to 2018-19

School Exclusion

Figure SE1: Groups over-represented in permanent exclusion, state-funded secondary schools in England,
2018/19
Figure SE2: Trends in permanent exclusion rate in state-funded secondary schools in England
Figure SE3: Trends in permanent exclusion rates by free school meal and special educational needs status,
state-funded secondary schools in England
Figure SE4: Trends in permanent exclusions in state-funded secondary schools in England, by region
Figure SE5: Permanent exclusion rates in state-funded secondary schools in England, by ethnicity, selected
years
Figure SE6: Trends in permanent exclusion rates in state-funded secondary schools in England, by gender
Figure SE7: Trends in appeals against permanent exclusion and their outcome, England, all schools
Figure SE8: Trends in permanent exclusion rate, England and Wales compared

4
School Absence

Figure SA1: Reasons why young people may be absent from school
Figure SA2: Behaviour Improvement Programme initiatives by frequency and theme
Figure SA3: Trends in number of penalty notices issued for attendance, England, all school types
Figure SA4: Trends in overall absence, persistent absence, and severe absence, state-funded secondary
schools, England
Figure SA5: Trends in overall absence in state-funded secondary schools in England by free school meals and
special educational needs status
Figure SA6: Overall absence in state-funded secondary schools in England: change by ethnic group. Selected
years
Figure SA7: Trends in overall absence in state-funded secondary schools in England by national curriculum
year
Figure SA8: Trends in overall absence in state-funded secondary schools, by region
Figure SA9: Trends in absence due to holidays in state-funded secondary schools

Teenage Pregnancy

Figure TP1: Trends in under-18 conception and birth rate: England

Figure TP2: Conception rates per 1000 by mother’s age at conception: England and Wales, 1998 to 2019
Figure TP3: Trends in under-18 conception rates in the most and least deprived local authorities in England
Figure TP4: Trends in under-18 conception rates by English region
Figure TP5: Trends in under-18 conception rates in England, Scotland and Wales
Figure TP6: Trends in live births to 15–19-year-olds: international comparisons
Figure TP7: Trends in proportion of 15-year-old female pupils in England who have had sexual intercourse by
age 15
Figure TP8: Trends in 15-year-old girls in England: use of pill at last intercourse
Figure TP9: Trends in under-17s attending contraceptive clinics in England using Long-Acting Reversible
Contraceptives

Adolescent Alcohol Use

Figure AD1: Trends in children who have ever had a ‘proper alcoholic drink’, England
Figure AD2: Trends in proportion of pupils who usually drink almost every day, or at least once a week,
England
Figure AD3: Trends in young people who have been drunk at least twice, England
Figure AD4: Trends in hospital admission rates for under-18s for alcohol-specific conditions, England
Figure AD5: Trends in proportion of different age groups who have drunk alcohol in the last week, England
Figure AD6: Trends in alcohol-specific hospital admissions of under-18s by region (three year averages)
Figure AD7: Trends in 11–15-year-old pupils who have drunk alcohol in last week, by broad ethnic group,
England
Figure AD8: Trends in weekly drinking by 15-year-olds, by gender, selected countries
Figure AD9: Trends in 11–15-year-old pupils who believe their parent(s) doesn't or wouldn't like them
drinking, England
Figure AD10: Trends in 11–15-year-old pupils' attitudes to drinking by pupils of their age, England
Figure AD11: Trends in self-purchase of alcohol by young people who drink, England
Figure AD12: Trends in locations of drinking for 11–15-year-olds who drink alcohol, England

Adolescent Drug Use

Figure DR1: Proportion of 15-year-old pupils using different types of drugs: 2018, England
Figure DR2: Trends in number of young people in substance use treatment in England by age
Figure DR3: Trends in proportion of 11–15-year-old pupils in England who have ever used drugs
Figure DR4: Trends in proportion of 13–¬15-year-olds in England who have used cannabis or Class A drugs in
last year
Figure DR5: Trends in 16–19-year-olds in England and Wales who have used cannabis or Class A drugs in last
year

5
Figure DR6: Trends in 15-year-old pupils in England who have taken drugs on more than ten occasions, by
gender
Figure DR7: Trends in proportion of 11–15-year-old pupils in England who took drugs in last year by region
Figure DR8: Trends in drug use by 15- and 16-year-old students in the UK and the ESPAD international survey
Figure DR9: Trends in HBSC data on cannabis use by 11–15-year-olds, England and Survey average
Figure DR10: Trends in attitudes to drug use among 11–15-year-old school pupils in England
Figure DR11: Trends in drug use by 11–15-year-old pupils who have truanted or been excluded, England
Figure DR12: Trends in proportion of 11–15-year-old pupils in England who have been offered drugs or think
they would be easy to obtain, by age

Conclusion

Figure S1: Long term trends in adolescent outcomes - summary

Appendix 1

Figure OY1: Trends in proportion of 11–15-year-old pupils who are regular smokers or e-cigarette users,
England
Figure OY2: Trends in proportion of 15-year-olds who have recently engaged in two or more of alcohol use,
drug use, and smoking, England
Figure OY3: Trends in proportion of 11–15-year-old pupils meeting the recommendation of one hour per day
of moderate physical activity, England
Figure OY4: Trends in proportion of 11–15-year-olds who are overweight or obese, England
Figure OY5: Trends in any mental disorder and specific disorders in 11–15-year-olds by sex, 1999, 2004,
2017, England
Figure OY6: Trends in young people’s happiness with different aspects of life, UK
Figure OY7: Trends in perceptions of school, 11–15-year-olds, England
Figure OY8: Secondary school pupils by broad ethnic minority group, selected years, England
Figure OY9: How young people mostly spend their spare time, comparison of the two LSYPE cohorts, England
Figure OY10: Trends in young people communicating with friends by telephone, e-mail or text every day,
England
Figure OY11: Trends in typical weekly hours spent online by UK 12–15-year-olds, self-report
Figure OY12: Trends in 11–15-year-olds who have experienced bullying in the last two months, England
Figure OY13: Trends in pupils who have been involved in a physical fight at least three times in the last year,
England
Figure OY14: Trends in crimes experienced by 10–15-year-olds, England and Wales
Figure OY15: Trends in young homicide victims, rate per million, England and Wales
Figure OY16: Percentage of 15-year-olds in England reporting they had self-harmed
Figure OY17: Trends in hospital admissions for self-harm, 10–14-year-olds and 15–19-year-olds, England
Figure OY18: Trends in suicide rates of 15¬–19s, England

List of Tables

Table SE1: Number and rates of permanent exclusion, state secondary schools, selected years, England
Table SE2: Percentage change in permanent exclusion rate by national curriculum stage, state funded
schools in England, change between 2012/13 and 2018/19
Table SE3: Breakdown of change in permanent exclusions over time by reason recorded
Table SA1: The rise in overall, persistent and severe absence after the pandemic, state funded secondary
schools, England
Table TP1 Conceptions, births, and abortions to girls under 18: England, 1998 and 2019
Table TP2: Births and conceptions by mother’s age, 1998 and 2019, England and Wales
Table AD1: Fiscal cost of alcohol duty cuts in Budgets 2013 to 2018 (£ million compared with indexation)
Table AD 2: Change in adolescent alcohol indicators in HBSC survey 2014 and 2018

6
Summary
Introduction
Over the last two decades, England has seen two enormous waves of change in the
government’s approach to children and young people. The first wave of change was
conducted by the 1997-2010 Labour government and involved significant public spending
increases, new cross-cutting approaches and large-scale national and area-based prevention
programmes. The second wave of change, after 2010, was conducted first by the Coalition,
then by the Conservatives, with the priority being deficit reduction and austerity across
government spending, alongside reducing central government intervention and making
structural changes to the school system. This report, covering the period 1997 to 2019,
focuses on how these changes played out in relation to 11–18-year-olds, across a range of
indicators – child poverty, attainment at age 16, post-16 participation, school exclusion,
school absence, teenage pregnancy, adolescent alcohol use, adolescent drug use, and youth
offending.

Overview
The report’s findings tell a complex story but with clear uniting themes.
(i) The resources and attention given to youth disadvantage increased very sharply
under the Labour government and decreased again very sharply after 2010.
Labour’s approach was characterised by national initiatives intended to deliver
specified outcome targets, but after 2010 national government set no targets,
and left local areas and institutions to decide their priorities and manage the
consequences.

(ii) The trend in outcomes changed equally starkly. For most of the outcomes
measured, substantial improvement started to become evident from the early or
mid-2000s. By the end of Labour’s term, some indicators had halved and many
continued to improve during the first few years of the Coalition. Then in the mid-
Coalition period, between 2012 and 2014, many of these indicators began a
period of stalling and, in some cases, deterioration.

(iii) During the period of improving outcomes, there were notable examples of
disadvantaged groups and areas improving faster than the average. For young
people eligible for free school meals, the attainment gap at Level 2 narrowed, as
did the gap in secondary school absence, and in the gap in permanent exclusion
rates. The difference in teenage conception rates between most and least
deprived local authorities also narrowed.

(iv) There was thus a striking change in prevalence of several forms of adolescent
disadvantage over a very short space of time, to the benefit of the generation
who turned 16 in the early 2010s. Compared with their predecessors born five or

7
 ten years earlier, this generation experienced much lower levels of teenage drug
taking, alcohol use, absence and exclusion from school, and involvement in the
criminal justice system, and also had fewer unwanted conceptions, better
attainment at 16 and higher staying on rates in school. On some of these
measures, this group of young people also fared better than the generation
which came after them.

(v) International comparisons underscore the importance of the changes described.

England has become less of an outlier in teenage birth rates and adolescent
drinking, UK levels of adolescent drug use came much closer to international
averages, and between 2006 and 2014 the UK caught up with the OECD average
for educational participation by 15–19-year-olds.

(vi) The data this report examines covers millions of young people over two decades,
whose childhood and adolescence were affected by multiple policy changes and
other factors. This makes it impossible to say with confidence what caused what.

(vii) However, each chapter considers plausible explanations for the trends it reports.
Overall, the conclusion is that the outcome trends this report examines are likely
to have been caused by a combination of exogenous social changes, direct policy
impacts, and wider reinforcement effects between different areas of young
people’s lives.

The conclusions of the report follow up the broader themes with recommendations arguing
for a more holistic approach to policies affecting young people, through more joining up and
leadership at national level; better local coordination; more emphasis on outcomes with
better use of data and evidence; and greater urgency in addressing problems when they
arise.

The report’s approach

The report’s approach is long-term and cross-cutting. It explores spending, policy design,
and outcome trends between 1997 and 2019 in relation to key aspects of teenage
disadvantage. In a very few cases, important data and policy developments since 2019 are
also mentioned for completeness. The report then considers possible reasons for the
trends, using the data itself and a range of wider research evidence, including government-
commissioned evaluations where they exist, the National Audit Office’s archive of value for
money reports, the work of parliamentary select committees, inspectorates, and bodies
such as the Children’s Commissioner, and a wide range of independent research from this
country and abroad.
The theme of the study is driven by my role as a Visiting Professor in Practice in the Centre
for the Analysis of Social Exclusion (CASE) at LSE. Cross-fertilisation between research and
practice is a key purpose of this scheme. Writing about youth disadvantage draws on my
professional experience in government as Director of the Social Exclusion Unit between

8
1997 and 2002, and as a Home Office official working on crime and criminal justice between
2002 and 2008.1 From CASE’s point of view, drawing together different aspects of youth
disadvantage complements past and present CASE programmes by adding a cross-cutting
dimension and longer timescale to analysis of individual subject areas and individual
administrations. Because I have had personal involvement in some of the policy and delivery
issues described at different points in time, this involvement is noted in individual chapters
where it arises.
The subjects for the study have been selected on the basis of several common features.
First, all of these issues pose significant cost to the individuals who experience them, and
the taxpayer, over the life course. Secondly, evidence from around the world suggests that
these issues are often inter-connected. Many young people experience several of these
issues in parallel; experiencing one of these issues has in many cases been found to make it
more likely that you will experience another; and ‘solutions’ to one problem often turn out
to lie in ‘solving’ one of the others. Thirdly, they represent a set of issues which the Labour
government reflected directly or indirectly in national targets, and which therefore provide
points of reference for assessing the priorities of youth policy at the time, and tracking these
issues through into the decade that followed.
‘Youth policy’ and ‘cross-cutting youth policy’ are perhaps unfamiliar terms in the English
policy context. Governments do not always articulate in one place their policies in relation
to children or teenagers, or manage them as a single system. But whether or not a
government thinks of it that way, the sum of what government does, as it affects young
people, is its youth policy, and that is the picture this report tries to paint.
The scope of the report is therefore wide, but inevitably it cannot cover every issue of
relevance. There are important other stories to be told that are not covered here, notably
on housing and homelessness, developments and pressures in the children’s social care
system, the broader aspects of children’s mental and physical health, and young people as
victims of crime. To give a flavour of the data on some of these areas, an Appendix sets out
data on other youth indicators, including: health behaviours; obesity and being overweight;
overall mental health; wellbeing measures; leisure time and face-to-face socialising;
technology use; experience of being bullied; young people’s involvement in fighting; young
people as victims of crime; and self-harm and suicide by young people.

Structure of the report

The report is structured as follows. Chapters 1 to 3 provide introductory context by outlining
the overall social policy posture of Labour, Coalition, and Conservatives in three
‘cornerstone’ policy areas: secondary education, youth crime and youth justice, and child
poverty and youth support. These are followed by detailed case studies of five cross-cutting
areas - school exclusion, school absence, teenage conceptions, adolescent alcohol use, and
adolescent drug use (Chapters 4 to 8). The case studies are looked at in terms of goals,
policies and spending, and trends both in national level outcomes and their socio-economic
and regional distribution. These chapters then discuss plausible explanations for the

1
Involvement in particular initiatives described will be noted in individual chapters.

9
changes in outcomes seen over the period, considering both policies and social
determinants, and drawing on the wider literature. Chapter 9 takes a step back again to
discuss the broader picture suggested by the analysis, and concludes with four
recommendations.

Summary of key findings

The key findings are set out in the summary below. Trends in eight key indicators are
illustrated in the panel of charts at Figure S1 later in the chapter. For sources and citations
please see the chapters that follow.

‘Cornerstone’ policy chapters

Chapter 1: Secondary education and post-16 participation
The Labour government's education policy was marked out by large year-on-year increases
in school funding, significant increases in the school workforce, hands-on intervention to
improve school effectiveness, and quantified targets for improvement applying to schools,
local authorities and the government itself. Labour devoted considerable attention to
making schools work for pupils from disadvantaged backgrounds, through the targeting of
funding and area initiatives, policies such as curriculum flexibility and the promotion of
vocational qualifications, initiatives to reduce absence and exclusion, and the introduction
of financial incentives to stay on in education after 16.
The defining policies of the Coalition and Conservatives were encouragement for schools to
take up Academy status and for new schools to be established, and large-scale changes to
GCSEs. During the Coalition administration, the previously announced policy of raising the
participation age to 18 entered into force, and major changes to the special educational
needs regime were introduced.
Between 1999/2000 and 2009/10 spending per pupil in secondary schools increased by 6
per cent in real terms annually. Spending per pupil in secondary schools then fell in real
terms. Pupil-teacher ratios fell under Labour and rose again under the Coalition and
Conservatives.
Attainment of Level 2 by age 16 increased significantly under Labour and until 2012, but
then fell and remained lower through until 2019. On this measure, the attainment gaps for
both pupils eligible for free school meals, and for pupils with special educational needs but
no statement or education and health care plan, narrowed until 2012 then widened again.
Between 2001 and 2019, there was a sizeable increase in the proportion of young people
staying in full-time education after 16. For 17-year-olds the increase was over 20
percentage points. Most of this increase occurred by 2010. Historically, the UK had lagged
behind OECD average levels of educational participation by 15- to 19-year-olds but by 2014
this gap was eliminated. The proportion of 16- and 17-year-olds not in education,
employment, or training (NEET) fell from the mid-2000s peak, but between 2015 and 2019
it edged up again for boys while improving for girls.

10
Chapter 2: Youth crime and youth justice
The Labour government's approach to crime emphasised both prevention of offending, and
detection and punishment of offenders. Early priorities were the creation of local Youth
Offending Teams and speeding up the operation of the youth justice system. National
targets were set for crime reduction, and police forces and local partners were held to
account for their performance. There was significant investment in law and order with a 50
per cent real terms increase in police spending in the government’s first ten years, funding
16,000 extra police officers and the roll-out of community support officers. Coordinated
action on drugs, alcohol, exclusion and school absence addressed some of the major drivers
of offending.
Labour met its pledge to halve the time from arrest to sentencing for persistent young
offenders. A separate target, to increase the number of offences brought to justice, proved
perverse in incentivising the police to pursue arrests for easy-to-detect crimes, and was
eventually dropped. After 2007, the focus switched to measures to divert young people
from the formal justice system where possible.
The Coalition introduced elected Police and Crime Commissioners at local level. There were
early real terms cuts in both police funding and other justice budgets. Police officer numbers
fell by 12 per cent between 2010 and 2015 and continued to fall until 2018.
The Coalition and Conservatives retained and intensified the move to divert young people
away from the formal justice system. But for youth cases that went to court, the average
time from offence to completion rose again.
The level of overall crime, as measured by the Crime Survey for England and Wales, fell by
44 per cent during Labour's period in office, and by 36 per cent between 2010 and 2017,
then flattened between 2017 and mid-2019. Recorded knife crime rose after 2014/15.
The level of overall youth crime can only be inferred through other indicators. This will be
very frustrating to readers. However, several different data sources – self-reported
offending measures from the 2000s, and longer-term trends in victim accounts, and criminal
justice data – suggest that youth crime fell substantially between the mid-2000s and the
mid-2010s.
Chapter 3: Child Poverty and Cross-cutting Support
In 1999, Labour announced its target to halve the number of children living in relative
poverty, and began a series of real-terms increases in benefits to children both for families
in and out of work. Overall, compared with spending in 1996/97, by 2010/11 there had been
a real terms increase of nearly £24 billion in cash transfers to families with children.
Under the Coalition and Conservatives, a series of policy changes made the benefit and tax
credit system for families with children less generous. Between 2009/10 and 2018/19 per
capita social security spending on children fell by 25 per cent. The Conservative government
legislated in 2016 to replace the child poverty targets with indicators which were not about
income poverty.

11
In terms of outcomes, under Labour, child poverty fell in both relative terms and against a
fixed real threshold, with the improvement concentrated in the early part of Labour’s term.
The measure shown in Figure S1 is the fixed real income threshold, also known as absolute
poverty: by 2010-11, the proportion of 11–18-year-olds in poverty on this measure was 17
percentage points lower than in 1996-97. A small rise and fall over the 2010s left the
numbers one percentage point lower overall in 2018-19 compared with 2010-11.
Labour put resources into several prevention programmes designed to address the risk of
multiple adverse outcomes for children and young people, including – for teenagers – the
Connexions service. Labour broadened the remit of the Department for Education and Skills
to include children’s social services and coordination of youth policy, and required local
authorities to bring education and social services together as one of many changes in the
Every Child Matters programme. In 2007, the Department for Education and Skills was
renamed the Department for Children, Schools and Families.
In 2010 the Coalition and Conservatives removed the words ‘children and families’ from the
title of the Department for Education. Connexions was effectively abolished in 2012, and
under the Coalition and Conservatives local authorities’ spending on children fell 20 per cent
in real terms between 2009/10 and 2019/20.

Case study policy chapters

Chapter 4: School exclusion
Between 1997 and 2010, the Labour government targeted high levels of school exclusion
through multiple measures, including (short-lived) national targets, statutory guidance with
an emphasis on preventing exclusion, and a range of funded programmes to tackle the
causes of exclusion and encourage good practice in behaviour management.
Under the Coalition and Conservatives, changes to the exclusion appeals system meant a
headteacher’s decision to exclude could no longer be overturned. A funding squeeze on
schools, workforce reductions, perverse incentives in the special educational needs funding
system, and changes to GCSEs, made it harder for vulnerable pupils to do well, and harder
for schools to help. Reductions in a range of local authority, health and other services left
schools and young people struggling to access support.
The rate of permanent exclusions in secondary schools halved between 2003/04 and
2012/13 but rose again thereafter. Inequalities were reduced then rose again. It seems likely
that a set of mutually reinforcing policy changes within education were important in driving
both the fall in exclusions and its subsequent rise, with potential contributions from wider
factors such as changes in drug and alcohol use. The chapter also describes the parallel
phenomenon of ‘off-rolling’ and what is known about its causes.
Chapter 5: School absence
The Labour government set itself targets to reduce school absence throughout its time in
government and drove multiple initiatives to improve attendance in partnership with
schools, local authorities, and police. These initiatives included better systems to track and
respond to absence, as well as policies to address the factors that kept children away from

12
school. More support was provided for pupils who were falling behind; new flexibility
allowed schools to make the curriculum more motivating for those at risk of disengaging;
and police worked more closely with schools to tackle and deter offending in and around
schools. Many of these policies were evaluated and found to be effective.
After 2010, absence had a lower priority. Few specific absence policies were introduced, and
evidence does not suggest that they were effective. Wider schools and youth policy was less
well resourced, and less able to tackle the many issues which underpin absenteeism.
Overall absence fell fairly steadily in secondary schools between 2000/01 and 2013/14, then
after a few years began to edge up again. ‘Severe absence’ (missing more than 50 per cent
of schooling) was first measured in 2006/07 at 1.6 per cent of secondary school pupils. It
halved to 0.8 per cent by 2013/14 then rose again to reach 1.3 per cent in 2018/19. The
‘absence gap’ for pupils eligible for free school meals and those with special educational
needs first reduced then grew again. Since the pandemic, absence rates are even higher,
with the rate of severe absence in secondary schools reaching 2.7 per cent in 2021/22, and
5.7 per cent for secondary school pupils eligible for free school meals.
The decline in absence rates seems to have been driven by four main factors: a step-change
in government commitment, sustained over a long period; investment and roll-out of
multiple policies effectively targeted on the causes of absence; a supportive climate of
wider schools policy and funding; and reductions in a range of wider risk factors. The
reasons for the later stalling and rise in absence include: de-prioritisation of absence; few
and ineffective absence policies; the impact of wider secondary schools policy and funding;
and rises in the wider risk factors for absence.
Chapter 6: Teenage conceptions
In the late 1990s the level of teenage pregnancy in the UK was a matter of great public
concern. While many other Western European countries had seen falling teenage birth rates
in the 1980s and 1990s, the UK had not. Soon after coming to office in 1997 the Labour
government set in hand a ten-year national strategy on teenage pregnancy in England,
incorporating information campaigns for young people and parents, a drive to improve sex
and relationships education, increased focus on making contraception available to young
people, dedicated funding for local work with at-risk groups, and improved support for
teenage parents. These actions were complemented by policies on educational under-
attainment, absence from school, and adolescent drinking, and the strategy was
underpinned by the principle of joined-up action at national, regional and local level.
After 2010, Coalition policy was to encourage local areas to maintain their efforts to reduce
teenage pregnancy, and there was no dramatic change in policy. However, national support
teams ended, and the Teenage Pregnancy Unit closed in 2012. Since 2014/15 funding for
sexual health services has declined, and several expert bodies have raised concerns about
access to some of the most effective forms of contraception.
Between 1998 and 2019 the rate of conceptions to under-18s fell by 66 per cent. The
milestone of halving the 1998 level was reached in 2014. Progress in the early 2000s was
mainly seen in London, particularly Inner London, but steep reductions were seen in all

13
other regions from 2007 or 2008 onwards. Internationally, the UK’s reduction in teenage
births was exceeded by that in the US but the UK improved relative to New Zealand, started
to catch up with Canada and Australia, and significantly narrowed the gap with its European
neighbours.
The reduction in teenage conceptions over this period reflects both teenagers becoming less
likely to have sex in their early teens, and sexually active teenagers becoming more likely to
use the pill or long-acting reliable forms of contraception (LARCs) such as implants and
injections. Plausible explanations for these changes include new policies to raise access to
LARCs, an increase in the proportion of young people getting their sex education mainly
from school, and significant changes in wider factors such as increased educational
attainment and staying-on rates, and reduced school absence and alcohol use.
Chapter 7: Adolescent Alcohol Use
Policy-making on adolescent drinking had a rather slow start under Labour. However, from
2003 onwards, a variety of government initiatives were introduced to reduce alcohol use by
young people, including significantly increased enforcement of the law on underage selling,
controls on drinking in public places, and public information campaigns on the risks of
teenage drinking. For a while, the government also increased alcohol taxation. The
government also addressed other adolescent issues which are risk factors for drinking, such
as drug use, school absence, and school exclusion.
In 2013 the Coalition abolished the policy of increasing alcohol duty by more than inflation
and began instead to freeze or cut duty rates. Alcohol duty freezes or cuts were enacted in
every budget but one for the rest of the decade, at a sizeable fiscal cost. And although the
Coalition announced in 2012 that it intended to introduce a minimum unit price for alcohol,
in 2013 it dropped the idea. After 2014/15 funding for alcohol treatment was cut, and risk
factors such as drug use, school absence, and school exclusion started to rise again.
Between 2003 and 2014, there was a reduction of more than half in the proportion of
under-14s who had ever tried alcohol. The proportion of 15-year-olds who had been drunk
at least twice also halved, and alcohol-related hospital admissions for under-18s fell.
Although adolescent alcohol use has fallen in many high-income countries over recent
decades, the size of the fall in England stands out in international comparisons. However,
the data suggest that reductions in teenage drinking in England stalled after 2014.
These striking trends are likely to have many causes but are likely to be best understood by
considering a series of related changes that occurred over the period. These include changes
in the age of first introduction to alcohol; changing attitudes to alcohol amongst parents and
young people; changes in school exclusion, absence, and rates of staying on in education
after 16; changes in levels of drug-taking and smoking; enforcement of the law on underage
alcohol purchasing; changes in where young people drink and in their patterns of socialising;
and changes to alcohol taxation. Every one of these factors was moving in a direction likely
to reduce adolescent drinking for part or all of the period when young people’s alcohol use
was declining. Some of the factors then took a different direction after 2013, which may
help explain the subsequent stalling of progress.

14
Chapter 8: Adolescent Drug Use
The Labour government made drugs policy an early priority on taking office, publishing a
cross-governmental strategy in 1998 and setting a target to reduce the proportion of under-
25s misusing illegal drugs. Between 2000/01 and 2005/06 spending on drug treatment in
England more than doubled and there was significant investment in the young people and
communities elements of the drugs programme. The strategy relied on significant police
investment of time, and local level partnerships known as Drug Action Teams, and the
National Treatment Agency was set up to oversee the expansion of drug treatment. Young
people benefited from increased investment in specialist drug treatment; from the ‘FRANK’
advice line for teenagers and parents; and from drugs prevention projects.
Under the Coalition, both the Department of Health and the Home Office pushed
responsibility down to local areas, and funding for drugs prevention and treatment fell. The
National Treatment Agency was subsumed into Public Health England. By 2018/19 the
number of young people receiving specialist drug and alcohol treatment was a third down
on 2010/11. Drug seizures by police forces fell by a quarter between 2010 and 2015, and a
reduced number of police officers were soon facing significant challenges from the new
‘County Lines’ model of drugs supply. A review of drug markets that began in 2019 led the
government to announce a major overhaul of policy in 2021, including promised investment
in treatment, a move to revive local drug partnerships, and the re-creation of a unit to
coordinate drugs policy across departments.
Most indicators of adolescent drug use fell during Labour’s period in government, and
England/the UK ceased to be such an outlier in international comparisons. But the
improving trends stalled and reversed after 2013 or 2014. The proportion of 15-year-olds
who had ever taken drugs halved between 2003 and 2011, then rose again after 2014.
Cannabis use by under-16s also halved in the decade after 2003 then edged up again. Class
A drug use by 13-to 15-year-olds and 16–19-year-olds showed a similar pattern of fall and
rise.
During the period that adolescent drug use was falling a set of significant related factors
were changing in a positive direction. Treatment for young people with drug problems
became more available. There was a decline in the number of young people persistently
absent or excluded from school. Young people’s attitudes to drugs became more negative.
Surveys found that school pupils had become less likely than previous generations to have
been offered drugs and less likely to know where to get drugs. Very early experimentation
with drugs became less common. These changes are likely to have been mutually
reinforcing.
During the 2010s, by contrast, early experimentation with drugs became a little more
frequent, attitudes to drugs became less cautious again, exclusion and severe absence rose
again, and treatment availability fell. A new model of drug dealing took hold which directly
recruited vulnerable young people, and young people became more likely to be offered
drugs. Again, these changes are likely to have been mutually reinforcing, but this time in the
wrong direction. The government’s overhaul of policy in the wake of the Black review

15
represents an important recognition of the drivers of this complex problem and the urgency
of addressing them.

Conclusions (Chapter 9)
Each of the policy stories described in the case study chapters is interesting in its own right.
But behind the individual subjects, a bigger overall picture emerges.
The policy story
The report recounts sizeable shifts in multiple social policies affecting young people, both
after 1997 when Labour took over from a Conservative government, and in the 2010s when
the Coalition and Conservative administrations in turn came to office. During the first period
there was a major increase of activity as Labour expanded services, addressed cross-cutting
issues and service inequalities, and put more money in the hands of poor families. After
2010, the focus was on deficit reduction and austerity across government spending,
alongside reducing central government intervention and making structural changes to the
school system.
The outcome story
The report also recounts sizeable changes in outcomes. Over Labour’s time in office as a
whole, the key indicators of youth disadvantage discussed in this report improved
substantially and many continued to improve during the first few years of the Coalition.
Then in the mid-Coalition period, between 2012 and 2014, many of these indicators began a
period of stalling and, in some cases, deterioration.
The outcome trends for eight key indicators of adolescent disadvantage are illustrated in the
panel of charts at Figure S1 below. The graphs show that at the end of Labour’s term all
eight were substantially lower than at the beginning, with some indicators having halved.
Many saw steep falls virtually in parallel over the decade from 2003 to 2012. (This period is
shaded green in the graph).
The picture is not uniform. For child poverty, the improvement was concentrated in the
early part of Labour’s term. And two indicators (10–17-year-olds entering the criminal
justice system, and secondary pupils missing over 50 per cent of schooling) show a steep fall
from a slightly later starting point, around 2007. In the case of 10–17-year-olds entering the
criminal justice system, 2007 marked a change in trend (discussed in Chapter 2). For
secondary pupils missing over 50 per cent of schooling (‘severe absence’) no data were
collected before 2006/07.

16
 0.0%
0.2%
0.4%
0.6%
0.8%
1.0%
1.2%
1.6%
1.8%

1.4%
10%
20%
30%
40%
50%

0%
60%
1997/98
1998/99 1996-97
1999/00 1997-98
2000/01 1998-99
2001/02 1999-00
2002/03 2000-01
2003/04 2001-02
2004/05 2002-03
2005/06 2003-04
2006/07 2004-05
2007/08 2005-06
2008/09 2006-07
2009/10 2007-08
2010/11 2008-09
2011/12 2009-10
2012/13 2010-11
2013/14 2011-12
constant prices, AHC)

2014/15 2012-13
2015/16 2013-14
2016/17 2014-15

50 per cent of schooling

Abolute poverty: 11 to 18s
(below 60 % of 2010/11 median in

2017/18 2015-16
2018/19 2016-17

Secondary pupils missing over

2017-18
2018-19
10
20
40
50

0
30
60
%

10
20
30
40
50
60
70

0
rate/
1,000
1998 1997/98
1999 1998/99
2000 1999/00
2001 2000/01
2002 2001/02

14 year olds
2003 2002/03

15 year olds
2004 2003/04
2004/05
2005

16 year olds
2005/06
2006
2006/07
2007
2007/08
2008
2008/09
2009
2009/10
mother
2010
2010/11
2 by age 16

2011
2011/12
2012
2012/13
2013 17 year olds 2013/14
2014
2014/15
2015
2015/16
2016
2016/17
Figure S1: Long term trends in adolescent outcomes - summary

Proportion not attaining Level

2017
2017/18
Teenage conceptions by age of

2018
2018/19
2019

10
20
30
40
50
60

0
%
0
100,000
120,000

40,000
80,000

20,000
60,000

1997

17
1998
1997/98
1999 1998/99
2000 1999/00
2001 2000/01
2002 2001/02
13 year olds

2003 2002/03
2004 2003/04
2004/05
14 year olds

2005
2006 2005/06
2007 2006/07
2007/08
2008
2008/09
2009
2009/10
2010 2010/11
2011 2011/12
15 year olds

2012 2012/13
2013 2013/14
2014 2014/15
10-17 year olds entering

drank alcohol in last week

2015 2015/16
13, 14 and 15 year olds who

2016/17
criminal justice system (E&W)

2016
2017 2017/18
2018 2018/19
10
12
14

0
2
6
8

4
%

0.15
0.25
0.35

0.05

0
0.1
0.2
0.3
0.4

1998
1999 1997/98
1998/99
2000
1999/00
2001
2000/01
2002
2001/02
2003
2002/03
2004
Green shading is decade 2003 to 2012. Refer to chapters for sourcing. Alcohol figures are from Smoking Drinking and Drug use by Young People 2018, Table 5.5

2003/04
2005 2004/05
2006 2005/06
2007 2006/07
2008 2007/08
2009 2008/09
2010 2009/10
2011 2010/11
in last month

2012 2011/12
secondary schools

2013 2012/13
2014 2013/14
2014/15
Permanent exclusion rate,

2015
2016 2015/16
2016/17
11-15 year olds who took drugs

2017
2017/18
2018
2018/19
The trends did not break abruptly with a switch of government. But after the mid-Coalition
period, things started to change for many indicators. After 2011/12, Level 2 attainment at
16 began to deteriorate and the cohorts who turned 16 in the next three years each had
slightly more young people not achieving Level 2. Indeed, despite a slight improvement later
in the decade, Level 2 attainment had still not recovered its 2011/12 level by 2018/19. The
trends in permanent exclusion, teenage drug use, and severe absence all bottomed out
between 2012 and 2014 then started to rise again. After 2014, alcohol use probably
plateaued, taking all indicators into account. Only two indicators continued to fall
consistently between 2010 and 2019 – teenage conceptions, and 10–17-year-olds entering
the criminal justice system. It is not clear how far the latter reflects changes in offending, as
against changing criminal justice practices and resourcing.
A changing picture on inequalities
During the period of improving outcomes, there are some striking examples of
disadvantaged groups and areas improving faster than the average.

• Pupils eligible for free school meals and pupils with special educational needs
narrowed the gap with other pupils in terms of Level 2 attainment, school absence,
and permanent exclusion. The deprivation gap in teenage pregnancy also narrowed.

• Black Caribbean pupils had been heavily overrepresented in permanent exclusions

and amongst those not achieving level 2 by age 16, but saw some of the largest
improvements until 2011/12. White pupils were the ethnic group with the highest
drinking prevalence in 2003, and then saw the biggest reduction by 2014.

• Regionally, London moved from having the highest rates of both teenage
conceptions and school absence to being among the lowest. Some North-South
inequalities began to narrow, as the North East had the greatest fall in alcohol-
related hospital admissions for under-18s and, along with Yorkshire and Humberside,
saw the largest increase in level 2 attainment between 2001/02 and 2011/12.
After 2012, progress in narrowing inequalities stalled, and some started to widen again.

• Increases in permanent exclusion were sharpest for pupils eligible for free school
meals, pupils with special needs but no statement or education and health care plan,
and pupils of Black Caribbean, Mixed White/Black Caribbean, and Gypsy Roma
ethnicities. The increases in exclusion were also particularly pronounced in the North
East, and North West.

• Pupils eligible for free school meals, pupils with special needs, and pupils of Black
Caribbean and Mixed White and Black Caribbean ethnicity all saw large increases in
overall absence and the regions with the largest increases in persistent absence after
2013/14 were the North East and Yorkshire and Humberside.

• The deterioration in Level 2 attainment at 16 was disproportionate for pupils on free

school meals, and those with special educational needs but no statement or
education and health care plan, and the North East was the region with the largest
deterioration between 2011/12 and 2017/18.

18
 • During the decade, the attainment gap in secondary schools has stopped narrowing
on all the key measures. Black Caribbean pupils saw a 12.9 percentage point
decline in their Level 2 attainment between 2012 and 2019.

Change between cohorts

The picture presented in Figure S1 represents a striking change in prevalence of several
forms of adolescent disadvantage over a very short space of time. This is particularly marked
for the generation who turned 16 around 2012 and 2013. Compared with their predecessors
born five or ten years earlier, this generation experienced much lower levels of teenage
drug taking, alcohol use, absence and exclusion from school, as well as lower rates of entry
into the criminal justice system. They also had fewer unwanted conceptions, better Level 2
attainment at Age 16, and higher staying-on rates in school. On some of these measures,
this group of young people also fared better than the generation which came after them.
This striking data about the teenage years of this generation create the possibility of a
cohort effect which may well become visible in a range of other outcomes in later life.
International comparisons
International comparisons underscore the importance of the changes described. During the
2000s, England became less of an outlier in teenage birth rates; had the largest fall in
weekly drinking for both boys and girls, out of 36 European countries; and UK levels of
adolescent drug use came much closer to international averages. Between 2006 and 2014
the UK caught up with the OECD average educational participation rate for 15–19-year-olds,
closing an eleven percentage point gap. A study of 23 countries between the mid-1990s and
2010 found that only eight of the countries covered reduced child poverty, with the UK
seeing the largest reduction. However, between 2013 and 2018 the United Kingdom had the
third largest increase in relative child poverty out of over 30 comparator countries. Between
2014 and 2018, cannabis use by boys in England started to diverge from the international
average again, and since 2014 adolescent drinking in England has not matched the further
falls seen in other countries.
How can we explain the trends?
These striking trends in overall outcomes and in their distribution are intriguing and call for
explanation. But there is no research method that can determine the causes with certainty.
The data this report examines is not the output of a controlled experiment, but instead
covers the entire adolescence of millions of young people over two decades, whose
childhood and adolescence were affected by multiple policy changes and many other
factors. This is an important caveat to any attempt to explain these trends. But it does not
mean that we cannot marshal the evidence we do have, and use that to consider plausible
hypotheses that might explain what has happened.
Having done this, my assessment is that the pattern of outcome trends this report examines
is likely to have been caused by a combination of exogenous social changes, direct policy
impacts, and wider reinforcement effects between different areas of young people’s lives.

19
Social, demographic, or technological change?

To take these in turn, there is clearly some role for social, demographic, and indeed
technological change in some of the trends observed. But, as discussed in the individual
chapters, exogenous factors of this kind cannot on their own explain the magnitude of the
outcome changes that have occurred, or their distribution, or their stalling. To take one
example, while it is likely that the changing ethnic mix of the youth population has made
some contribution to falling drinking levels, this cannot account for the scale of the fall, the
change that has been seen in locations of drinking and how alcohol is obtained, the fact that
the fall was greatest among white young people, or the stalling of the change after 2014. To
take a different example, it is clear that the extraordinary growth of mobile phone
ownership and use has played a part in changing patterns of socialising among young people
that may have impacted upon the use of alcohol, drugs and sexual behaviour. But many
other factors appear likely to have played a part, and research cited in the chapters finds
that those who communicate most online are more likely than others to use alcohol and
drugs.
Policy
The report suggests instead that the changes in the outcomes discussed are multi-causal,
with policy changes likely to have been part of the explanation. The report has drawn on a
wide range of analysis to reconstruct the policy picture, including government-
commissioned evaluations where they exist, the National Audit Office’s archive of value for
money reports, the work of parliamentary select committees, inspectorates, and bodies
such as the Office of the Children’s Commissioner, and a wide range of independent
research from this country and abroad. These help to reconstruct the picture of what was
done at different points in the period, and the inputs and outputs that were delivered. The
evidence varies in its rigour, and many policies were not evaluated to the most robust
standard (or even evaluated at all). Nonetheless, there is substantial evidence in every
chapter suggesting that government policy initiatives, in different ways and at different
times, are likely to have contributed directly to the changes in outcomes that have occurred,
both when they were improving, and when they deteriorated.
Reinforcement effects
It is on the question of indirect and reinforcement effects that the research literature
becomes rather scarce. The debate about the impact of past policy on real world outcomes
tends to be conducted within the confines of each policy and academic discipline. It is
widely known that risk factors and adverse outcomes tend to cluster together, and the
literature on ‘what works’ rightly stresses the importance of multi-component programmes
to address adolescent risk behaviours. Despite this, there is not a great deal of research
looking into whether these ‘crossover’ and reinforcement impacts have actually occurred in
the recent policy environment, for example through changes in educational participation
affecting changes in crime, or falling teenage alcohol use impacting on the teenage
conception rate.
The individual chapters of the report draw attention to several possible reinforcement
effects of this type. Chapter 7, for example, identifies nearly a dozen factors across multiple
policy areas which all, for a period, were moving in a direction likely to reduce adolescent

20
drinking, and could have been very powerful in combination. Many other chapters end with
a similar list. At the aggregate level, the trends shown in Figure S1 are consistent with the
interpretation that there were multiple mutually reinforcing changes in young people’s
environment, which together caused them to go through adolescence in a very different
way from their predecessors.
Cohort effects
This brings us back to the possibility of a cohort effect. The young people whose outcomes
improved so sharply in the 2000s went through childhood and adolescence in an era of
falling family poverty and rising school resourcing, and some of them will have benefitted
from Sure Start in their early years. Growing up, they were exposed to lower levels of crime,
and their peer groups engaged less in alcohol and drug use. In secondary school they
benefited from a very different 14-19 curriculum, with much more support to help them
stay in learning after age 16. These changes in their environment could easily have had
substantial effects on their performance in education as well as on their involvement in risky
behaviours.
The generation of young people who turned 16 in the early coalition years were also subject
to multiple policy changes compared with their predecessors. Many will have been affected
by changes to benefits; young people were no longer able to study vocational qualifications
to the same extent; spending on mental health, youth services, and drug services was
falling; Connexions and the Education Maintenance Allowance had been abolished; and
virtually all frontline services that young people might need to draw on were going through
a time of funding pressure and organisational turbulence. Looked at in that context, the
slowing and stalling of progress from mid-Coalition onwards becomes easier to understand,
as a cumulative result of multiple policies whose likely interaction appears not to have been
recognised in advance.
Policy is not just about spending
Policy is a very broad term, and as will be obvious from the detailed chapters, it covers many
different dimensions of the way governments operate. These include the priorities that
governments set; the areas they decide not to prioritise; how these priorities are reflected
in spending budgets; the design of incentives for institutions, markets and individuals;
collaboration and accountability structures; national/local relationships; and the degree of
interest in tracking performance. The analysis in the detailed chapter suggests that all of
these are important, although they do not all attract the same attention in political debate.
In weighing the impact of government policies, therefore, we need to look at more than just
spending levels. This is an important point to recognise in relation to the three
administrations we are discussing. Labour did indeed raise spending on many social
programmes – both to expand existing services and to create whole new spending
programmes in areas such as the Education Maintenance Allowance, the Connexions
service, or the teenage pregnancy strategy. But some of its policies cost very little – for
example, increasing the enforcement of alcohol licensing, giving parents clear guidance on
adolescent drinking, making contraception services more youth-friendly, and giving young

21
people clear information on drugs. One policy – increasing alcohol duties – actually raised
money as well as contributing to public health.
Just as Labour’s policy changes cannot all be categorised simply as spending increases, the
Coalition and Conservative governments’ policy changes cannot all be bracketed under the
umbrella of deficit reduction. Certainly, austerity loomed large across many youth spending
areas. But many Coalition and Conservatives policies cost money rather than saving it (for
example the academy and free schools programmes, and the repeated reductions in alcohol
duty). And many of the most significant policy changes under the Coalition and
Conservatives were structural and design changes which sprang from concerns other than
austerity - such as the desire for a smaller state, the preference against regulation, the
desire to reduce local authorities’ role in running schools and the view that secondary
education should be more traditionally academic.
Lessons and recommendations for the future
This consideration of policy and outcomes on adolescent disadvantage offers many
potential lessons for the future. But those lessons will have to be applied in a new context
that is made much more difficult by the experience and aftermath of the pandemic, and the
state of the economy.
There are many issues competing for political attention in 2023. Policy on children and
young people is always important and should be particularly prioritised at present. This is
because the factors that are new in our current context - the educational, social and mental
health issues which young people have experienced during the pandemic, and the impact of
the cost of living crisis – have been overlaid on a system that was already struggling in many
aspects.
The detailed chapters of this report offer plenty of insight into specific elements of policy
that need attention, with drugs, special education needs, and school absence all high on the
list. But these examples are just a few of the items amongst many youth policy issues that
need to be tackled.
The report suggests four overarching recommendations for policy-makers which are
relevant to all the youth policy areas discussed in this report.
i. Joining-up and leadership at national level. Individual youth policy areas will be
easier and cheaper to tackle if the government does it in a coordinated way. This will
maximise the synergies between policies, help to avoid perverse incentives, reduce
the scope for gaps and duplication, and help develop preventive investments. These
points have been powerfully made in several recent reports in relation to children
and young people.2 I strongly agree with their arguments that underplaying the
government leadership role is a false economy.

2
Commission on Young Lives, Hidden in Plain Sight: Final Report by the Commission on Young Lives (Oasis
Charitable Trust, 2022); House of Commons Committee of Public Accounts, Support for Vulnerable
Adolescents: Thirty-Seventh Report of Session 2022-23 (HoC, 2023).

22
 ii. Local coordination. Neglecting the cross-cutting role at local level is also a mistake. It
matters enormously whether local services are encouraged to work together, or not.
Central government should take a more consistent and constructive interest in this,
and above all needs to satisfy itself that there are enough frontline staff, with enough
time, to support young people and their families who are at risk of, or already
experiencing, serious difficulties. There are many different models for this –
Connexions was one model in the past, and the Commission on Young Lives idea of a
‘Sure Start for teenagers’ is another possible model. Currently, however, capacity of
this kind is fragile and overstretched in many parts of the country. It is in central
government's interests to develop a sustainable model for the future. Indeed, it is
very hard to see how effective or lasting solutions can be found to the large number
of children missing school, the scale of criminal exploitation, or the number of 16- to
17-year-olds whose activity is unknown, without a well-resourced and visible
frontline resource taking on such a role on a continuing basis.

iii. More emphasis on outcomes with better use of data and evidence. As this report has
shown, a great deal of data is already available, most of it produced by the
government itself. But it does not appear that government is currently bringing this
picture together at the aggregate level to track the overall outcome picture.3 Good
policy would be served by an investment in collating this data for the public and
policy-makers, and by investment in filling the key data gaps.4 Taking this forward
would support the NAO’s recommendation that central government needs ‘a shared
strategic, data-led view of the complex set of problems it is trying to address, and a
strong evidence base to determine the most efficient and effective ways of
addressing them’. 5

iv. Urgency: Above all, the analysis in this report points to the need for greater urgency
in addressing the problems of youth policy. One can understand the extraordinary
pressures that Ministers and civil servants have been under in recent years. But
children only get one childhood. While the clock ticks on, it is the frontline and
children themselves who are left trying to cope in the system as it is.

There is, without doubt, a demanding agenda to tackle. But this report’s conclusions end
on a more hopeful note. Significant change in young people’s outcomes has been achieved
in our own recent past. Those issues appeared entrenched when efforts to tackle them
first began. But it turned out that there were many things that could be tried, and many of
them appear to have worked. Targets that some thought fanciful were, in many cases,
achieved. Forms of disadvantage that were thought to be intractable turned out not to be.
Many young people overcame challenges that had impeded previous generations. These
lessons of the past can offer encouragement for the future, at a time when effective youth
policies are much needed.

3
Public Accounts Committee, Supporting Vulnerable Adolescents: Transcript of Oral Evidence (HoC, 2022).
4
Amongst the data gaps mentioned in the report are: restoring annual drug and alcohol surveys of young
people; reinstituting self-report crime surveys; gathering more data on hard-to-reach groups; and tracking
trends in the numbers of young people experiencing multiple disadvantage.
5
National Audit Office, Support for Vulnerable Adolescents (NAO, 2022).

23
 About this report

The age focus of the study is 11- to 18-year-olds.

The geographical focus is for the most part England. However:

• On some policy topics a wider geographical area is covered to reflect policy

responsibilities. For instance, justice and policing policies cover England and
Wales, the drug classification system is UK wide, as is alcohol taxation.
• Where another UK nation has taken a very different approach to one of the policy
areas discussed, this is covered briefly where space permits.
• The report uses data for England and Wales, or at GB or UK level, if suitable data
for England alone is not published, for example in some international
comparisons.

The time frame is from 1997 to 2019. The starting point is determined by the start of the
Labour government and the stopping-point represented the last year before the
pandemic. Policies are discussed by administration, but analysis of trends focuses on sub-
periods defined by turning-points in the data. In a very few cases, important data and
policy developments since 2019 are also mentioned for completeness.

The data sources used are mainly long-established government data series, most of which
have the status of National Statistics. The report also uses cohort studies such as the two
cohorts of the Longitudinal Survey of Young People in England,6 and international survey
data such as the World Health Organisation’s long-standing survey, Health Behaviour in
School-aged Children, which is conducted every four years.7

6
https://www.closer.ac.uk/study/lsype-2/
7
http://www.hbsc.org/

25
Chapter 1 Secondary Schools and Post-16
Participation
Introduction
Young people’s experience of the secondary phase of schooling is crucial to their transition
to adulthood and their long-term welfare. Acquisition of secondary qualifications at GCSE
and A-level or equivalent has become more important as the structure of the economy has
changed to require higher skills, and low-skill jobs have become fewer in number. There is a
close and two-way relationship between struggling in school, or becoming disengaged, and
the other forms of disadvantage described in this report. Disadvantaged young people face
multiple additional barriers in thriving academically.8 And, as will be discussed later in this
report, those who don’t do well in school, or who end up absent or excluded, are over-
represented in many other forms of disadvantage in their teenage years and beyond.
This chapter begins to set the context for the rest of the report by outlining the secondary
schools and post-16 participation policies of Labour, Coalition and Conservatives, alongside
the trends in outcomes over the period. It describes significant changes over time, both in
how schools are resourced and organised, and in what they are asked to prioritise. And it
sets out some striking data showing the scale of the growth in young people staying on in
education after 16, the improvements in attainment at age 16, the forward and backward
steps on attainment gaps, and the continuing problem of some 16- and 17-year-olds still
being out of education, employment or training. The chapter discusses some of the possible
explanations for these trends and looks ahead to the interactions with other outcomes
discussed later in the report.
The chapter will cover:
i. overall priorities
ii. funding for schools and its distribution
iii. teacher workforce and pupil teacher ratios
iv. success measures and league tables
v. structures, roles and responsibilities within the education system
vi. the impact of academies and changes in the local education landscape
vii. the 14-19 phase, GCSEs and vocational options
viii. post-16 education and policy on young people not in education, employment or
training (known as ‘NEET’)
ix. support for special educational needs.
It will then look at outcomes and inequalities focusing on three indicators:

8
Kerris Cooper and Kitty Stewart, Does Money Affect Children’s Outcomes? A Systematic Review (Joseph
Rowntree Foundation, 2013); Whitney Crenna-Jennings, Key Drivers of the Disadvantage Gap: Literature
Review (EPI, 2018).

26
 • attainment at age 16, measured by the proportion achieving ‘Level 2’ (five or more
GCSEs at A* to C, or equivalent)
• the proportion of young people staying on in full-time education after 16
• and the proportion of young people NEET at age 16 and 17.
In relation to these outcomes, we will see that:

• attainment of Level 2 by age 16 increased significantly under Labour and until

2011/12, but then fell and remained lower through until 2018/19

• pupils on free school meals, and pupils with special educational needs but no
statement or education and health care plan improved faster on this measure until
around 2011/12 but then attainment gaps widened again.

• the North East and Yorkshire and Humberside were the regions with the largest
increases in level 2 attainment at 16 between 2001/02 and 2011/12 and the North
East was the region with the largest decline between 2011/12 and 2017/18

• London’s attainment improved before other regions and sustained an advantage.

• the proportion of young people continuing in full-time education after 16 increased

substantially over the period but the increase since 2010 has been slower.

• NEET rates for 16- and 17-year-olds fell from their mid-2000s peak, then after 2015
edged up for boys.
For fuller accounts of education policies and outcomes under Labour, Coalition and
Conservative governments, please refer to studies published as part of CASE programmes.9
Departmental name changes
The Whitehall department responsible for schools has had several name changes during the
period we are studying, sometimes accompanied by changes in its other responsibilities.

1995 – 2001 Department for Education and Employment (DfEE)

2001 – 2007 Department for Education and Skills (DfES)

2007 - 2010 Department for Children, Schools and Families (DCSF)

2010 – present Department for Education (DfE)

9
Ruth Lupton and Polina Obolenskaya, Labour’s Record on Education: Policy, Spending and Outcomes 1997-
2010 (Centre for the Analysis of Social Exclusion, LSE, 2013); Ruth Lupton and Stephanie Thomson, The
Coalition’s Record on Schools: Policy, Spending and Outcomes 2010-2015 (Centre for the Analysis of Social
Exclusion, LSE, 2015); Ruth Lupton and Polina Obolenskaya, The Conservatives’ Record on Compulsory
Education : Spending , Policies and Outcomes in England , May 2015 to Pre-COVID 2020 (Centre for the Analysis
of Social Exclusion, LSE, 2020).

27
Labour policies 1997–2010: overview
Overall priorities
As detailed below, the Labour government’s education policy was marked out by large year-
on-year increases in school funding, significant increases in the school workforce, hands-on
intervention to improve school effectiveness, and quantified targets for improvement
applying to schools, local authorities and the government itself. Labour devoted
considerable attention to making schools work for pupils from disadvantaged backgrounds,
through the targeting of its funding and area initiatives, as well as national policies such as
extended schools, curriculum flexibility, the promotion of vocational qualifications,
initiatives to reduce absence and exclusion, and support for young people to stay on in
education after 16.
Funding for schools, and its distribution
The 1997 Labour government was elected on a manifesto with education at its centre, but
its principal headline pledge related to primary school class sizes. Once in government, a
more ambitious and developed agenda began to unfold, including significant change for
secondary age pupils.
This was evident both in spending totals and in their allocation. Over three terms of office,
Labour implemented significant real terms increases in spending on both primary and
secondary education. Between 1999/2000 and 2009/10 spending per pupil in primary and
secondary schools increased by around 6 per cent in real terms annually (78 per cent over
the entire period).10
Within overall funding, area-based programmes and specific grants directed extra help to
poor areas. An early example of such a targeted programme for secondary schools was
‘Excellence in Cities’ which targeted extra resources at schools in disadvantaged urban
areas, to pay for the development of learning support units, learning mentors, and
educational enrichment programmes. Excellence in Cities began in 1999 with 25 local
authority areas, including all 12 Inner London boroughs, and was extended in two further
phases to cover a third of secondary schools in England. Evaluation showed a positive,
albeit small, improvement in pupil attainment, and a strong reduction in absences. In
2000/01 its budget was £139 million rising to some £300 million by 2002/03.11
Another example of an initiative aimed at disadvantaged areas was the ‘Extended Schools’
initiative. Launched in 2003, the aim of this programme was to support the development of
schools that provided a comprehensive range of services including access to health services,
adult learning and community activities, as well as study support and 8:00 am to 6:00 pm
childcare. By the third year of the initiative nearly 150 schools were involved, of whom two
thirds were secondary schools. The evaluation found that the approach was impacting
positively on pupils’ attainment, with the strongest impacts in the case of pupils facing

10
Luke Sibieta, School Spending in England : Trends over Time and Future Outlook (IFS, 2021).
11
Stephen Machin, Sandra McNally, and Costas Meghir, ‘Improving Pupil Performance in English Secondary
Schools: Excellence in Cities’, Journal of the European Economic Association, 2.2–3 (2004), 396–405; Lesley
Kendall and others, Excellence in Cities: The National Evaluation of a Policy to Raise Standards in Urban Schools
2000-2003 (DfES, 2005).

28
difficulties. The schools were also having a range of other impacts including on engagement
with learning and family stability. The cost-benefit analysis suggested that the approach
represented a good investment, and in 2005 the government committed to extend the
approach nationwide.12 By 2010, two thirds of schools were offering all five elements of the
full core offer (childcare, a varied menu of activities, parenting support, community
facilities, and specialist support), and the remaining third were offering some elements.13
These two examples were two among many funding streams intended to modernise schools
and tackle low attainment, the overall effect of which was strongly redistributive and
became more so as time went on. The Institute for Fiscal Studies analysed the overall
quantum of school funding and how it changed during the mid-2000s. It found that in
2003/04, for every pound of funding allocated universally to secondary schools (via the
basic per-pupil amount), an additional 61 per cent was allocated for pupils eligible for free
school meals. By 2006/07 this extra had risen to 77 per cent. A second study found that this
implicit premium for free school meals pupils grew three times faster than the overall
increase in per pupil funding between 2005/06 and 2008/09. 14
Teacher workforce
Extra schools funding provided the resources for an expansion of the teacher workforce. By
2004, maintained secondary schools had 21,600 more teachers and 17,500 more teaching
assistants than in 1997. This represented an 11 per cent increase in teacher numbers, and
more than trebled the number of teaching assistants. (Figure ED1)
Figure ED1: Trends in workforce in maintained secondary schools, England, 1997 to 2010
400,000

350,000

300,000

250,000

200,000

150,000

100,000

50,000

0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

Teachers (full time equiv) Teaching assistants (full time equiv) Other support staff (full time equiv)

Source: DCSF / DfE, School Workforce in England, 2007 and 2010 releases. *Figures do not include staffing in academies and
city technology colleges

12
C Cummings and others, Evaluation of the Full Service Extended Schools Initiative: Final Report (DfES, 2007).
13
Hannah Carpenter and others, Extended Services Evaluation: End of Year One Report (DfE, 2010).
14
Luke Sibieta, Haroon Chowdry, and Alastair Muriel, Level Playing Field? The Implications of School Funding
(IFS, 2008); Haroon Chowdry, Ellen Greaves, and Luke Sibieta, The Pupil Premium: Assessing the Options
(Institute for Fiscal Studies, 2012).

29
There were new policies to attract teachers – including bursaries for teacher training
courses, ‘golden hellos’ to attract applicants, and support for Teach First, a not-for-profit
company which brought graduates from top universities to the most challenging schools. In
areas with high housing costs, key worker housing programmes were instituted.
The rise in the workforce, combined with the slight fall in secondary school pupil numbers
meant that pupil teacher ratios improved considerably from 14.5 pupils per adult in 1997, to
10.4 pupils per adult in 2010. (Figure ED2, later in this chapter).
Surveys of newly qualified teachers (NQTs) one year into the profession were conducted
from 2003 onwards, and there were substantial increases in new teachers’ satisfaction. The
number of NQTs saying they had been well prepared to manage behaviour rose from 60 per
cent in 2003 to 69 per cent in 2010. Over the same period satisfaction with preparation to
teach learners from all ethnic backgrounds rose from 32 per cent to 44 per cent, and
satisfaction with preparation to work with special educational needs rose from 46 per cent
to 57 per cent. 15
Success measures and league tables
In its first spending review in 1998, the Labour government announced quantified outcome
targets for departments alongside their spending review settlements. These national targets
were known as Public Service Agreements or PSAs, and progress was publicly reported and
closely monitored within government.16 For education, throughout its time in government,
Labour had targets for the percentage of the population achieving certain standards at
primary and secondary level. For example, the 1998 PSAs included these targets for
secondary achievement:

• an increase in the proportion of those aged 16 who achieve one or more GCSEs at
grade G, or equivalent, from 92 to 95 per cent by 2002.
• also by 2002, an increase from 45 to 50 per cent in the proportion of those aged 16
who achieve Level 2.17
The choice of these targets reflected a desire to improve attainment across the spectrum of
ability. Later in the government’s period in office, objectives were modified to have an
explicit focus on the parts of the country, and indeed the schools, which had the lowest
performance at age 16. In 2000, these ‘floor targets’, as they became known, were set at
the level of the Local Education Authority (at least 38 per cent of pupils were to achieve
Level 2 in every LEA by 2004).18 This was then narrowed down to school level (in 2004 the
PSA set the ambition that by 2008, 30 per cent of pupils in every school were to achieve
Level 2).19

15
Department for Education, Newly Qualified Teachers : Annual Survey 2013 (DfE, 2013).
16
Michael Barber, Instruction to Deliver: Tony Blair, Public Services and the Challenge of Achieving Targets
(London: Politico’s, 2007).
17
HM Treasury, Public Services for the Future : Modernisation, Reform, Accountability (The Stationery Office,
1998).
18
HM Treasury, Prudent for a Purpose : Building Opportunity and Security for All (HMT, 2000).
19
HM Treasury, 2004 Spending Review Public Service Agreements 2005-2008 (HMT, 2004).

30
In parallel, Labour introduced ‘value-added’ measures in school league tables to take more
account of differences between the intakes of different schools. This started by adjusting
purely for the prior attainment of the school’s intake at Key Stage 2, then broadened to
‘contextual value added’ which included data on the pupils’ broader characteristics,
including age, gender, ethnicity, socioeconomic status and special educational needs.20
The government’s aspirations for attainment in secondary schools sat alongside similar
aspirations and policies in primary schools and in the early years (including the development
of Sure Start). They were complemented by wider policies designed to address the factors
that contributed to, and entrenched, educational underperformance. These included family
poverty, special educational needs, school absence, drug and alcohol use, and the UK’s
unusually low rates of staying on in education after 16. These issues are covered in other
chapters of this report, with the exception of special educational needs and post-16
participation, both of which are discussed further below.
Structures, roles and responsibilities within the education system
The role of central government
Under Labour, Ministers took responsibility for national targets, and Whitehall departments
were reshaped to develop and roll out initiatives that would help the frontline to improve
outcomes. In education, this work was driven by the Standards and Effectiveness Unit in the
Department for Education and Skills, under Sir Michael Barber, through ‘National Strategies’
which covered multiple aspects of primary and secondary school improvement – literacy
and numeracy, school leadership, attendance and behaviour management, and teacher
recruitment and training.21 This degree of involvement with the frontline, and accountability
for frontline performance, was a significant change from the way the department had seen
its role in the past and was subsequently extended to other central government
departments, alongside the creation of a central Delivery Unit in No 10. 22
Pressure and support
This increase in ambition brought extra resources for schools but also carried a cost for
schools and their leaders who often faced overlapping questions, advice and scrutiny from
Whitehall, local authorities and Ofsted. School leaders put pressure on the government to
deliver its objectives with more empowerment and less bureaucracy. The ‘London
Challenge’, established in 2003, offered a new model of reform, based on a more equal
partnership, bringing central and local government together with families of schools to raise
performance across the whole city. 23 in 2008 the model was extended also to Greater
Manchester and the ‘Black Country’ area of the West Midlands. In 2013, Ofsted summarised

20
George Leckie and Harvey Goldstein, ‘The Evolution of School League Tables in England 1992–2016:
“Contextual Value-Added”, “Expected Progress” and “Progress 8”’, British Educational Research Journal, 43.2
(2017), 193–212.
21
Department for Education, The National Strategies 1997–2011: A Brief Summary of the Impact and
Effectiveness of the National Strategies (DfE, 2011).
22
Nehal Panchamia and Peter Thomas, Public Service Agreements and the Prime Minister’s Delivery Unit
(Institute for Government, 2014).
23
Marc Kidson and Emma Norris, Implementing the London Challenge: Beyond Top-down and Bottom-Up
(Institute for Government, 2014).

31
the evaluations of the London and City challenges as ‘pointing to measurable improvements
in reducing the number of underperforming schools, increasing the number of good or
outstanding schools and raising educational attainment for disadvantaged pupils. 24
In 2004 the government also announced a ‘New Relationship with Schools’ so that in future
there would be a single annual review with the school on its performance, combining the
interest of both Local Authorities and the Department. Local authorities would be required
to appoint a ‘school improvement partner’ within a system of national training and
guidance: many of these school improvement partners were serving or recently retired
headteachers.25
Both local authorities and schools were expected to improve and challenged if they did not
make enough progress: this was seen as important from an equity point of view, as poorly
performing schools often served more disadvantaged communities. 26 The ‘academy’
model, developed by Labour after 2000, became part of this challenge from 2008, with the
threat of closure and replacement with an academy if schools did not meet floor targets.
Impact of academies: initial phase
Evidence suggests that the pre-2010 academies programme was successful in raising
performance. The National Audit Office reviewed progress in 2010 and found that the
proportion of pupils achieving Level 2 in secondary academies improved at a faster rate than
maintained schools with similar intakes, and absence rates also fell faster than comparators.
However, disadvantaged pupils in academies saw less improvement than pupils who were
less disadvantaged.27 A later review conducted jointly by the Education Policy Institute and
LSE economists found strong positive effects on GCSE attainment for pre-2010 academies,
of around one grade in each of five GCSE subjects four years after conversion, but with
considerable variation amongst the group, and performance in some academies
deteriorating. 28
Special Educational Needs
Special educational needs is a significant but often ignored aspect of education policy, highly
relevant to disadvantaged groups. During its first parliament, Labour introduced a new
Special Educational Needs Code of Practice, designed to increase the attention devoted to
the early identification of needs. It established three levels of intervention: School Action,
School Action Plus, and the SEN statement. 29 The first of these, as its name implies, was
organised within the school, the second drew on external help, and the third involved
specialist support arranged by a local authority. Alongside the new Code, other policies
encouraged greater identification of special needs. For example, as noted above, changes to

24
Merryn Hutchings and others, Evaluation of the City Challenge Programme (DfE, 2012); Ofsted, Unseen
Children Evidence Report: Access and Achievement 20 Years On (Ofsted, 2013).
25
Department for Education and Skills, Five Year Strategy for Children and Learners (DfES, 2004).
26
Pam Sammons, ‘Zero Tolerance of Failure and New Labour Approaches to School Improvement in England’,
Oxford Review of Education, 34.6 (2008), 651–64.
27
National Audit Office, The Academies Programme (NAO, 2010).
28
Andrew Eyles and others, ‘The Impact of Pre-2010 Sponsored Academies on Educational Attainment’, in The
Impact of Academies on Educational Outcomes, ed. by Jon Andrews and Natalie Perera (EPI, 2017).
29
Department for Education and Skills, Special Educational Needs - the Code of Practice (DfES, 2001).

32
league tables (‘contextual value added’) began to take more account of pupil characteristics,
including the school’s proportion of children with special educational needs.30
The proportion of children recorded with some level of special educational needs fluctuated
over the years that followed, but from 2003 onwards the numbers grew substantially, from
13 per cent of secondary pupils to a peak of 19.7 per cent in 2010. (Figure ED 4, later in this
chapter.) Ofsted found in a 2010 review that achievement for children with special
educational needs and disabilities was good or outstanding in 41 per cent of the provision
they visited, but it was inadequate in 14 per cent, and there was continuing evidence of
considerable inconsistency between areas.31
The 14 to 19 phase
Many of the changes in secondary education had as their goal to help young people from
disadvantaged backgrounds to stay engaged and do well in their exams. But the legal age at
which young people could leave school was 16, and educational participation rates for
teenagers were lower in the UK than in other comparable European countries.32 The
government wanted to address that and reduce the proportion of young people not in
education, employment or training at 16 and 17.
These linked policy concerns drew attention to the difficulties many young people faced as
they navigated the years before and after 16. In school the essential components of the
phase at 14+ were about choosing and taking GCSEs and other qualifications, starting to
think about careers, and deciding about post-16 plans. But these years were also the peak
years for absence and exclusion, so the young people who most needed help were often not
in school to receive it. The educational challenges of 14+ coincided with the biological and
social changes of the teenage years, with increased experimentation including alcohol, sex
and drugs, and for some young people, looming pressures to get a job, move out of the
family home, or leave care. It became progressively clearer to the government that
navigating these years was hardest for the most disadvantaged, and that improving
attainment and post-16 participation was going to require a much more coherent approach
to supporting young people through the process.
Several different policy processes were in play here. The Social Exclusion Unit published a
report on young people not in education, employment or training in 1999. 33 34 Some of the
policy measures introduced after that are discussed below. But the SEU report was just one
element of a wider change in mainstream education policy as it began to focus on the need

30
Leckie and Goldstein; Anthony Kelly and Christopher Downey, ‘Value-Added Measures for Schools in
England: Looking inside the “Black Box” of Complex Metrics’, Educational Assessment, Evaluation and
Accountability, 22.3 (2010), 181–98.
31
House of Commons Education and Skills Committee, Special Educational Needs: Third Report of Session
2005–06 (HoC, 2005); Brian Lamb, Lamb Inquiry: Special Educational Needs and Parental Confidence (DCSF,
2009); Ofsted, The Special Educational Needs and Disability Review (Ofsted, 2010).
32
OECD Centre for Educational Research and Innovation, Education Policy Analysis 1998 (OECD, 1998).
33
Social Exclusion Unit, Bridging the Gap: New Opportunities for 16-18 Year Olds Not in Education,
Employment or Training (Stationery Office, 1999).
34
This is a report I was involved in developing, as Director of the Social Exclusion Unit at the time.

33
for the school experience to change to be more flexible to the needs and interest of the
whole cohort.35
GCSEs, including vocational options
The content and delivery of GCSEs was an important part of this process. Several
government strategies set out ambitions to increase curriculum flexibility.36 In 1998,
schools became able to disapply the National Curriculum and set aside up to two three
subjects so that a student could follow an extended work-related learning programme. From
2002 the ‘Increased Flexibility for 14- to 16-Year-Olds’ Programme (IFP) allowed 2000
schools and around 40,000 year 10 pupils to access a more vocational and practical
curriculum organised through partnerships usually led by an FE college. 37 GCSEs in
vocational subjects were introduced in September 2002, replacing a previous qualification
(the GNVQ). The new GCSEs covered subject groupings such as Applied Art and Design,
Engineering, Health and Social Care, Manufacturing, and Leisure and Tourism. Over the
course of around a decade, vocational qualifications came to represent a significant
proportion of the achievement of many young people at age 16.
Policy on increasing participation and reducing NEETS
As secondary school attendance and attainment improved during the 2000s, more young
people had the academic qualifications to carry on in school after the age of 16. But this left
other issues to be addressed – lack of support and guidance about post-16 options, lack of
post-16 places, and financial barriers.
Connexions and careers advice
One change, recommended by the Social Exclusion Unit in 1999, was the introduction of a
youth support service.38 This was agreed by the government, and implemented by the
Department of Education with the creation of the ‘Connexions’ service. Connexions was
piloted, then rolled out nationally on a phased basis from 2001. It was built on a Personal
Advisor model and worked with a range of agencies to offer a one-stop advice, support and
guidance service to young people before and after the end of compulsory education. It
subsumed the Careers Service, but had double the budget to enable it to offer more help.
By 2004 it was staffed by over 7,700 Personal Advisers with an annual budget of £492
million which remained at around that level for most of the rest of the decade. 39 40

35
Department for Education and Skills, Five Year Strategy for Children and Learners.
36
Department for Education and Skills, 14–19: Extending Opportunities, Raising Standards (DfES, 2002);
Department for Education and Skills, Five Year Strategy for Children and Learners.
37
Sarah Golden and others, Evaluation of Increased Flexibility for 14 to 16 Year Olds Programme: Outcomes for
the Second Cohort. Research Report No. RR786 (DfES, 2006).
38
Social Exclusion Unit, Bridging the Gap: New Opportunities for 16-18 Year Olds Not in Education,
Employment or Training.
39
National Audit Office, Connexions Service: Advice and Guidance for All Young People (NAO, 2004).
40
Department for Children Schools and Families, Departmental Report 2009 (DCSF, 2009). p216.

34
 Evaluations of Connexions

Qualitative evaluation of the Connexions pilots in 2001 found that:

• 96 per cent of clients said they found Personal Adviser sessions useful.
• 87 per cent said they now did something different as a result, including improved
attendance and motivation.
• 79 per cent said the help had made them more interested in education and
training.
• 75 per cent said it had helped them cope with other problems.
• 77 per cent said it was better than the support they had received in the past. 41

An NAO report on Connexions in 2004 found that the Connexions Service had made good
progress in improving the way that young people received advice and guidance, that
young people who had had contact with Connexions personal advisors rated the service
they received highly, and that schools were confident in the work that Connexions did
with young people who needed the most help. However only half of schools surveyed
were satisfied with the level of response to the needs of other young people in schools. 42

A 2004 study on how Connexions operated highlights the wide-ranging issues it found
itself dealing with. This study found that ‘many of the young people in the sample faced
multiple risks in their lives and needed intensive attention’ and that the impact of work
with Connexions was multi-faceted, ‘involving outcomes in different areas of young
people’s lives, including personal development and dealing with urgent or underlying risks
as well as destination outcomes in education, employment or training.’43 A later study of
its operation found that, for some young people faced with complex and challenging
circumstances, the relationship with their personal advisor provided ‘a uniquely stable
and valued source of support’.44

Education Maintenance Allowance

To tackle financial disincentives to staying on in education, the government introduced a
new means-tested payment, the Education Maintenance Allowance (EMA) – paid weekly
direct to lower-income young people in education or unwaged training at three income-
related levels. Evaluation of the pilots estimated that it had increased the proportion in full-
time education at both age 16 and 17 by 6.1 percentage points, and by more for young
people from the least well-off socio-economic groups.45 The EMA was rolled

41
Peter Dickinson, Lessons Learned from the Connexions Pilots (DfES, 2001).
42
National Audit Office, Connexions Service: Advice and Guidance for All Young People.
43
Liz Hoggarth, D.I. Smith, and G. Britain, Understanding the Impact of Connexions on Young People at Risk
(DfES, 2004).
44
Kieron Sheehy, Rajni Kumrai, and Martin Woodhead, ‘Young People’s Experiences of Personal Advisors and
the Connexions Service’, Equality, Diversity and Inclusion: An International Journal, 30.3 (2011), 168–82.
45
Sue Middleton, Kim Perren, and others, Evaluation of Education Maintenance Allowance Pilots: Young
People Aged 16 to 19 Years (DfES, 2005).

35
out nationally to 16-year-olds in academic year 2004/05, 17-year-olds in academic year
2005/06 and 18-year-olds in 2006/07. By 2008 the total cost was £532 million a year.46
Education and training provision
Supply and access to suitable post-16 courses was another issue. From September 2007, the
government introduced the ‘September guarantee’ to offer every 16-year-old an
appropriate place in full or part-time education or training. The following year, this was
extended to 17-year-olds, to ensure that young people on short courses or who dropped out
during the year had an opportunity to re-engage. 47 In November 2009, a January guarantee
was introduced to address the same issue of young people who had not found or stayed in a
suitable course.48
NEET targets and raising the participation age
The government reflected all these aspirations in a PSA target, set in 2004, to reduce the
proportion of 16- to 18-year-olds who were NEET from 10 per cent to 8 per cent between
2004 and 2010.49 Looking further ahead, in March 2007, the government proposed a longer-
term change to require 16- and 17-year-olds to participate in education or training.50 This
idea became law in the Education and Skills Act 2008, with implementation scheduled for
the generation that would turn 16 in 2013. This legislation was implemented, with some
changes, by the Coalition government which took office in 2010.

Coalition and Conservatives policies 2010 to 2019: overview

Overall priorities
The Coalition agreement negotiated by the Liberal Democrats and Conservatives in May
2020 set clear aspirations in relation to schools: tackling educational inequality, high
standards of discipline, robust standards, high quality teaching, and helping parents,
community groups and others to start new schools.51 The defining policies of this decade
for secondary schools were funding constraints, academisation, reduction of the role of
local authorities, and multiple changes to Key Stage 4 qualifications, alongside
implementation of the new legal participation age of 18.
Funding for schools and its distribution
Coalition
Relative to the deep cuts made in some other Whitehall budgets, schools enjoyed some
protection in the 2010 spending review. A new pupil premium was introduced in funding
allocations to reflect the proportion of pupils on free school meals. For pupils in years 7-11,

46
John Clark and Paul Simmonds, Evaluation Study to Assess the Economic Impact of ESRC Research: Case
Study of the Education Maintenance Allowance (Technopolis, 2010).
47
Department for Education and Skills, Departmental Report 2007 (DfES, 2007); Department for Children
Schools and Families, Departmental Report 2009.
48
House of Commons Children Schools and Families Committee, Young People Not in Education, Employment
or Training (HoC, 2010).
49
HM Treasury, 2004 Spending Review: New Public Spending Plans 2005 - 2008 (HMT, 2004); HM Treasury, PSA
Delivery Agreement 14: Increase the Number of Children and Young People on the Path to Success (HMT, 2007).
50
Department for Education and Skills, Raising Expectations: Staying in Education and Training Post-16 (DfES,
2007).
51
HM Government, The Coalition: Our Programme for Government, 2010.

36
the premium started as a flat rate of £430 in respect of children currently in receipt of free
school meals (2011/12). By 2014/15 it was paid at a higher level of £935, for children who
had been in receipt of free school meals at any point in the last six years. 52
Overall, however, the headroom in school funding was tilted towards primary rather than
secondary schools: spending per pupil rose in real terms for primary schools between
2011/12 and 2015/16 (by 7 per cent) but spending per pupil in secondary schools fell by 3
per cent in real terms.53 The pupil premium did not protect all schools with very
disadvantaged intakes: the National Audit Office found that the effect of the premium was
sometimes outweighed by the loss of other funding, with the result that the per-pupil
funding of 16 per cent of the most disadvantaged secondary schools fell by over 5 per cent
in real terms between 2010-11 and 2014-15.54
Conservatives
The squeeze on secondary spending tightened after 2015 when the Conservatives governed
on their own. Between 2015–16 and 2019–20 there was a further 9 per cent real terms fall
in secondary school spending per pupil overall. Within that the distribution of resources
was unfavourable to deprived areas: the most deprived fifth of schools saw the largest
reduction, a 14 per cent real terms drop in per pupil funding, for a variety of reasons
including a new funding formula.55 The pupil premium was not increased from its 2014/15
level until 2020/21. In 2016 and 2017, while Justine Greening was Secretary of State, twelve
areas of the country were selected for a new initiative, ‘Opportunity Areas’, which was
targeted at social mobility and included funding of £72 million over three years. As of 2020,
the Education Policy Institute found mixed results at Key Stage 4 for these areas: the
disadvantage gap had fallen in seven opportunity areas, but risen in five.56
Teacher workforce and pupil teacher ratios
With reduced funding, it was inevitable that the number of teachers and other staff would
fall. Figure ED2 below shows that in secondary schools the peak in teacher numbers was in
2012/13, and the peak in teaching assistants was 2013/14. By 2018/19 there were around
11,000 fewer teachers and 7,000 fewer teaching assistants in state funded secondary
schools than in 2011/12. A Sutton Trust Survey in 2019 found that 70 per cent of senior
leaders in secondary schools had had to cut teaching staff or other staff for financial
reasons, 47 per cent had had to cut subject choices at GCSE, and 40 per cent at A level, and
27 per cent of senior leaders said that pupil premium funding was being used to plug gaps in
the schools budget.57

52
Tim Jarrett, Robert Long, and David Foster, School Funding and the Pupil Premium (House of Commons
Library, 2015).
53
Jack Britton, Christine Farquharson, and Luke Sibieta, 2019 Annual Report on Education Spending in England:
Schools (Institute for Fiscal Studies, 2019). Takes account of the pupil premium, the effect on schools of 16-19
funding cuts and also of reductions in Local Authorities’ ability to spend on schools.
54
National Audit Office, Funding for Disadvantaged Pupils (NAO, 2015).
55
Sibieta, p10 and p12; see also National Audit Office, School Funding in England Department for Education
(NAO, 2021).
56
Jo Hutchinson, Mary Reader, and Avinash Akhal, Education in England: Annual Report 2020 (EPI, 2020).
57
Sutton Trust, School Funding and the Pupil Premium: NFER Teacher Voice Survey (Sutton Trust, 2019).

37
There were recruitment and retention problems too. Recruitment targets were missed
every year from 2013/14 onwards.58 Retention rates for newly qualified teachers fell, and
the overall number of teachers leaving the profession before retirement age increased from
25,000 in 2010-11 to 36,000 in 2016-17.59
Figure ED2: Trends in workforce in state-funded secondary schools, England, 2010/11 to
2018/19
400,000

350,000

300,000

250,000

200,000

150,000

100,000

50,000

0
2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 2018/19

Teachers (full time equiv) Teaching assistants (full time equiv) Other support staff (full time equiv)

Source: DCSF / DfE, School Workforce in England, 2019

The fall in school staffing after 2013 occurred just as the baby boom of the early 2000s was
approaching secondary school age.60 The pupil : teacher ratio began to worsen after 2012,
as shown in Figure ED3 below. By 2019 the pupil : teacher ratio was the worst it had been
in two decades. Pupil : adult ratios also deteriorated after 2010.

58
Beng Huat See and Stephen Gorard, ‘Why Don’t We Have Enough Teachers?: A Reconsideration of the
Available Evidence’, Research Papers in Education, 35.4 (2020), 416–42.
59
Jack Worth, ‘Latest Teacher Retention Statistics Paint a Bleak Picture for Teacher Supply in England - NFER’
(National Foundation for Educational Research, 2018) <https://www.nfer.ac.uk/news-events/nfer-blogs/latest-
teacher-retention-statistics-paint-a-bleak-picture-for-teacher-supply-in-england/>.
60
Department for Education, Schools, Pupils and Their Characteristics: January 2019 (DfE, 2019).

38
Figure ED3: Trends in pupil: teacher and pupil: adult ratios in England, maintained / state-
funded secondary schools
20
17.2 17.6
18 16.7
15.5
16
Pupils per teacher / adult*

14
12 14.5 14.5

10 11.9
10.4
8
6
4
2
0

Pupil Teacher Ratio (qualified teachers)

Pupil Teacher Ratio (qualified and unqualified)
Pupil adult ratio (includes teachers and support staff excluding administrative/clerical)

* Figures before 2010 are LA maintained and do not include academies. Figures after 2010 cover all state-funded secondary schools
and do include academies. Source: DfE / DCSF, School workforce in England, 2007, 2010 and 2019 releases

One factor in recruitment and retention is workload and low wellbeing in the teaching
profession. An Ofsted survey in 2019 found that teachers’ satisfaction with life was lower
than that of the general public, and that the main causes of heavy workload were the
volume of administrative tasks, marking, staff shortages, lack of support from external
specialist agencies, challenging behaviour of pupils, changes to external examinations,
frequently changing government policies and regulations, and in some cases, lack of skills or
training.61
Success measures and league tables
The Coalition and Conservative governments introduced multiple changes in the Key Stage 4
framework (discussed in more detail below) which changed the qualifications that could be
counted in measuring performance, as well as what was taught and how it was assessed.
This made it hard to compare trends over time. In any event, from 2010 onwards, the
government did not set quantified targets for the educational outcomes expected at
national level. What the government took credit for, after five years in office, was school
inspection ratings. At the 2015 election, both Conservative and Liberal Democrat manifestos
of 2015 used as a measure of school improvement that there were now a million more
pupils in schools rated by Ofsted as ‘good’ or ‘outstanding’.62

61
Ofsted, Teacher Well-Being at Work in Schools and Further Education Providers (Ofsted, 2019).
62
Liberal Democrats, Manifesto 2015: Stronger Economy. Fairer Society. Opportunity for Everyone (LIberal
Democrats, 2015); Conservative Party Manifesto 2015 (Conservative Party, 2015).

39
After 2012, however, referring to Ofsted rating had become an increasingly unreliable
measure, as the Ofsted inspection regime was scaled back to end routine inspections of
schools rated outstanding. 63 This, and other changes to the inspection regime, meant that
by August 2017, 1,620 schools (secondary and primary) had not been inspected for 6 years
or more, and 296 ‘outstanding’ schools had not been inspected for at least 10 years. 64 The
exemption for outstanding schools was lifted in 2020. Of those formerly outstanding schools
reinspected in 2021/22, over 80 per cent did not retain the outstanding grade. The majority
were judged to be good, but 17 per cent were rated ‘requires improvement’, and 4 per cent
were rated ‘inadequate’. The average time since last inspection was 13 years. 65
In school level performance information, ‘contextual value added’ measures were dropped,
the government arguing that taking account of the socio-economic characteristics of a
school’s intake in inter-school comparisons would entrench low expectations for deprived
pupils.66 After 2016, the Conservative government brought in a new headline progress
measure for schools, ‘Progress 8’. This was a relative measure (comparing pupils with other
pupils) so did not measure what progress was being made nationally. In comparing schools,
it took no account of pupils’ background characteristics other than their attainment at Key
Stage 2. 67
Roles and responsibilities within the system
Coalition and Conservative governments put significant effort into the expansion of school
autonomy, through new powers to set up free schools and a much broader push to
academisation. The Academies Act 2010 enabled all schools to apply to become an
Academy (so-called ‘converter academies’) with outstanding schools being fast-tracked,
followed by those that were ‘performing well’, and other schools if they joined a chain. The
‘sponsored’ academy model also remained in place as a required solution for low-
performing schools.
The scale of the change during the Coalition years was dramatic, as Figure ED4 shows. In
two years, the proportion of secondary schools that were academies grew from 11 per cent
to 50 per cent.
The National Audit Office estimated that the additional cost of the programme in the first
two years was £1 billion, covering items such as central programme administration,
transition costs, academy insurance, support for academies in deficit, and reimbursing
academies’ VAT costs. As the programme had run ahead of forecast, the department had to
reallocate money from other budgets to accommodate it. 68 In March 2016, the
Conservative Green Paper set the ambition that by the end of 2020, all schools would be

63
Ofsted, The Framework for School Inspection (Ofsted, 2012).
64
National Audit Office, Ofsted’s Inspection of Schools (NAO, 2018).
65
Ofsted, ‘Hundreds of Formerly Outstanding Schools Reinspected’, 2022
<https://www.gov.uk/government/news/hundreds-of-formerly-outstanding-schools-reinspected> [accessed 7
January 2023].
66
Department for Education, The Importance of Teaching (DfE, 2010).
67
Leckie and Goldstein.
68
National Audit Office, Managing the Expansion of the Academies Programme (NAO, 2012).

40
academies or in the process of becoming academies, but subsequently stepped back from
this policy in the face of opposition.69
Figure ED4: Trends in percentage of state-funded secondary schools that are academies,
England

100
90
80 72
65 68
70 61
57
60 50
Per cent

50
36
40
30
20 11
4 6
10 0 0 0 1 1 2
0

Source: NAO, Converting maintained schools to academies, 2018

Research into the motivation of schools that became academies by their own choice reveals
a range of motivations. A 2012 survey by the think tank Reform and the Specialist Schools
and Academies Trust found that 78 per cent of schools chose to become an academy at
least partly because they thought they would receive additional funding and 39 per cent
said this was the main reason for conversion. The other main reasons cited were financial
and educational autonomy, and freedom to buy services from providers other than the local
authority.70
In the early years of academisation many schools became academies on a standalone basis.
But over time, maintained schools converting to academies became more likely to form or
join multi-academy trusts (MATs), and the government encouraged this. So, while in
2011/12 and 2012/13 respectively 70 per cent and 43 per cent of new academies were
standalone, by 2016/17 only 5 per cent of new academies were standalone. 71 When a
school is in a MAT, it is the Trust that is the legal entity with responsibility for governance of
the school. By 2022, the largest MATs were chains which had upwards of 50 schools, but the
majority of academies were in trusts of 9 or fewer schools.72 Some studies have found that
the process of joining a MAT has impacted on the sense of autonomy that originally

69
Department for Education, Educational Excellence Everywhere (DfE, 2016).
70
Dale Bassett and others, Plan A + Unleashing the Potential of Academies (Reform and The Schools Network,
2012).
71
National Audit Office, Converting Maintained Schools to Academies (NAO, 2018).
72
Nerys Roberts, Schools White Paper March 2022 (House of Commons Library, 2022).

41
motivated schools to convert, although some school leaders felt this was necessary trade-
off with the support they gained from being in a MAT.73
Impact of academisation second phase, and impact of multi-academy trusts
Academies policy is often seen as a point of continuity between the pre-2010 Labour
administration and the Coalition and Conservative administrations that followed. But the
academisation programme introduced by the Coalition (‘converter academies’) was very
different from the sponsored academy model that operated pre-2010. The schools involved
were also very different, with the converter academies having generally much higher
attainment and lower proportions of pupils eligible for free school meals.74 Research has
found that the effect of converter academies on GCSE attainment was ‘far smaller than the
effects of the pre-2010 sponsored academies and is, in some cases, undetectable’. 75
Analysis of the impact of academies’ membership of multi-academy trusts found that ‘taken
in aggregate there appears to be little difference in the improvement seen in schools within
local authorities and schools within multi-academy trusts’.76 Ofsted has no powers to
inspect multi-academy trusts. Since 2018 it has introduced ’summary evaluations’ of MATs,
but does not have the power to insist that trusts engage with summary evaluations. Ofsted
research has found ambiguity in the school system about the role of MATs with trust
leaders’ conception of their responsibilities ranging from ‘accountable for everything’ to
‘not accountable for anything’.77
Local authorities’ support for schools
This substantial move to academisation coincided with a significant squeeze on local
government funding in general, and for children’s services and prevention (see Chapter 3) as
well as new requirements for local authorities to delegate budgets and move to traded
services. This hollowed out the support that local authorities could provide for schools to
help them support the most disadvantaged young people. A 2014 DfE-funded study of
change in local education systems found that support for vulnerable pupils was the area
where the new landscape was working least well, with 44 per cent of school leaders saying
there was not the provision in their school and across the local area to ensure vulnerable
children received a high quality education.78 More recent DfE-funded research (2021)
analyses in some detail the misalignment between, on the one hand, local authorities’
responsibilities in relation to providing sufficient school places, managing high needs
budgets, and securing alternative provision, and, on the other, the perverse financial

73
Ofsted, Multi-Academy Trusts: Benefits, Challenges and Functions (Ofsted, 2019); Greg Thompson, Bob
Lingard, and Stephen J. Ball, ‘“Indentured Autonomy”: Headteachers and Academisation Policy in Northern
England’, Journal of Educational Administration and History, 53.3–4 (2021), 215–32.
74
Andrew Eyles, Stephen Machin, and Olmo Silva, ‘Academies 2 – The New Batch: The Changing Nature of
Academy Schools in England’, Fiscal Studies, 39.1 (2018), 121–58.
75
Jon Andrews and Natalie Perera, The Impact of Academies on Educational Outcomes (EPI, 2017). p7.
76
Jon Andrews, School Performance in Multi-Academy Trusts and Local Authorities - 2015 (EPI, 2016); For a
similar analysis with similar conclusions see also Daniele Bernardinelli and others, Multi-Academy Trusts: Do
They Make a Difference to Pupil Outcomes, 2018.
77
Ofsted, Multi-Academy Trusts: Benefits, Challenges and Functions.
78
Leigh Sandals and Ben Bryant, The Evolving Education System in England : A ‘Temperature Check’. Research
Report, July 2014 (DfE, 2014).

42
incentives on schools to be ‘isolationist’. It set out how few levers exist for local authorities
to challenge this since most of the power lies in the hands of multi-academy trusts, Regional
Schools Commissioners, or indeed the Secretary of State.79
Free schools
Between 2011 and 2019, over 500 ‘free schools’ were opened in England (including both
primary and secondary). This programme is led by applications rather than a national needs
assessment and as a result has not always matched the profile of where more places are
needed. This phenomenon was noted early on in the programme’s life and has continued to
be an issue: analysis by the Education Policy Institute found that between 2016 and 2018
secondary free schools had added 4 places per 1,000 pupils in the areas of greatest demand
but 15 places per 1,000 pupils in areas where there was excess capacity.80
GCSEs and other Key Stage 4 qualifications
The Coalition government set in hand multiple changes to the framework for GCSEs and
other Key Stage 4 qualifications, covering performance measures, curriculum and
assessment. One change of particular significance was the decision that from summer 2014,
a number of qualifications, mainly vocational, would no longer count in performance tables.
The overall number of non-GCSE qualifications that could count for any pupil was capped at
two. This followed analysis in the Wolf report which critiqued the labour market value of
some vocational qualifications previously classified as equivalent to GCSE A*-C.81 The
qualifications that were de-recognized had been taken by many young people, but played a
particularly important role for young people on free school meals and with less good prior
attainment. A study by Burgess and Thomson in 2019 looked at the attainment of a group of
pupils with characteristics associated with high rates of entry in the de-recognised
qualifications. They found that overall, the percentage of this group achieving Level 2 fell
from 72 per cent to 61 per cent between 2013 and 2014, an eleven percentage point
reduction in one year. This group did not manage to catch up with the preceding cohort in
post-16 study: by age 18 only 75 per cent of them had achieved Level 2, compared with 82
per cent for the previous cohort. 82
Several other changes affected pupils’ Key Stage 4 options and experience.

• One major change was to bring all GCSEs back to a linear format and abandon
modular assessment. This change was announced in 2013 and took effect for those
sitting examinations in 2014. An independent evaluation found that modular or
linear exams did not appear to favour male or female students, or affect high- and
low-income students differently. However, the majority of teachers who expressed
an opinion thought the linear structure was more stressful, both in terms of the

79
Ben Bryant, Natalie Parish, and Jodie Reed, Research into How Local Authorities Are Ensuring Sufficient
Places and Supporting Vulnerable Children (DfE, 2022).
80
National Audit Office, Establishing Free Schools (NAO, 2013); Bobbie Mills, Emily Hunt, and Jon Andrews,
Free Schools in England: 2019 Report (Education Policy Institute, 2019).
81
Alison Wolf, Review of Vocational Education: The Wolf Report (DfE, 2011).
82
Simon Burgess and Dave Thomson, ‘The Impact of the Wolf Reforms on Education Outcomes for Lower-
Attaining Pupils’, British Educational Research Journal, 45.3 (2019), 592–621.

43
 weight of final exams, and losing the advantage that being able to track progress
reduced anxiety over results.83

• Second, there were significant changes to the content of GCSEs, ‘to make them more
challenging so pupils are better prepared for further academic or vocational study,
or for work’. Some changes were introduced from summer 2012 onwards, and
significant content changes were made in English language, English literature and
mathematics for courses taught from September 2015, and for a large range of
additional subjects including sciences and languages from September 2016. Content
changes included using longer texts in English literature, requiring better reading
skills in English, making several subjects more mathematically challenging, and
increasing the focus on the UK in history and geography.84 85

• Third, the introduction of ‘Progress 8’ as the key measure for performance league
tables strongly tilted subject choices towards a defined list of academic subjects.
Progress 8 is constructed on the basis of performance across eight qualifications,
including Maths and English (both double-weighted), three further qualifications
from the English Baccalaureate list (sciences, languages, history, geography) and
three further qualifications that can be GCSEs or non-GCSE qualifications approved
by the DfE. DfE research found that many schools said Progress 8 had led them to
narrow the curriculum.86

Post-16 education and policy on NEETs

The cohorts who experienced the new regime of GCSEs were also among the first to be
affected by the raising of the participation age to 17 in 2013, and 18 in 2015. This was
intended to raise skill levels and further reduce the number of young people who were
NEET. The policy was underpinned by duties on local authorities to assist young people to
participate, to secure sufficient suitable provision, and to collect information so that 16- and
17-year-olds who were not participating, or who were NEET, could be identified and
supported to re-engage.87

83
Jo-Anne Baird and others, Examination Reform: Impact of Linear and Modular Examinations at GCSE (Ofqual
and Oxford University Centre for Educational Assessment, 2019).
84
https://www.gov.uk/government/publications/2010-to-2015-government-policy-school-and-college-
qualifications-and-curriculum/2010-to-2015-government-policy-school-and-college-qualifications-and-
curriculum
85
Lupton and Thomson.
86
Cooper Gibson Research, Understanding Schools’ Responses to the Progress 8 Accountability Measure (DfE,
2017).
87
Department for Education, Participation of Young People in Education, Employment or Training: Statutory
Guidance for Local Authorities, 2016
<https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/561546
/Participation-of-young-people-in-education-employment-or-training.pdf>.

44
By the time of implementation, however, the new participation age coincided with multiple
other developments that reduced the resourcing for post-16 support.88

• The Connexions model of personal support and careers guidance came to an end.
The funding Local Authorities had received to fund Connexions support for 16-18s
was reduced, and merged into an overall early intervention pot which was then
further cut.89

• Responsibility for careers education and guidance was placed upon individual
schools, but with no additional resources. In 2013, a year after implementation,
Ofsted found that only a fifth of secondary schools were effective in careers
education and guidance, and that ‘vocational training and apprenticeships were
rarely promoted effectively, especially in schools with sixth forms’.90 The quality of
provision has been repeatedly criticised since, but despite changes, when a DfE
survey in 2019 asked secondary school pupils what careers activities they had
participated in at school in the past 12 months, 30 per cent (and 39 per cent of
children on free school meals) said they had participated in none. 91 92

• The Education Maintenance Allowance in England was abolished for new applicants
from 2011 and replaced with a much smaller bursary scheme, which was
discretionary and to be run by local authorities. Evaluation of the switch from
Education Maintenance Allowance to a discretionary bursary scheme estimated that
it had led to a 1.6 percentage point fall in full time participation for those year 12
students who would otherwise have been eligible for a full EMA, and 1.4 percentage
point reduction in participation or the larger group of pupils who would have been
eligible for any level of EMA support. The effects were estimated to be larger in year
13.93 Another study found that the new bursary scheme was not well known
amongst year 11 students, and that payments were not always timely, with the
result that participation was affected and some students experienced hardship.94

• In terms of study options post-16, funding for educational provision for 16- to 18-
year-olds fell significantly after 2010. Between 2010–11 and 2018–19, real spending
per student fell by around 12 per cent in further education and sixth form colleges,
and by 23 per cent in real terms for school sixth forms.95

88
Sue Maguire, ‘Will Raising the Participation Age in England Solve the NEET Problem?’, Research in Post-
Compulsory Education, 18.1–2 (2013), 61–76.
89
Tristram Hooley and A.G. Watts, Careers Work with Young People: Collapse or Transition? (University of
Derby International centre for Guidance Studies, 2011).
90
Ofsted, Going in the Right Direction? Careers Guidance (Ofsted, 2013).
91
House of Commons Business Innovation and Skills and Education Committees (Sub-Committee on Skills and
the Economy), Careers Education, Information, Advice and Guidance (HoC, 2016).
92
Department for Education, Omnibus Survey of Pupils and Their Parents or Carers: Wave 6 (DfE, 2019).
93
Jack Britton and Lorraine Dearden, The 16 to 19 Bursary Fund : Impact Evaluation (DfE, 2015).
94
Cheryl Lloyd and others, The 16 to 19 Bursary Fund : Year 3 Process Evaluation (DfE, 2015); Britton and
Dearden.
95
Britton, Farquharson, and Sibieta.

45
 • Apprenticeships for 16- to 18-year-olds were initially high on the Coalition agenda.
But annual apprenticeship starts for under-19s never rose above the level of
2010/11 (132,000) and then fell to 98,000 between 2015/16 and 2018/19.96 This
decline disproportionately affected the most disadvantaged and was driven by
several policy factors including perverse incentives in the design of the
apprenticeship levy.97
Special educational needs
The policy and funding framework for special educational needs changed significantly under
the Coalition and Conservative administrations.
A first change to note is the impact of changes to school performance measures. The move
away from ‘contextual value added’ league tables after 2010 meant, among other things,
that allowance was no longer made in league tables for the impact on a school if it had a
large number of pupils with special educational needs. This has been seen as penalising
schools that are more inclusive, as well as contributing to incentives to exclude or off-roll
pupils.98
A second important change, announced in 2012 and implemented in 2013/14, introduced a
new rule that in future, schools would have to find the first £6,000 of high needs support for
each pupil with special educational needs.99 This replaced multiple different arrangements
that had been in place previously. This meant that schools now faced large and lasting
financial consequences of having a pupil with special needs on their roll, for which they
might or might not have sufficient budget. The incentive effects of this in relation to
inclusion have been widely noted: the National Audit Office, for example, commented that it
‘risk[ed]incentivising mainstream schools to be less inclusive’.100 This issue is discussed again
later in this report, in Chapter 4, in connection with exclusion and off-rolling.
These changes were followed by the implementation in 2014 of major changes to the SEN
system, intended to improve it and extend support up to the age of 25 if needed. The
package of changes also combined ‘School Action’, and ‘School Action Plus’ into the new
category to be known as ‘SEN support’ which would be tailored locally, in consultation with
parents, to produce a local offer. The reform also broadened the content of SEN Statements

96
Niamh Foley, Apprenticeship Statistics Briefing Paper (House of Commons Library, 2021).
97
For a fuller discussion see: Neil Amin-Smith, Cribb Jonathan, and Luke Sibieta, ‘Reforms to Apprenticeship
Funding in England’, in IFS Green Budget 2017, ed. by Carl Emmerson, Paul Johnson, and Helen Miller (IFS,
2017); Ruth Lupton, Stephanie Thomson, and others, Moving on from Initial GCSE ‘Failure’: Post-16 Transitions
for ‘Lower Attainers’ (Nuffield Foundation, 2021); Jenna Julius, Henry Faulkner-Ellis, and Sharon O’Donnell,
Putting Apprenticeships to Work for Young People (NFER, 2021).
98
House of Commons Education Committee, Forgotten Children: Alternative Provision and the Scandal of Ever
Increasing Exclusions (HoC, 2018); Steve Preston, ‘Progress 8 Fails Pupils with Special Needs’, Schools Week,
November 2019.
99
Department for Education, School Funding Reform : Next Steps towards a Fairer System (DfE, 2012);
Department for Education, School Funding Reform : Arrangements for 2013/14 (DfE, 2013).
100
National Audit Office, Support for Pupils with Special Educational Needs and Disabilities in England (NAO,
2019).

46
to become ‘Education and Health Care Plans’ (EHCPs) and required Statements to be
transferred to EHCPs in the new form.101
As all these different changes worked their way through, against the backcloth of very tight
funding, significant problems started to become evident. The House of Commons Education
Committee reported in 2019 that the 2014 reforms had resulted in ‘confusion and at times
unlawful practice, bureaucratic nightmares, buck-passing and a lack of accountability,
strained resources and adversarial experiences’.102 The National Audit Office found that
many pupils were not being supported effectively, that pupils with special needs who did
not have EHCPs were particularly exposed, and that costs were rising because funding
pressures caused mainstream schools to pass pupils on to local authorities, who then used
independent provision as state special schools were full.103 Ofsted found that in January
2019, almost 3,500 children and young people with EHCPs (all ages) were waiting for the
provision due, and 2,700 were not in school or education at all.104
Figure ED5 illustrates the sharp reduction during the 2010s in the proportion of pupils in
mainstream secondary schools who were recorded with special educational needs that fell
short of a full statement. The size of this group shrank by 100,000 between 2010 and 2013,
and a further 150,000 between 2013 and 2015, and by 2019 reached a level much lower
than the level of 1997. The proportion of pupils on free school meals in state-funded
secondary schools receiving SEN support nearly halved in nine years, falling from 36 per
cent in 2010 to just 19 per cent in 2019. 105 Figure ED5 also shows the rise in appeals as the
decade went on. By 2018/19, 92 per cent of decisions were in favour of the appellant (at
least in some respects) compared with 69 per cent in 2011/12.
The government responded to the multiple concerns expressed in September 2019 by
announcing extra funding outside the normal spending reviews and promising to review the
whole special educational needs system.106 The review appeared, for consultation, in March
2022 and final proposals were published in March 2023. 107 The Children's Commissioner has
welcomed some elements of the plan but also expressed concern that it ‘does not go far
enough swiftly enough’, with planned implementation by the end of 2025 leaving ‘two more
years of children being fed into this cycle with the commensurate poor outcomes that has
necessitated this review in the first place’.108 109

101
Robert Long, Nerys Roberts, and Shadi Danechi, Special Educational Needs: Support in England (House of
Commons Library, 2020).
102
House of Commons Education Committee, Special Educational Needs and Disabilities (HoC, 2019).
103
National Audit Office, Support for Pupils with Special Educational Needs and Disabilities in England.
104
Her Majesty’s Chief Inspector of Education Children’s Services and Skills, Annual Report 2018/19 (Ofsted,
2019).
105
Not shown in chart. Figures from Department for Education, Special educational needs in England: January
2010 (Table 7) January 2019 (Table 5)
106
https://www.gov.uk/government/news/major-review-into-support-for-children-with-special-educational-
needs
107
https://www.gov.uk/government/consultations/send-review-right-support-right-place-right-time.
108
Children’s Commissioner for England, Statement in Response to the SEND Improvement Plan (OCC, 2023).
109
For a summary of wider reactions see: Robert Long and others, The Special Educational Needs and
Disabilities and Alternative Provision Improvement Plan (House of Commons Library, 2023).

47
Looked at over the longer term, it is clear that special educational needs policy went in
radically different directions in the pre-2010 and post-2010 period. Pre-2010, several
features of policy incentivised schools to identify pupils with special needs and to support
them to remain in mainstream schooling. Post-2010, several features of policy had the
opposite effect. The impact of these policies on school inclusion will be discussed again in
chapter 4 (Permanent Exclusion) and Chapter 5 (School Absence).

48
Figure ED5: Trends in Special Educational Needs, England

Number of pupils in state-funded Proportion of pupils in state-funded Special educational needs: appeals
secondary schools with special secondary schools with special educational outcomes decided 2011/12 to 2018/19
educational needs needs
25 3,000
800,000

700,000 19.7 2,500

20

600,000
15.9
2,000
12.4
500,000 15

Per cent
1,500
400,000

10
300,000 10.8
1,000

200,000
5
500
2.4
100,000 1.7

0 0 0

Pupils with SEN without statements or EHC Percentage with statements/EHCPs

plans Outcomes decided
Pupils with statements or EHC plans Percentage with SEN but no statement/EHCP Decision in favour of appellant

Source: Pupil numbers from DfE, Special Educational needs in England 2019. Appeals data from Ministry of Justice, Tribunal Statistics, 2021

49
Outcomes: attainment at 16
We now turn to the data about trends in educational outcomes at 16, covering data up to
2019.
Level 2 achievement
Overall
The most prominent target Labour set itself for secondary education in its first round of
PSAs was to increase the proportion of pupils obtaining Level 2 (5 A*-C grades at GCSE or
equivalent). Labour pledged to raise this proportion to 50 per cent by 2002 and achieved
this a year early. 110
This was the start of a long trend of improvement Figure ED6 shows that comparing the
cohort of young people turning 16 in 2011/12 with those of the same age a decade earlier,
an extra 20 percent of the school population obtained Level 2 at age 16.
But from 2012/13 the trend altered. Cohorts of pupils who turned 16 in each of the next
three years achieved slightly lower Level 2 performance than the cohort before them – a 6.6
percentage point fall in total. After 2014/15, this fall stopped and overall, between 2014/15
and 2018/19 performance rose slightly, by 1.3 percentage points.
Free schools’ meals, special educational needs, deprivation
This pattern of rise then fall is more pronounced for disadvantaged groups. Figure ED7
shows that for pupils receiving free school meals, the decade to 2011/12 was a period of
catching up with other pupils, and the three years that followed re-entrenched a wider gap.
It also shows the position for pupils assessed with different levels of special needs. For
pupils with special educational needs but no statement/education and health care plan,
significant improvements were seen until 2011/12, again followed by a sharp decline. (But
note that the SEN trends are complicated by the significant changes – first up, then down -
in the proportion of pupils with SEN support over the period, as described above.)

110
Department for Education and Skills, Departmental Report 2003 (DfES, 2003).

50
Figure ED6: Trends in Level 2 attainment by year turned 16, England

80
69.2
70 63.9
62.6

60
49.0
Per cent of cohort

50

40

30

20

10

Source: From 2001/02 figures are from DfE, Level 2 attainment in England at 16 and exclude independent schools.
Earlier figures are from National Performance Tables.

Figure ED7: Trends in proportion of cohort attaining Level 2 by age 16, by free school meal
eligibility and SEN status

90 90

78.0
80 80
72.5 70.1 70.2
66.6 67.9
70 70

58.5
60 60
53.4
Per cent of cohort

49.4
50 50
43.1
39.8
37.7
40 40 28.5
32.0
30 30
22.7
20 20
16.6 10.3
13.0
10.7
10 10 5.9

0 0
2017/18
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17

2018/19
2001
1998
1999
2000

2002
2010/11
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10

2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
2017/18
2018/19
2000
1998
1999

2001

No identified SEN
FSM not eligible
SEN support
FSM eligible
Education Health and Care plan

Source: From 2001/02 figures are from DfE, Level 2 attainment in England at 16 and exclude independent schools. Earlier
figures are from National Performance Tables.

51
Area deprivation
Turning to area deprivation, Figure ED8 shows that the period 2001/02 to 2011/12 saw
significant levelling up of Level 2 attainment. Performance improved overall and deprived
areas caught up. The gap between richest and poorest quartiles narrowed from over 39
percentage points in 2001/02 to under 22 percentage points a decade later. This was then
followed by levelling down – as national performance plateaued, poor areas fell backwards
and the area gap widened again, to 28.7 percentage points in 2015/16.

Figure ED 8: Trends in Level 2 attainment at 16, comparing least and most deprived areas
Area deprivation
90 80.5 77.8
80
68.4
70
58.9
60 50.5
50
40 29.1
30
20
10
0

25% least deprived 25% most deprived

Source: DfE, Level 2 attainment in England at 16. Excludes independent school pupils

Ethnicity
Inequalities by ethnicity also show some cases of narrowing then widening gaps. The
full picture of Level 2 achievement by ethnic group is set out at Figure ED9 below. It
shows that the largest gains made between 2003/04 and 2011/12 were for the
following ethnic groups:
• Any other Black background (+27.3ppt)
• Black Caribbean (+26.5 ppt)
• White and Black African (+24.3 ppt)
• Black African (+24.2 ppt)
• White and Black Caribbean (+23.6ppt)
Three of these groups then featured in the list of ethnicities losing most between 2011/12
and 2018/19, joined by White British pupils and Gypsy Roma pupils. All these groups are
picked out on the chart below.

• Black Caribbean (-12.9 ppt)

• Gypsy Roma (-10.6 ppt)
• White and Black Caribbean (-9.3 ppt)
• Any other Black background (-6.2 ppt)
• White British (-6.2 ppt)

52
Figure ED9: Trends in Level 2 achievement at age 16 by ethnicity, 2004 to 2019

(Ethnicities highlighted were the groups which saw largest performance declines between 2011/12 and 2018/19)

100

90

80 Any other
White
British White & Black Black
Black background Caribbean
70 Caribbean

60
Per cent of cohort

50

40

30 Gypsy
Roma

20

10

0
2003/04
2006/07
2009/10
2012/13
2015/16
2018/19
2004/05
2007/08
2010/11
2013/14
2016/17

2005/06
2008/09
2011/12
2014/15
2017/18
2003/04
2006/07
2009/10
2012/13
2015/16
2018/19
2004/05
2007/08
2010/11
2013/14
2016/17

2005/06
2008/09
2011/12
2014/15
2017/18
2003/04
2006/07
2009/10
2012/13
2015/16
2018/19
2004/05
2007/08
2010/11
2013/14
2016/17

2005/06
2008/09
2011/12
2014/15
2017/18
2003/04
2006/07
2009/10
2012/13
2015/16
2018/19
2004/05
2007/08
2010/11
2013/14
2016/17

2005/06
2008/09
2011/12
2014/15
2017/18
2003/04
2006/07
2009/10
2012/13
2015/16
2018/19
2004/05
2007/08
2010/11
2013/14
2016/17

2005/06
2008/09
2011/12
2014/15
2017/18
2003/04
2006/07
2009/10
2012/13
2015/16
2018/19
2004/05
2007/08
2010/11
2013/14
2016/17
Chinese Indian White and Asian Any other Asian White Irish Any other White Any other mixed White British Bangladeshi White and Black Pakistani Black African White and Black Any other Black Black Caribbean Traveller of Irish Gypsy Roma
background background African Caribbean background Heritage

Source: DfE, Level 2 attainment in England at 16 in England. Excludes independent school pupils

53
Intersection of gender, ethnicity and poverty
The patterns are similar when the intersection of different characteristics is considered.
Figure ED10 shows that white pupils eligible for free school meals saw large gains in
attainment between 2001/02 and 2011/12 GCSE rounds – an increase of more than 24
percentage points for girls and over 21 percentage points for boys (more than doubling
boys’ performance). But both groups then saw a sharp drop in performance after 2012. By
2018/19 level 2 attainment was 11 percentage points lower than in 2011/12 for white girls
eligible for free school meals, and 11.4 percentage points lower for boys.
Figure ED10: Trends in Level 2 attainment at age 16, white pupils in mainstream schools,
by gender and free school meal status

90

76.9
80
72.6

70
61.4
68.2
60
per cent of cohort

50.0 62.1
50
White female 51.0
39.0
40 non-FSM
40.1
30 White male 25.6
non-FSM
28.7
20
White female
FSM eligible 18.7
10
White male
0 FSM eligible
2005/06
2003/04

2004/05

2006/07

2007/08

2008/09

2009/10

2010/11

2011/12

2012/13

2013/14

2014/15

2015/16

2016/17

2017/18

2018/19

Source: : DfE, Level 2 attainment in England at 16 in England. Excludes independent school pupils

Regions
The regional pattern of change was also striking. As Figure ED11 shows, the upswing in
performance between 2001/02 and 2011/12 was greatest in regions which had had low
performance at the beginning of the period – London, Yorkshire and the Humber, and the
North East. London’s attainment improved sharply even before 2002.111
After 2011/12, some of the regions that had gained most saw big falls again. The North East
region moved from biggest gainer between 2001/02 and 2011/12 to losing the most
between 2011/12 and 2017/18 with an 11.8 percentage point drop in the proportion
achieving level 2. But London saw relatively small declines in attainment.

For pre-2002 data see Jo Blanden and others, Understanding the Improved Performance of Disadvantaged
111

Pupils in London (Centre for Analysis of Social Exclusion (CASE), 2015).

54
Figure ED11: Trends in Level 2 attainment by region, mainstream state-funded schools,
selected years

80

70

60

50
Per cent

40

30

20

10

0
South East South East of East West North London Yorkshire North East
West England Midlands Midlands West and the
Humber
2001/02 2011/12 2017/18

Source: DfE. Level 2 and 3 attainment age 16 to 25. Latest published figures for regional split.

Other ways to measure performance at age 16

As noted above, there are many other ways to measure performance at secondary level. To
cover some of the possibilities briefly:

• Labour’s 1998 target to reduce below 5 per cent the proportion of the cohort not
achieving any pass at GCSE was met, one year late, in 2003/04. It continued to fall
further and the proportion of pupils on free school meals achieving no passes fell
from the shocking 2002 level (12.8 per cent) to 2.3 per cent by 2010.112

• Labour’s ‘floor targets’ for schools, referred to earlier in this chapter, were designed
to ensure the poorest performing schools raised their performance, rather than
being masked by national averages. The target to reduce to zero the number of
schools where fewer than 30 per cent of pupils achieved Level 2 was not quite met,
but the number of schools missing the benchmark fell sharply, from 2003/04 to
2007/08, from more than 340 to just 26.113

• From 2005/06 onwards, the government began to publish attainment on the

measure of Level 2 including English and Maths. The percentage of pupils attaining

112
Lupton and Obolenskaya, Labour’s Record on Education: Policy, Spending and Outcomes 1997-2010.
113
Department for Children Schools and Families, Departmental Report 2009.

55
 this benchmark by age 16 rose every year until 2013, then fell in 2014 on both old
and new qualification rules.114

• Level 2 attainment by age 19 (i.e. including also the achievement of pupils who
attain Level 2 between 16 and 19) peaked at 87.5 per cent in 2014/15. This was same
cohort as the peak for achievement at 16. Performance then fell four years in
succession to 83.4 per cent for pupils aged 19 in 2018/19. 115

• In the post-2015 period, changes to curriculum, grading systems and what

information is collected data mean there are only a few years of comparable data on
the measures the Conservatives have prioritised. These show that the percentage of
pupils in state funded schools entering for the EBacc rose by 1.3 percentage points
between 2014/15 and 2018/19, to 40 per cent. The proportion of pupils achieving
both English and Maths at grade 5 or above increased by 0.6 percentage points
between 2016/17 and 2018/19.116

Other measures of attainment gaps

There are several other ways to measure attainment gaps in secondary schools, comparing
overall points scores or the attainment of qualifications in specific subjects. All of these
analyses show similar patterns of narrowing then widening gaps:
• The free school meals gap in terms of pupils attaining Level 2 including English and
Maths fell from 2006 until 2012, albeit modestly, from 28.1 points to 26.3 points. It
then rose in 2013 and 2014 on both old and new counting rules.117

• The Department for Education publishes a Key Stage 4 (GCSE) disadvantaged

attainment gap index which goes back as far as 2010. This shows an initially
improving trend until 2013/14, but since then it has fluctuated within a small margin
before rising again in the pandemic. 118

• The Education Policy Institute publishes regular reports on disadvantage gaps. On

their measure, progress in narrowing the ‘headline’ disadvantage gap at Key Stage 4
has stalled since 2017, and for persistently disadvantaged pupils there has been no
progress in closing the Key Stage 4 attainment gap since 2011. 119

114
Lupton and Thomson. Table 8.
115
Department for Education, Level 2 and 3 Attainment in England: Attainment by Age 19 in 2019 Statistical
Commentary (DfE, 2020).
116
Lupton and Obolenskaya, The Conservatives’ Record on Compulsory Education : Spending , Policies and
Outcomes in England , May 2015 to Pre-COVID 2020.
117
Lupton and Obolenskaya, Labour’s Record on Education: Policy, Spending and Outcomes 1997-2010; Lupton
and Thomson.
118
https://explore-education-statistics.service.gov.uk/find-statistics/key-stage-4-performance-revised
119
Emily Hunt and others, Covid-19 and Disadvantage Gaps in England 2020 (EPI, 2022).

56
Were the improvements real?
The striking improvement shown in the official data under Labour and in the early years of
the Coalition is not taken at face value by all commentators. Three criticisms ought to be
noted and addressed here.
• Borderline effects: one suspicion expressed is that the improvement in attainment
may have been concentrated amongst pupils whose attainment placed them at the
C/D borderline.120 It is probably fair to say that schools made great efforts to get
pupils a C grade wherever possible, given its significance for both pupil and school
and there is some evidence of this.121 But overall, the data shows that improvements
in attainment in Labour’s period up to 2013 were not simply a product of students
being pushed over particular examination thresholds, but that improvement was
seen across the spectrum of attainment, especially towards the bottom. 122

• Vocational qualifications: another challenge to the published data focuses on the

difference in the improvement trajectory between the broad Level 2 measure and
other measures which only count level 2 if it includes a C in English and Maths, or
measures that also exclude all vocational equivalents.123 It is a mathematical fact that
the trajectories for each of these measures are different, but that is no basis to
dismiss entirely the gains for those pupils who took up vocational pathways. 124
Moreover, as we shall see, evidence elsewhere in this report will underline the value
of curriculum flexibility at Key Stage 4 in combating disaffection, reducing absence
and supporting post-16 participation.

• International comparisons: A final challenge is whether apparent improvements in

domestic results are also apparent in international tests. This is a complicated area,
examined in some detail in a paper by John Jerrim, following claims from
Conservative Education Secretary Michael Gove that England's rankings in PISA
(Programme for International Student Assessment) tests had fallen during Labour’s
time in office. Jerrim found that, although England’s PISA scores appeared to have
fallen between 2000 and 2009, the surveys in different years were not comparable
for various reasons (for example, the 2006 and 2009 tests were taken five months
earlier in the school year). In the other major international study, TIMSS (Trends in
International Mathematics and Science Study) England’s maths performance
improved between 2003 and 2007: however, Jerrim points out that this series too
has some methodological problems. Jerrim’s conclusion is that ‘the coalition

120
Anastasia de Waal, School Improvement – or the ‘ Equivalent ’ (Civitas, 2009).
121
Simon Burgess and Dave Thomson, School Accountability and Fairness: Does ‘Progress 8’ Encourage Schools
to Work More Equitably? (Nuffield Foundation, 2020).
122
Lupton and Thomson; Rikki Dean and Moira Wallace, ‘New Labour and Adolescent Social Exclusion: A
Retrospective’, in Social Policy Review 30, ed. by C Needham, E Heins, and J Rees (Policy Press, 2018), pp. 269–
90.
123
Anthony Heath and others, ‘Education under New Labour, 1997-2010’, Oxford Review of Economic Policy,
29.1 (2013), 227–47.
124
Geoff Whitty and Jake Anders, ‘(How) Did New Labour Narrow the Achievement and Participation Gap ?’
(Centre for Learning and Life Chances in Knowledge Economies and Societies, 2014).

57
 government should not base educational policies on the assumption that the
performance of England’s secondary school pupils has declined (relative to
that of its international competitors) over the past decade’.125 126

What did pupils think?

In all the argument about outcomes, the question of what pupils think about their school
experiences is all too often overlooked. Some of the indicators in the Appendix to this report
cast light on pupils’ changing attitudes to school over this period. Figure OY6 shows a rise in
10-to-15-year olds’ happiness with school between 2004 and 2011/12, followed by a
significant reduction over the period to 2018/19. Figure OY7, from a four-yearly survey,
shows an overall picture of improvement in positive perceptions up to 2014 (in some cases
measured back to 2002, in some cases only measured back to 2010), but a decline in all the
positive perception indicators between the 2014 and 2018 surveys.
A more recent study explores data from a nationally representative survey completed by
over 10,000 15- and 16-year-olds in the summer term of 2021. This is a cohort who had
spent their entire secondary school career in schools whose curriculum and approach reflect
the new GCSEs and greater emphasis on traditional academic subjects. The cohort was also
significantly affected by the pandemic. Only 55 per cent of survey respondents agreed with
the statement that on the whole they liked being at school. (When the same question was
asked in the LSYPE survey in 2006, the comparable proportion was 84 per cent.) The
qualitative strand of the research identified two broad forms of alienation - curriculum
alienation, and identity-based alienation - potentially underpinning the low level of
happiness with school.127

Outcomes: post-16
Participation in education post 16
As noted earlier, many of the changes in secondary education were designed not just to
improve attainment at 16 but also to support and enable young people to carry on in
education after 16 (the then minimum school leaving age).
As Figure ED12 illustrates, there has been a significant increase in the proportion of the
cohort continuing in full-time education over the last two decades. Between 2001 and 2019,
the proportion of 16-year-olds remaining in full-time education increased by just over 17
percentage points. For 17-year-olds the increase was over 20 percentage points. Most of
this increase occurred by 2010. In the case of 17-year-olds, participation rose by 15
percentage points between 2001 and 2010, then just over five points between 2010 and
2019.

125
John Jerrim, ‘England’s “Plummeting” PISA Test Scores between 2000 and 2009 : Is the Performance of Our
Secondary School Pupils Really in Relative Decline?’, Institute of Education DoQSS Working Papers, 11.09
(2011).
126
For a discussion of results of the PISA tests in 2018 see Lupton and Obolenskaya, The Conservatives’ Record
on Compulsory Education : Spending , Policies and Outcomes in England , May 2015 to Pre-COVID 2020.
127
Charlotte McPherson and others, Schools for All? Young People’s Experiences of Alienation in the English
Secondary School System (Edge Foundation, 2023).

58
Figure ED12: Trends in participation in full time education by 16 and 17 year olds, England

16 year olds 17 year olds

100%
100%
88.4% 90%
90%
84.4% 79.0%
80% 73.8%
80% 71.0%
70%
70%
58.7%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%

0% 0%

end 2003

end 2005
end 1997
end 1998
end 1999
end 2000
end 2001
end 2002

end 2004

end 2006
end 2007
end 2008
end 2009
end 2010
end 2011
end 2012
end 2013
end 2014
end 2015
end 2016
end 2017
end 2018
end 2019
end 2014
end 2015
end 1997
end 1998
end 1999
end 2000
end 2001
end 2002
end 2003
end 2004
end 2005
end 2006
end 2007
end 2008
end 2009
end 2010
end 2011
end 2012
end 2013

end 2016
end 2017
end 2018
end 2019

Source: DfE, Participation in education, training and employment: 2021

The trends for 16- and 17-year-olds are shown by gender in Figure ED13 gap between boys
and girls narrowed somewhat after 2007, but widened for 17-year-olds after 2012.
Figure ED13: Trends in participation in full time education, by gender

16 year olds, 17 year olds,

by gender by gender
100% 100%
91.3%
86.4%
90% 90% 82.7%
76.3%
80%
73.3% 80%

70% 82.5% 85.7% 70%

61.6%

60% 75.4%
67.2% 60% 71.4%

50% 50%
55.8%
40% 40%

30% 30%

20% 20%

10% 10%

0% 0%
end 1997
end 1998
end 1999
end 2000
end 2001
end 2002
end 2003
end 2004
end 2005
end 2006
end 2007
end 2008
end 2009
end 2010
end 2011
end 2012
end 2013
end 2014
end 2015
end 2016
end 2017
end 2018
end 2019
end 1997
end 1998
end 1999
end 2000
end 2001
end 2002
end 2003
end 2004
end 2005
end 2006
end 2007
end 2008
end 2009
end 2010
end 2011
end 2012
end 2013
end 2014
end 2015
end 2016
end 2017
end 2018
end 2019

16 year old boys 16 year old girls 17 year old boys 17 year old girls

Source: DfE, Participation in education, training and employment: 2021

59
International comparisons of participation
Historically, the UK had lagged behind OECD average levels of educational participation
by 15–19-year-olds. Between 2006 and 2014 this gap was eliminated as Figure ED14
shows.
Figure ED14: Trends in 15–19-year-olds in education; UK compared with OECD average
100
90
81

80
70
60 70
Per cent

50
40
30
20
10
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

UK OECD
Source: OECD Education at a Glance, 2012 and subsequent years. Consistent pre 2006 data not available

16- and 17-year-olds not in education, employment and training (NEET)

NEET rates for 16- and 17-year-olds are shown Figure ED15 below. This chart shows that:

• NEET rates peaked in 2002 for 16-year-olds and in 2005 for 17-year-olds

• these two groups saw generally declining rates from that point onwards, including in
the years surrounding the raising of the participation rate in 2013

• NEET rates for 18-year-olds rose substantially in 2009, during the recession, and their
fall came later

• since the late 2000s, there has been a much wider gap between NEET rates for 18-
year-olds and the younger age groups.
The fall in the NEET rate for 16- and 17-year-olds was mainly the result of the growing
participation in full time education discussed above, but this was partly offset by falls in the
proportion of young people in employment and work-based learning, including
apprenticeships. The PSA target of a two percentage point reduction between 2004 and
2010 was not met, even for 16- and 17-year-olds.

60
Figure ED15: Trends in NEET at age 16, 17 and 18, England
18% 16.5%

16%

14%
11.4%
12%
9.5%
10%

8%
4.8%
6% 7.6%
4%

2% 3.8%

0%

16 year olds 17 year olds 18 year olds

Source: DfE, Participation in education, training and employment: 2021

The recent trend in NEET rates for 16- and 17-year-olds looks rather flat. But this conceals a
gender split. Figure ED16 shows that NEET rates worsened for boys after 2015, but
improved for girls.
Figure ED16: Trends in NEET rates of 16- and 17-year-olds by gender

16 year olds 17 year olds

12% 12%

10% 10%

8% 8%

6% 6%

4% 4%

2% 2%

0%
0%
2009
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008

2010
2011
2012
2013
2014
2015
2016
2017
2018
2019

2007
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006

2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019

16 female 16 male 17 female 17 male

Source: DfE, Participation in Education and Training of 16-18 year olds, 2021

61
NEET status and the participation age

How does it come about that NEET numbers remain at this level in the wake of the Raising
Participation Age legislation? It turns out that there are many ways for young people to find
themselves outside education. The reasons could include some or all of the following:

• one course or placement has ended and another one has not been found or has not
started. ‘Churning’ between different post-16 courses is a phenomenon widely found
amongst young people who are NEET 128
• a young person has health issues or a disability, or is a young parent. Government
guidance recognises that a small number of young people will need breaks in, or
periods out of, education or training129
• a young person is off the radar of local authorities and therefore not getting help to
find and take up options.
In relation to the last of these issues, aggregated statistics show that for 2.3 per cent of the
16- and 17-year-old cohort in England, their current activity was not known to the local
authority in 2019, with figures much higher in some local authorities.130

Discussion
The big picture on outcomes
Looking at educational outcomes over a period of two decades brings out very clearly trends
and turning points in secondary school attainment. Across a range of secondary school
attainment indicators, the decade up to 2012 saw improvement and narrowing
disadvantage gaps. Over the same period, secondary school absence levels fell by 40 per
cent, permanent exclusion rates halved, and an extra fifth of the teenage population
remained in full-time education after 16. After 2012 Level 2 attainment fell from its peak
with the largest losses occurring for poorer children, those with special educational needs
but no statement/education and health care plan, some ethnic groups, and some regions.
NEET rates for 16- and 17-year olds have fallen since their peak in the early 2000s, but the
NEET rate for 18-year-olds remains close to where it was in 1997, and for 16- and 17-year-
old boys, NEET rates edged up again after 2015.
Explaining the trends
These outcomes trends do not map neatly onto the lifespan of political administrations, with
some striking turning points occurring two or three years into the Coalition government. We
will follow the turning points to look at explanations of these trends in two phases,
pre-2012, and the post-2012 period.

128
Sue Maguire, ‘Who Cares? Exploring Economic Inactivity among Young Women in the NEET Group across
England’, Journal of Education and Work, 31.7–8 (2018), 660–75; Lupton, Thomson, and others; Lisa Russell,
‘The Realities of Being Young, Unemployed and Poor in Post-Industrial Britain’, Power and Education, 8.2
(2016), 160–75.
129
Department for Education, Participation of Young People in Education, Employment or Training: Statutory
Guidance for Local Authorities.
130
Department for Education, 16-17 Year Olds Recorded in Education and Training and NEET by Local
Authority, 2019 (DfE, 2020).

62
Pre-2012
Several overview studies have synthesised the evidence on educational outcomes over this
the period and there is a large evidence base covering education initiatives under the
Labour government, comprising both research on individual policies and overview studies.
The main themes are set out below. In a pattern that will become familiar, many of them
potentially reinforce each other.

• Innovation: the chapter has already referenced evaluation evidence suggesting

positive impacts on attainment and participation from policies including Excellence in
Cities, Extended Schools, the London Challenge, the early Academies programme, the
Education Maintenance Allowance and Connexions. In a broad study of evidence on
the attainment gap, Whitty and Anders pick out several key initiatives as likely
contributors to narrowing inequalities, notably the literacy and numeracy strategies
and reading recovery programme, London Challenge, and the extended schools
initiative.131 Anthony Heath and colleagues, who believe that participation improved
under Labour but that Level 2 improvements were overclaimed, nonetheless agree
that there is strong evidence on the impact of the literacy hour, and the Education
Maintenance Allowance.132
• Resourcing and capacity: This chapter has already noted the IFS studies on the
increase in secondary school spending and its more pro-poor distribution between
1997 and 2010. A 2013 paper by Lupton and Obolenskaya lists multiple Labour policy
initiatives which brought about an overall increase in capacity in the system between
1997 and 2010, such as improved pupil-teacher ratios, greater access to other
support in and out of school, better paid staff with more access to professional
development, more subject choice, and extended schools. Taken together these
meant that ‘the experience of schooling for children in 2010 would have been
substantially different from that of their counterparts in 1997, especially if they lived
in disadvantaged areas’.133

• Changes in the 14-19 phase: a study of 14-19s by Rogers and Spours characterises
the period 2004-2012 as a phase of system growth where improvements in
achievement and participation at 16 and up to 18, especially for ‘middle attainers’,
arise from growing recognition and take-up of vocational qualifications at GCSE;
modularisation of GCSEs, as well as increased funding, and the introduction of
Education Maintenance Allowances. 134 Other studies of the changes in post-16
participation and the fall in NEET rates also focus on policy drivers: in addition to the

131
Whitty and Anders.
132
Heath and others.
133
Lupton and Obolenskaya, Labour’s Record on Education: Policy, Spending and Outcomes 1997-2010.
134
Lynne Rogers and Ken Spours, ‘The Great Stagnation of Upper Secondary Education in England: A Historical
and System Perspective’, British Educational Research Journal, 46.6 (2020), 1232–55.

63
 Education Maintenance Allowance, credit is given to the impact of the Connexions
service, and the ‘September Guarantee’ of a college place for school leavers.135

• London: A specific focus for research has been on why the relative performance of
London pupils at secondary school improved earlier than other regions (and
outstripped them). Possible explanations have included the impact of improvements
in London’s primary schools between 1999 and 2003, better secondary school
performance, improved funding, teacher recruitment and leadership, and the effects
of the London Challenge programme after 2003. To some extent these may reflect
London’s high level of exposure to early pilots. London’s greater proportion of pupils
from high performing ethnic minorities has also been seen as important. 136

• The impact of wider factors: Finally, we should note the relevance of other forms of
disadvantage that often sit slightly outside mainstream educational research. As
other chapters of this report set out, the years of improving attainment coincide
with a period of considerable government focus on reducing child poverty, school
non-attendance, and adolescent drug and alcohol use. The declines in those
indicators, and the role of government policy in driving those declines, also need to
feature in a list of possible explanatory factors.
After 2012
As we explore what changed around or after 2012, the research literature again consists of
some studies of specific policies, and some broader overviews. Six main themes stand out.
Again, these overlap and potentially reinforce each other.

• Funding for disadvantaged schools: Research on the impact of the Pupil Premium
shows a mixed picture.137 As noted earlier, while the principle is widely supported,
there have been questions about whether the money is always being spent on the
pupils for whom it was intended. However, more fundamentally, its apparent goals
have been undercut by other policy decisions - the abolition of other grants at its
inception and by other funding cuts which were disproportionately targeted on
schools with the highest levels of deprivation. The IFS research noted above found
that the differential in spending per pupil between the most-deprived set of schools

135
Sue Middleton, Sue Maguire, and others, The Evaluation of Education Maintenance Allowance Pilots: Three
Years’ Evidence: A Quantitative Evaluation (DfES, 2003); H. Chowdry, L. Dearden, and C. Emmerson, Education
Maintenance Allowance Evaluation with Administrative Data: The Impact of the EMA Pilots on Participation
and Attainment in Post-Compulsory Education (IFS, 2008); Maguire, ‘Will Raising the Participation Age in
England Solve the NEET Problem?’
136
Simon Burgess, Understanding the Success of London’s Schools (Centre for Market and Public Organisation,
2014); Alex Macdougall and Ruth Lupton, The ‘London Effect’: Literature Review (University of Manchester /
Joseph Rowntree Foundation, 2018); Ellen Greaves, Lindsey Macmillan, and Luke Sibieta, Lessons from London
Schools for Attainment Gaps and Social Mobility (Social Mobility and Child Poverty Commission, 2014).
137
Jake Anders and Morag Henderson, ‘Socioeconomic Inequality and Student Outcomes in English Schools’, in
Socioeconomic Inequality and Student Outcomes: Cross-National Trends, Policies and Practices, ed. by Volante
et al. (Springer 2019). 64
 and the least-deprived had fallen from 34 per cent in 2010/11 to 23 per cent in
2019/20. 138
• Workforce reductions and difficulties meeting additional needs. Lupton and
colleagues analysed the record of both the Coalition and the Conservatives on
schools and compulsory education. The 2015 study of the Coalition covered 2014
outcome data and identified the beginning of the stalling in performance and
widening socio-economic gaps shown earlier in this chapter. The 2020 study of
compulsory education found a system ‘increasingly under strain, evidenced in
teacher shortages, rising pupil-teacher ratios and difficulties meeting additional
needs’ with ‘no real evidence that the efforts put into remodelling the system were
substantially improving it’.139

• Limited benefits from academisation: This chapter has already covered research into
the impact of academisation and MAT membership on school performance. This
shows that, overall academisation and MAT membership appear to have made no
systematic difference to attainment.

• Policies with negative effects on the attainment of disadvantaged groups: many

research studies listed earlier in the chapter, and in Chapters 4 and 5, outline clear
negative impacts on disadvantaged groups from specific policies under the Coalition
and Conservatives. This list includes: limits on the recognition of vocational
qualifications in performance measures; the incentives created by the funding rules
for special educational needs; the impact of performance measures that take no
account of special needs or disadvantage; curriculum narrowing; reductions in
external support available to schools; reduction in the availability of careers advice;
abolition of the EMA; and the reduction in 16- to 18-year-olds’ access to
apprenticeships.

• Changes in the 14-19 phase: looking at the 14 -19 phase as a whole, Rogers and
Spours characterise the post 2012 phase of policy as a period of stagnation and
falling attainment, caused by: reform of qualifications and assessment; changes to
accountability measures, inspection and funding; as well as schools responding to
national incentives with more selective and exclusionary institutional behaviours.140
In another assessment, Hodgson and Spours concluded that middle attainers in the
14 to 19 phase had been only ‘half-served’ by Labour, ‘because of its incomplete and
contradictory 14–19 reforms’ and had subsequently been ‘overlooked’ by Coalition
policy because of its emphasis on high attainers.141

138
Kate Ogden and others, Does Funding Follow Need? An Analysis of the Geographic Distribution of Public
Spending in England (IFS, 2022).
139
Lupton and Thomson; Lupton and Obolenskaya, The Conservatives’ Record on Compulsory Education :
Spending , Policies and Outcomes in England , May 2015 to Pre-COVID 2020.
140
Rogers and Spours.
141
Ann Hodgson and Ken Spours, ‘Middle Attainers and 14-19 Progression in England: Half-Served by New
Labour and Now Overlooked by the Coalition?’, British Educational Research Journal, 40.3 (2014), 467–82.

65
 • Wider factors: the above list again mainly focuses on topics such as institutions,
courses, and qualifications that sit very centrally in mainstream educational
research. Such an approach risks overlooking wider factors which may also have
influenced attainment and attainment gaps. Drawing on evidence from other
chapters of this report, we should also include in our analysis the impact on children
of family poverty (Chapter 3), increases in school exclusion and severe absence
(Chapters 4 and 5), evidence of increased use of drugs by adolescents (Chapter 8)
and the wider picture set out in Appendix 1, including declining happiness at school,
and increasing numbers of young people suffering anxious disorders.
Secondary education has been an area of intense policy activity for governments of all
complexions through the two decades discussed in this report. The changes in secondary
school attainment and post-16 participation are an important foundation for understanding
the wider picture of outcomes for young people. As we shall see in the chapters that follow,
many of these other outcomes show rapid improvement and narrowing inequalities until
around 2013, and stalling of progress and widening inequalities thereafter. The possible
interactions between these trends will be explored further in the chapters that follow.

66
Chapter 2 Youth Crime and Justice
Introduction

This chapter continues the survey of ‘cornerstone’ areas of social policy relevant to young
people by summarising policy on policing, crime, and youth justice from 1997 to 2019, and
discussing trends in youth offending. This is a policy area in which I was involved between
2002 and 2008, first as Chief Executive of the Office of Criminal Justice Reform142, then as
Home Office Director General of Crime and Policing.
The policy story, in a nutshell, is a tale of interventionist policies under Labour, with
increased funding for policing and prevention, followed by a more devolved approach and
reduced funding under the Coalition and Conservatives. The outcome trend is dominated by
a long decline in overall crime from its 1995 peak: the fall continued, with fluctuations,
during Labour’s period in office, and fell further in the second decade, but flattened out
between 2017 and mid-2019.
Many of the statistics in this area are about overall crime and offending, and do not (and
cannot) separate out young people specifically. So some of the analysis in this chapter deals
with the overall picture, before exploring what we know about young people and offending
specifically. The level of offending perpetrated by young people appears to have fallen
substantially during the fall in overall crime, but the recent trend is hard to establish, and
young people have been significantly involved in the rise in knife crime after 2014.
The policies described in this chapter - policing, crime and youth justice - are, of course, only
one piece of the jigsaw in terms of what influences youth offending. Many of the drivers of
youth offending, such as alcohol, drugs, school absence and school exclusion, are covered in
other chapters of this report: a full account of government policy on crime only really
emerges when all these factors are considered in the round.

Note on coverage

Geographical coverage: Most chapters of this report cover England only. However, the
Home Office and main criminal justice institutions cover both England and Wales, so this
chapter reflects that.

Children as victims: Note also that the chapter focuses on offending by young people and
does not cover the equally large subject of children and young people as victims of crime.
Some key statistics on trends in children and young people’s experience of crime as
victims are included in Appendix 1. Figure OY14

142
The Office for Criminal Justice Reform was a joint team set up by the Home Office, Department of
Constitutional Affairs and Crown Prosecution Service, to improve joint working between criminal justice
agencies. It was dissolved in 2010.

67
Labour policies 1997–2010: overview
Overall approach
In campaigning for office, the Labour Party set as the essence of its crime policy the notion
of ‘tough on crime and tough on the causes of crime’, listing social deprivation and drugs
amongst the prominent causes.143 Juvenile offending attracted particular focus and this was
reflected in the inclusion on the party’s five-point pledge card of a commitment to ‘fast-
track punishment for young offenders by halving the time from arrest to sentencing’.
Youth Offending Teams and the Youth Justice Board
Once in government, in November 1997, Home Secretary Jack Straw set out reforms to
improve the operation of the youth justice system, and to encourage more work on
prevention of youth offending. 144 The Crime and Disorder Act 1998 created YOTs (youth
offending teams) as local, multi-agency bodies , convened by local authorities but including
representatives of probation, social services, police, health, and education, under the
oversight of a new national Youth Justice Board for England and Wales which would monitor
local performance and oversee standards for secure accommodation.
Resources and police numbers
The focus on youth crime sat within a context of multiple other policies to tackle crime by
offenders of all ages. Increased resources for law and order were one part of this. Spending
on the police, criminal justice system and wider public order and safety rose by more than
50 per cent in real terms in Labour’s first decade.145 Amongst other things, this funded
16,000 additional police officers between 1998 and 2010, as well as the development of
police community support officers (PCSOs) who boosted the capacity for visible policing.
(See Figure YJC1 later in this chapter)
The strategy: crime reduction targets, performance pressure
National policy during this period was characterised by strong government pressure on local
agencies to intervene to reduce crime. The Crime and Disorder Act 1998 required local
authorities and the police, in partnership with other agencies, to set up Crime and Disorder
Reduction Partnerships. Performance was overseen by the Home Office, operating through
the Police Standards Unit, the Inspectorate of Constabulary, and Government Regional
Offices. Policing techniques developed considerably throughout this period, in the UK as in
the US, as police forces increasingly adopted new models of policing which evidence
suggested were more effective in reducing crime and responding to communities’ needs. 146
Many of these strategies were strongly data-led and targeted their efforts on ‘repeat crime
places, victims, suspects, situations, and other patterns’.147

143
Labour Party, New Labour: Because Britain Deserves Better (Labour Party, 1997).
144
https://publications.parliament.uk/pa/cm199798/cmhansrd/vo971127/debtext/71127-05.htm
145
HM Treasury, Meeting the Aspirations of the British People: 2007 Pre-Budget Report and Comprehensive
Spending Review (HMT, 2007).
146
Nick Tilley, ‘Modern Approaches to Policing: Community, Problem-Oriented and Intelligence-Led’, in
Handbook of Policing, ed. by Tim Newburn, 2nd edn (Cullompton: Willan Publishing, 2008), pp. 373–403.
147
Lawrence W Sherman, ‘The Rise of Evidence-Based Policing: Targeting, Testing, and Tracking’, Crime and
Justice, 42.1 (2013), 377–451.

68
PSA targets were set for the reduction of total crime and specific crime types: vehicle crime,
domestic burglary, and robbery were singled out for attention in early PSA rounds.148 Anti-
social behaviour was also the subject of considerable attention, and the government policies
on this divided opinion. For some, their approach was an overdue prioritisation of behaviour
that made victims’ lives a misery. Others objected to the fact that the ‘anti-social behaviour
order’ or ASBO, mingled a civil order (for example to cease harassment) with criminal
sanctions for breach, and that it drew more people into the criminal justice net. 149
Later in Labour's time in government, the focus of the PSA targets switched to total crime
(from 2004 to 2007) and in 2007 the emphasis was further shifted to ‘the most serious
violence’. 150 151 152 This was accompanied by an action plan for reducing violence, including
the use of guns, knives, and gang-related violence. Part of this plan was to collect better
data on knife offences, to complement data about ‘assaults by sharp objects’ which
hospitals had started to collect and collate.153
Tough on crime?
Improving the rigour of the criminal justice system was seen by the Labour government as
integral to its approach, and important for public confidence. The 1997 manifesto pledge to
halve the average 142 days from arrest to sentencing for persistent young offenders
reflected the frustration that the criminal justice system did not act quickly enough to
prevent criminal careers developing. The pledge was met by 2002 and broadly sustained
thereafter. 154 The Crime and Disorder Act 1998 also significantly changed the caution
system for young offenders, with a new framework which involved youth offending teams at
an early stage but also escalated virtually automatically, if there were further offences, from
reprimand to final warning, then formal charge on a third offence.155
In the 2002 spending review the government set a new joint PSA target for the criminal
justice system, to increase the number of offences brought to justice to 1.2 million. This was
intended to raise the profile of detection and investigation and improve joint working
between police, prosecutors and courts. But it proved to have a perverse side-effect: as all
crimes counted equally towards the target, it had the consequence of incentivising the
police to pursue arrests for easy-to-detect crimes and thus distorting priorities within

148
HM Treasury, Public Services for the Future : Modernisation, Reform, Accountability; HM Treasury, Prudent
for a Purpose : Building Opportunity and Security for All.
149
Sarah Hodgkinson and Nick Tilley, ‘Tackling Anti-Social Behaviour: Lessons from New Labour for the
Coalition Government’, Criminology and Criminal Justice, 11.4 (2011), 283–305; John Muncie, Youth and Crime
(Sage Publications, 2021).
150
HM Treasury, 2004 Spending Review Public Service Agreements 2005-2008; HM Treasury, Meeting the
Aspirations of the British People: 2007 Pre-Budget Report and Comprehensive Spending Review.
151
HM Government, Youth Crime Action Plan (HMG, 2008).
152
As the senior Home Office official on crime and policing from 2005 to 2008, I was involved in policy and
delivery for these targets.
153
Home Office, Saving Lives. Reducing Harm. Protecting the Public (HM Government, 2008).
154
Ministry of Justice, Average Time From Arrest To Sentence For Persistent Young Offenders: April 2008 (MoJ,
2008).
155
Home Office and Youth Justice Board, Final Warning Scheme: Guidance for the Police and Youth Offending
Teams (HO, 2002).

69
policing.156 It also caused the number of young people entering the criminal justice system
to rise, even though crime was falling.157 These perverse effects caused the target to be
dropped in 2007. The Youth Justice system instead acquired an objective to reduce the
number of 10–17-year-old first-time entrants to the Criminal Justice System. 158 159
Diversion
This reframing of targets in 2007 led to greater emphasis on informal responses to low-level
offending, with the government piloting a Youth Restorative Disposal (a process designed to
show victims the impact of their actions and allow victims greater involvement in the
process) and the Youth Justice Liaison and Diversion scheme (which involved identifying
arrestees who had mental health or developmental difficulties, and – if appropriate -
providing them with support rather than a formal criminal justice response). 160 Several local
areas also began to pilot their own approaches to diverting young people out of the formal
justice system.161
The causes of crime?
These criminal justice policies sat alongside actions to tackle the drivers of crime such as
drugs and alcohol. Drug testing and treatment requirements were built into the criminal
justice process. Some of these programmes took effect from the point of arrest, where the
suspect would immediately be referred for drug treatment if they tested positive. Others
were part of a community sentence. These programmes, known collectively as the Drug
Interventions Programme are discussed in more detail in Chapter 8. That chapter also
summarises the evidence of their cost-effectiveness, and impact in reducing offending
behaviour amongst those treated.162
The government’s wider efforts to tackle the causes of crime spanned many other
departments. The policies most relevant to young people are set out in other chapters of
this report including: efforts in the education system to help more young people thrive in
education, cut exclusion and cut school absence (see Chapters 1, 4 and 5); initiatives to
reduce alcohol and drug use by young people (Chapters 7 and 8); and a wide range of
broader child and youth policies discussed in Chapter 3.

156
Sir Ronnie Flanagan, The Review of Policing: Final Report (Criminal Justice Inspectorates, 2008).
157
House of Commons Home Affairs Committee, Police Funding (HoC, 2007).
158
As the senior Home Office official on criminal justice from 2002 to 2005 and for crime and policing from
2005 to 2008, I was involved in this area both while the Offences Brought to Justice target was in force, and in
the period of its abandonment.
159
HM Treasury, PSA Delivery Agreement 14: Increase the Number of Children and Young People on the Path to
Success; HM Government, Youth Crime Action Plan.
160
HM Government, Youth Crime Action Plan.
161
A Haines and others, Evaluation of the Youth Justice Liaison and Diversion Scheme: Final Report (University
of Liverpool, 2012); Roger Smith, ‘Re-Inventing Diversion’, Youth Justice, 14.2 (2014), 109–21.
162
Tim McSweeney, Paul J Turnbull, and Mike Hough, The Treatment and Supervision of Drug-Dependent
Offenders: A Review of the Literature (UK Drug Policy Commission, 2008).

70
Coalition and Conservative policies, 2010 - 2015, and 2015 onwards:
overview
Overall approach
The priorities for crime and policing in the Coalition Agreement focused on giving police
forces greater freedom from ministerial control and making them more accountable to the
public they served. There were pledges to reduce bureaucracy and introduce better
technology. Local accountability was to be led by ‘a directly elected individual’.163 No
quantified targets for crime reduction were set at national level between 2010 and 2019.
Resources and police numbers
The 2010 spending review announced that police resource funding was to fall by 14 per cent
in real terms by 2014/15, with savings to be made from efficiencies in IT, procurement and
back-office functions. Policing would be overseen locally by democratically elected Police
and Crime Commissioners (enacted in 2012) and the government would support
productivity by ending central targets and cutting out bureaucracy. 164
However, in the event, the impact of this spending settlement was not confined to back-
office functions. Police officer numbers fell by 12 per cent between 2010 and 2015,
dropping to below the 1997 level, a cut of more than 17,000 officers.165 Community support
officer numbers fell by more than a quarter from their 2010 level. The number of both
police officers and community support officers continued to fall under the Conservatives
after 2015, although officer numbers rose slightly in 2019 compared with 2018. A
programme was announced in 2019 to recruit an additional 20,000 officers by 2023.166 The
trend in police workforce numbers over the period to 2019 is shown below in Figure YJC1.
Figure YJC1: Trends in police workforce (full-time equivalents) in England and Wales

300,000

250,000

200,000

150,000

100,000

50,000

Police officers Police Community Support Officer Police staff and designated officers
Source: Police Workforce Statistics, March 2021

163
HM Government, The Coalition: Our Programme for Government.
164
HM Treasury, Spending Review 2010 (HMT, 2010).
165
Home Office, Police Workforce Statistics
166
National Audit Office, The Police Uplift Programme (NAO, 2022).

71
The wider criminal justice system
Other criminal justice agencies also faced spending reductions in the 2010 spending review.
Over the four years to 2014/15 the Ministry of Justice budget fell 23 per cent in real terms,
and there was a 24 per cent real cut in the Law Officers’ Departments (which include the
Crown Prosecution Service and the Serious Fraud Office). From 2014, the Probation Service
was split in two, with parts of its operation transferred to privately run Community
Rehabilitation Companies. The effectiveness and value for money of this programme was
widely criticised, with the Chief Inspector of Probation describing it as ‘irredeemably flawed’
and the National Audit Office finding it had cost at least £467 million more than expected
and achieved poor value for money for the taxpayer.167 It was reversed in 2019.
Youth justice: more diversion, fewer court cases
Youth justice was already changing fast when the Coalition entered office, as the emphasis
had shifted from formal justice-based approaches to an ethos of maximum possible
diversion. To reverse this policy would have required more resources, and the Coalition did
not seek to do this: instead it embraced the approach, with an early policy statement
arguing for informal approaches and promising to legislate ‘to allow police and prosecutors
greater discretion in dealing with youth crime before it reaches court’ and ‘to end the
current system of automatic escalation’ - code for ending the final warning system.168
This change was made law in the Legal Aid, Sentencing and Punishment of Offenders Act
2012. Since that legislation, there are two statutory out-of-court disposals for offenders
who are not charged - youth cautions and youth conditional cautions. Both of these are
recorded and count in the data as a young person entering the criminal justice system. A
third option – community resolution – is informal and non-statutory, and does not lead to a
formal criminal record. The use of this third option is widespread: a 2018 Joint Inspectorates
report found that community resolutions accounted for 39 per cent of the Youth Offending
Team cases examined.169 But nationally, there is no published data on trends in the scale of
community resolutions, their impact, or any disproportionality in how they are used. This
should change soon as the Youth Justice Board has been collecting data from April 2020 and
is due to publish it in due course.170
With greater diversion, far fewer young defendants went through the courts in a formal
criminal justice process. For those that did, processing times rose. Between 2011 and 2019,
the average time from offence to completion for youth criminal cases rose again from 70 to
118 days.171

167
Glenys Stacey, Annual Report of the Chief Inspector of Probation (Criminal Justice Joint Inspection, 2019);
National Audit Office, Transforming Rehabilitation: Progress Review (NAO, 2019).
168
Breaking the Cycle: Effective Punishment, Rehabilitation and Sentencing of Offenders (Ministry of Justice,
2010).
169
H M Inspectorate of Probation and H M Inspectorate of Constabulary, Out-of-Court Disposal Work in Youth
Offending Teams (Criminal Justice Joint Inspection, 2018).
Ministry of Justice, Children and Young People in Custody: Entry into the Youth Justice System: Government
170
Response to Justice Committee’s Twelfth Report of Session 2019 – 21 (HoC, 2021).
171
Ministry of Justice, Youth Justice Statistics: 2018 to 2019 Additional Annexes
https://www.gov.uk/government/statistics/youth-justice-statistics-2018-to-2019
72
The Troubled Families programme
A new Coalition initiative with a potential bearing on crime was the Troubled Families
Programme, which began in 2012 and funded local services to work together with families
experiencing multiple problems. The list of issues to be addressed included (in the first
round) crime or anti-social behaviour, unemployment, and children not attending school. A
second round of funding extended the programme from 2015 to 2020, and added other
issues to the criteria for inclusion.
Formal evaluation of the programme progressed in stages. The first evaluation found
overall, across nearly all outcome measures ‘no consistent evidence that the Troubled
Families Programme had led to any systemic or significant improvements in families’
outcomes over the period that it was possible to observe changes’. The exception to this
was that significantly more of the families going through the programme reported they
were managing well financially compared with the comparison group.172 A later evaluation
of the programme found more impacts, including – in relation to crime - fewer adults and
juveniles receiving custodial sentences compared with the comparison group.173
The causes of crime and prevention
Under the Coalition and Conservatives, austerity and reduced priority affected many other
public services that have a bearing on the causes of crime. Youth crime prevention was no
exception: central government grants to youth offending teams halved between 2010/11
and 2017/18, and the ring-fence which earmarked a fifth of this funding for prevention was
removed.174 As other chapters of this report describe, over a similar period, drug and
alcohol treatment funding declined, the Connexions service was effectively abolished,
spending per pupil in secondary schools started to fall in real terms, pressure on local
government led to significant reduction in early intervention and youth service budgets, and
rates of permanent exclusion and severe absence rose.
Serious violence and the county lines problem
From 2014 onwards, reports began to emerge of a new model of drug supply known as
‘County Lines’ which involved significant recruitment of young people as runners, and was
associated with high levels of violence between competing gangs. This was widely believed
to be a factor in rising levels of knife crime amongst young people. The government
response to this issue, through the Conservative government’s Serious Violence Strategy, is
discussed in Chapter 8 of this report.

172
Laurie Day and others, National Evaluation of the Troubled Families Programme Final Synthesis Report
(DCLG, 2016).
173
Ministry of Housing Communities and Local Government, National Evaluation of the Troubled Families
Programme 2015-2020: Evaluation Overview Policy Report (MHCLG, 2019).
174
Manon Roberts, Gemma Buckland, and Harvey Redgrave, Examining the Youth Justice System : What Drove
the Falls in First Time Entrants and Custody, and What Should We Do as a Result? (Crest Advisory, 2019).

73
Outcomes
We now turn to the data about trends in crime. We will consider first of all the picture on
overall crime, and then what we know about trends in youth crime.
Overall crime
Overall crime, as measured by the Crime Survey for England and Wales, grew through the
1980s and early 1990s to peak in 1995, at 19.8 million offences. It had begun to fall by 1997,
then fell by 44 per cent between 1997 and 2010. Between 2010 and 2017 it fell by 36 per
cent, then was stable until mid-2019. (Figure YJC2)
Figure YJC2: Trends in crimes as measured by the Crime Survey for England and Wales

20,000

18,000

16,000

14,000
Thousands of crimes

12,000

10,000

8,000

6,000

4,000

2,000

Crime excluding fraud and computer misuse

Crime including fraud and computer misuse
Notes:
1. Year to December in 97 and 99. Year to March thereafter
Source Crime in England and Wales, year ending March 2021 2. Figure including faud and computer misuse only from
- Appendix tables, Table A1 2017

All the main volume crime types covered in the survey - violence, theft, and criminal
damage - saw significant reduction over the period, illustrated overleaf in Figure YJC3.
Most of the targets set by Labour for specific crime reductions were met. By 2005, the
Home Office reported that it had beaten its PSA objectives for overall crime, vehicle crime,
and burglary, and reduced the gap between high crime areas and others, but had not met
its robbery target. 175

175
Home Office, Departmental Report 2004/05 (HO, 2005).

74
Figure YJC3: Trends in main volume crimes included in the Crime Survey for England and
Wales

12,000

10,000
Thousands of crimes

8,000

6,000

4,000

2,000

Violence Theft offences Criminal Damage

Source Crime in England and Wales, year ending March 2021 - Appendix tables, Table A1

Recent trends in overall crime

The two charts above portray a rather flat trend towards the end of the period. The Office
for National Statistics, who are responsible for the Crime Survey, interpreted the trend from
2017 until mid-2019 as ‘broadly stable’.176 The March 2020 figures then estimated a further
fall.177 The latest crime figures at the time this report was finalised were for the year ending
September 2022. These figures showed a decrease compared with the March 2020 survey
data. However the ONS advise that caution is to be taken when using these data, which are
not National Statistics. The figures cover some experiences of crime that took place during
the social restrictions associated with the pandemic, and were also affected by lower
response rates during fieldwork.178
Recorded violence, knife crime, homicide
On recorded crime (which measures more serious but less frequent crimes) the ONS judged,
as of 2020, that recent increases in recorded crimes of violence were driven largely by
improvements in recording but that there had been ‘a real rise in the occurrence of knife or
sharp instrument offences’. (See Figure YJC4.) 179
The homicide rate (all ages) ended a long period of increase in 2002 at a rate of 15.1 per
million. Between 2002 and 2015 it fell, to low point of 8.8 per million population in the year

176
Office for National Statistics, Crime in England and Wales: Year Ending September 2019 (ONS, 2019).
177
Office for National Statistics, Crime in England and Wales, Year Ending March 2020 (ONS, 2020).
178
Office for National Statistics, Crime in England and Wales: Year Ending September 2022 (ONS, 2023).
179
Office for National Statistics, The Nature of Violent Crime in England and Wales: Year Ending March 2020
(ONS, 2021).

75
ending March 2015. The rate then increased until the year ending March 2018 (11.9 per
million) before a fall to 11.0 per million in the year ending March 2019.180
Figure YJC4: Trends in offences involving knife or sharp instrument in England and Wales

180

160

140

120

100

80
Indexed figures: March
60
2011 = 100

40

20

Police Recorded Crime NHS admissions data

Source: ONS, The nature of violent crime in England and Wales: Year ending March 2020, Figure 5

What can we say about trends in youth crime?

It is not possible to know from any of this data how much crime has been committed by
young people. Trends in youth crime can only be inferred through other indicators which
we will look at in turn.
Young people in the criminal justice system
Data about young people in the criminal justice system is the first indicator to look at. It
presents a dramatic, but distorting, picture. Figure YJC5 below shows this: the number of
‘first time entrants to the criminal justice system’ has fallen by over 85 per cent since
2000/01. However, the series does not compare like with like, for reasons alluded to earlier.
First, in the mid-2000s, the ‘offences brought to justice’ target inflated the trend, because
more minor offending was drawn into the criminal justice system. Later, the opposite
happened, as a large but unmeasured number of minor offences by young people came to
be dealt with more informally. So, although there probably has been a fall in youth
offending since 2000, no-one seriously suggests the fall has been 85 per cent.

180
Office for National Statistics, Homicide in England and Wales: Year Ending March 2020, Office for National
Statistics (ONS, 2021).

76
Figure YJC5: Trends in first-time entrants to the criminal justice system in England and
Wales (10-17 year olds receiving first reprimand, warning, caution or conviction)

120,000
110,815
Introduction of
100,000 90,180 community
resolution
Objective
to reduce
80,000 number of
10-17 year
old first-
60,000 time
entrants to
the CJS
40,000
Offences brought
to justice target in
20,000 operation 02-07 11,928

Source: MOJ, Youth Justice Statistics

More detail on the composition of this trend can be found in the annual Youth Justice
Statistics published by the Youth Justice Board. The reductions have been greater for girls
than for boys, and there has been widening ethnic disproportionality. Youth Justice Board
analysis for 2016/17 found that Black Asian and minority ethnic children and young people
made-up 24 per cent of first-time entrants to the criminal justice system in the year ending
March 2017, while only representing 18 per cent of the 10- to 17-year-old population.181
Other data on youth crime trends
There are several other data sources we can look at to try to understand underlying trends
in youth crime.
Self-report and cohort data
First, we can look at self-report studies. Figure YJC6 below shows the trend from a Youth
Justice Board survey conducted in several years in the 2000s in mainstream schools. This
shows that self-reported offending by 10-16 year olds rose between 1999 and 2005, then
fell by a third at some point between 2005 and 2009. (This study was not continued, so
there is no data after this point.)

181
Youth Justice Board and Ministry of Justice, Youth Justice Statistics 2016/17 England and Wales (MOJ and
YJB, 2018).

77
Figure YJC6: Trends in self-reported offending by 10-16 year olds in mainstream schools

30 26 26 26 27
25
25 22 23
per cent of pupils

survey not run in

20 2006 and 2007
18

15 12 survey not
11 continued
10 8

Committed at least one offence 5 or more offences

Source: Youth Justice Board: Youth Survey 2009

Other data collected in similar ways also shows signs of a substantial fall in youth offending
after the mid 2000s:

• Analysis of the Offending Crime and Justice Survey in 2009 found evidence of a
decline in offending rates between cohorts. Young people born between 1992 and
1996 had a lower rate of offending at age 12 or 13 than the generation born
between 1989 and 1991. (This study was also not continued, so there is no data after
this point.)182

• A slightly longer time period can be considered by comparing the first and second
LSYPE cohorts, who were aged 14/15 in, respectively, 2005 and 2014. Comparison of
these cohorts found large falls between cohorts in the proportion who had engaged
in graffiti, vandalism and shoplifting. (Graffiti fell from 6 to 2 per cent; vandalism
from 9 to 2 per cent; and shoplifting from 8 to 3 per cent.) 183

• Similarly, comparison of the Millennium Cohort at age 14 (in 2015) with the ALSPAC
cohort at the same age in 2005 found that most anti-social behaviours were
substantially lower in the later cohort.184

182
J Hales and others, Longitudinal Analysis of the Offending, Crime and Justice Survey 2003--06 (Home Office,
2009).
183
Carli Lessof and others, Longitudinal Study of Young People in England Cohort 2 : Health and Wellbeing at
Wave 2 (DfE, 2016).
184
Praveetha Patalay and Suzanne H. Gage, ‘Changes in Millennial Adolescent Mental Health and Health-
Related Behaviours over 10 Years: A Population Cohort Comparison Study’, International Journal of
Epidemiology, 48.5 (2019), 1650–64.

78
Victims’ accounts of offender age
An alternative source of insight comes from victim reports. The Crime Survey has for years
asked victims what they can say about the person who committed the crime against them
(in cases where they encountered the offender). Between 2007 and 2019 there has been a
significant reduction in the proportion of victims saying that their offender was under 16 (or
between 16 and 24) and a proportionate increase in the number saying the offender
appeared to be over 25.
Figure YJC7 below shows this phenomenon for the broad group of offences classified as
criminal damage. Over that period, the proportion of offenders who were thought to be
under 16 halved, while the proportion thought by victims to be older increased. Figure YJC8
shows answers to the same question for Crime Survey respondents who had been victims of
violent offences. For these offences too, the proportion of offenders who were thought to
be under 16 or 16-25, fell (at least until 2014 or 2015) while the proportion thought to be
older increased.
Figure YJC7: Trends in age of offender as estimated by victims of criminal damage, England
and Wales

Proportion of criminal damage victims Proportion of criminal damage victims

who think offender was under 16, or who think offender was over 40 or
16 to 24 aged 25 to 29

60 60

50
50 50

40 40
34
40 27

30 30

20 20 13
24
21

10 10

7
0 0

Aged under 16 Aged 16 to 24 Aged 25 to 39 Aged 40 or older

Figures may not sum to 100 as more than one response may be given. Source: Crime Survey for England and Wales,
Nature of Crime Tables

79
Figure YJC8: Trends in age of offender as estimated by victims of violence, England and
Wales
Proportion of victims who think Proportion of victims who think
offender was under 16, or 16 to 24 offender was over 40, or 25 to 29
60 60
50
50 50

37
40 40
34

30 27 30
Per cent

30
20 20
12 13
8
10 10

0 0

2015

2017

2019
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014

2016

2018
1999

2014
1998

2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013

2015
2016
2017
2018
2019

Aged under 16 Aged 16 to 24 Aged 25 to 39 Aged 40 or older

Figures may not sum to 100 as more than one response may be given. Source: Crime Survey for England and Wales,
Nature of Crime Tables

So did youth crime fall?

This data does not prove definitively that youth crime fell over the last two decades, but it
does suggest a fall in prevalence of several forms of youth offending. In a study covering
both Scotland and England and Wales, Lesley McAra and Susan McVie reach a similar view
(describing the evidence as ‘tentative, albeit not conclusive’).185 An analysis by Tim
Bateman also reaches the same conclusion, using similar evidence, and also trends in youth
victimisation. 186 (Trends in young people’s experience as victims are covered in the
Appendix to this report.)
Intriguingly though, even if we suspect that there has been a significant fall in youth crime in
England and Wales, we are unable to gauge its scale over the period as a whole. This means
that it could be a very significant contributor to the crime drop, without our knowing it.
McAra and McVie note that despite many studies of the crime drop showing
disproportionate falls in arrests and convictions amongst young people, there has as yet
been no systematic examination of the extent to which crime falls are accounted for by

185
Lesley McAra and Susan McVie, ‘Transformations in Youth Crime and Justice across Europe’, in Juvenile
Justice in Europe: Past, Present and Future, ed. by Barry Goldson (Routledge, 2018).
186
Tim Bateman, The State of Youth Justice 2020 (National Association for Youth Justice, 2020). See also Alex
Sutherland and others, An Analysis of Trends in First Time Entrants to the Youth Justice System (Ministry of
Justice, 2017).

80
young people, and no specific theory that identifies young people as a primary driver of the
crime drop.187 188
And is youth crime still falling?
The idea that the underlying level of youth offending has fallen is an encouraging one. But
most of the evidence on which it is based is now rather dated. Offence data from the second
half of the 2010s confirms that knife offences among young people rose after 2014. This
data, only collected in this form since the late 2000s, shows that cautions and sentences of
10- to 17-year-olds for knife offences fell until 2013, then rose again by 67 per cent between
2014 and 2019.
Figure YJC9: Trends in knife and offensive weapon offences committed by 10- to 17-year-
olds, resulting in caution or sentence
7,000

6,000

5,000

4,000
Title

3,000

2,000

1,000

Source: MOJ, Youth Justice Statistics 2018/19, Table 4.7

Looking ahead, the Ministry of Justice now forecasts that the number of children in custody
will increase, with the 15- to 17-year-old population in custody expected to double between
2021 and 2025. This increase partly reflects post-Covid recovery in the courts, and new
legislation, but also an expected increase in police activity when officer numbers increase.189

Discussion
The sequence of events presented in this chapter in some respects resembles the picture
seen in other chapters of this report, with significant policy and funding differences
between Labour and its successors, as rising funding gave way to austerity, and an
interventionist outcome-based national approach shifted to one based on devolution to
local areas. But in other respects, the policy picture is more complicated, with Labour first
toughening, then softening, the criminal justice approach to young offenders, and the trend
to more diversion continuing and growing under the Coalition and Conservatives.
It is beyond doubt that this period saw a reduction in overall crime. This began in the mid-
1990s, before Labour took office, and seems to have continued, although with fluctuations.

187
McAra and McVie.
188
See also André M. van der Laan and others, ‘The Drop in Juvenile Delinquency in The Netherlands: Changes
in Exposure to Risk and Protection’, Justice Quarterly, 38.3 (2021), 433–53.
189
National Audit Office, Support for Vulnerable Adolescents.

81
This crime drop is not unique to England and Wales, and most developed countries have
enjoyed similar phenomena to some extent over the last three decades, albeit with
variations in timing, and disagreement as to whether it applies to all types of crime.190
Many theories have been advanced to explain these reductions in overall crime and
violence, in this country and worldwide.191 Possible causes include a variety of policy
initiatives and social and technological changes, such as:

• improvements in building and vehicle security 192

• changes in policing practice193
• improved partnership working and targeting of alcohol-driven crime194
• the impact of drug ‘epidemics’ and drug treatment. 195 196
These explanations are not, of course, mutually exclusive and to some extent different
factors may have reinforced each other.
As discussed above, it is harder to know whether youth crime fell by the same amount, or
more, or less. Piecing together the evidence does suggest that youth crime fell markedly.
However, for the most recent years all we really have to go on is criminal justice data that
may be distorted by policy and resourcing changes in the criminal justice system. This
means that, in this policy area, there is an uncomfortable element of ‘flying blind’. In such
circumstances, it would be wise for policy makers to try to track youth crime as closely as
possible. A return to commissioning regular self-report studies on juvenile offending would
be a very good start.
Current crime trends are difficult to be confident about, given the after-effects of the
pandemic. But it will always make sense for policy makers to keep an eye on drivers of
crime such as drug use, alcohol use, as well as the trends in school exclusion and school
absence. These issues are explored further in later chapters of this report.

190
Jan van Dijk, Andromachi Tseloni, and Graham Farrell, The International Crime Drop: New Directions in
Research (Palgrave, 2012); Marcelo Aebi and Antonia Linde, ‘Is There a Crime Drop in Western
Europe?’, European Journal on Criminal Policy and Research, 16.4 (2010), 251–77.
Why Crime Rates Fall and Why They Don’t, ed. Michael Tonry (University of Chicago Press, 2014).
191

Graham Farrell, Nick Tilley, and Andromachi Tseloni, ‘Why the Crime Drop?’, Crime and Justice, 43.1 (2015)
192

421–90.
193
Anthony A. Braga, Brandon C. Welsh, and Cory Schnell, ‘Disorder Policing to Reduce Crime: A Systematic
Review’, Campbell Systematic Reviews, 15.3 (2019); Rachel Tuffin, Julia Morris, and Alexis Poole, An Evaluation
of the Impact of the National Reassurance Policing Programme (Home Office, 2006); David Weisburd and
others, ‘The Effects of Problem-Oriented Policing on Crime and Disorder’, Campbell Systematic Reviews, 4.1
(2008), 1–87.
194
Curtis Florence and others, ‘Effectiveness of Anonymised Information Sharing and Use in Health Service,
Police, and Local Government Partnership for Preventing Violence Related Injury: Experimental Study and Time
Series Analysis’, Bmj, 342.7812 (2011); Vaseekaran Sivarajasingam and others, ‘Trends in Violence in England
and Wales 2010 – 2014’, J Epidemiol Community Health, 2016, 616–21.
195
Nick Morgan, The Heroin Epidemic of the 1980s and 1990s and Its Effect on Crime Trends (Home Office,
2014).
196
McSweeney, Turnbull, and Hough, The Treatment and Supervision of Drug-Dependent Offenders: A Review
of the Literature. See especially Chapter 3. 82
Chapter 3 Child Poverty and Cross-cutting Youth
Policies
Introduction
This chapter completes the survey of ‘cornerstone’ areas of social policy affecting young
people by summarising policy developments in relation to child poverty, and cross-cutting
youth policies. The aim is to clarify different administrations’ overall ambitions for children
and young people, to set out their stance on issues such as prevention and the joining up of
services, and to paint more clearly the background context against which specific problems
of adolescence discussed in later chapters are played out.

Child poverty
Labour policies on child poverty: 1997–2010

Labour's approach to benefits was initially cautious and controversially so, declining to
reverse cuts to lone parent benefits that had been put in place by the previous
administration. In 1999, a different direction emerged, with the announcement of a target
to halve the number of children living in relative poverty, and the beginning of a series of
real terms increases in benefits to children both for families in and out of work. These
included increases in the level of both Child Benefit and Income Support allowances for
children, and the replacement of Family Credit (which supported low-earning families) with
a more generous ‘Working Families Tax Credit’.

Why does child poverty matter?

The rationale for Labour’s child poverty pledge rested in part on the links between
childhood poverty and other outcomes, on the basis that ‘children who grow up in
disadvantaged families generally do less well at school, and are more likely to suffer
unemployment, low pay and poor health in adulthood’.197

There is strong evidence that these links are causal. A 2013 systematic review of
controlled studies found clear indications that money makes a difference to children’s
outcomes, and that poorer children have worse cognitive, social-behavioural and health
outcomes in part ‘because they are poorer, and not just because poverty is correlated
with other household and parental characteristics’. The evidence supported two central
theories as to why income matters: one relating to the stress and anxiety caused by low
income, and the other relating to parents’ ability to invest in goods and services that
further child development. 198

197
Department for Social Security, Opportunity for All : Tackling Poverty and Social Exclusion : First Annual
Report (DSS, 1999).
198
Cooper and Stewart.

83
Further measures followed over the remainder of the decade to improve the incomes of
families with children both in and out of work. For example, tax credits for families with
children were uprated in line with earnings rather than just prices, and in some years the
indexation was higher still.199 200 Overall, compared with spending in 1996/97, by 2010/11
there had been a real terms increase of nearly £24 billion in cash transfers to families with
children.201 As well as benefit increases, the child poverty strategy included policies to
increase employment amongst lone parents, and increased access to childcare.202
One of the last acts of the Labour government was to attempt to embed the commitment to
child poverty reduction in the long term: the Child Poverty Act 2010 set in law four targets
for the future, relating to relative low income, a low-income measure fixed in real terms, a
combined low income and material deprivation measure, and a ‘persistent poverty’
measure.203
Coalition and Conservatives policies on child poverty: 2010 to 2019

The 2010 Coalition agreement signed by the incoming Conservatives and Liberal Democrats
said that the Coalition would maintain the goal of ending child poverty in the UK by 2020.
But a series of policy changes under the Coalition and Conservatives significantly affected
the benefit and tax credit system for families with children. Changes to the benefits system
included:

• from 2011 onwards, freezes or under-uprating of child benefit, changes to the

uprating of most working age benefits, changes to make tax credits less generous
• from 2013, introduction of a cap on the total amount of benefits that working age
people and families can receive.
• from 2013, the ‘bedroom tax’
• from 2016, removal of the family premium in housing benefit, and localisation and
shrinking of Council Tax Benefit
• from 2016, cuts to the work allowance in Universal Credit and reduction of the
benefit cap
• from 2017, introduction of the two-child limit.204

199
HM Treasury, Autumn Performance Report Progress Report on HM Treasury Public Service Agreement
Targets (HMT, 2007).
200
For more detail on benefits indexation compared with RPI and Rossi measures, see Table 5.1 in Mike
Brewer and others, Child Poverty in the UK since 1998-99: Lessons from the Past Decade (IFS, 2010).
201
John Hills, Labour’s Record on Cash Transfers, Poverty, Inequality and the Lifecycle 1997 - 2010 (Centre for
the Analysis of Social Exclusion, LSE, 2013). p13.
202
Jane Waldfogel, Britain’s War on Poverty (Russell Sage Foundation, 2010).
203
Steven Kennedy, Child Poverty Act 2010: A Short Guide (House of Commons Library, 2014).
204
John Hills, The Coalition’s Record on Cash Transfers, Poverty and Inequality 2010-2015: Social Policy in a
Cold Climate Working Paper 11 (Centre for the Analysis of Social Exclusion, LSE, 2015); Child Poverty Action
Group, Universal Credit: Cuts to Work Allowances (CPAG, 2016); Kitty Stewart and Mary Reader, The
Conservatives’ Record on Early Childhood: Policies, Spending and Outcomes from May 2015 to Pre-COVID 2020
(Centre for the Analysis of Social Exclusion, LSE, 2020).

84
The result of these changes was that between 2009/10 and 2018/19 per capita social
security spending on children fell by 25 per cent. 205
The Coalition and Conservative governments made a series of other policy changes that
were justified on the basis of the benefit to low earners – such as raising income tax
thresholds and increasing the minimum wage. However, the benefit of these for poor
working households was more than offset by changes to in-work benefits and tax credits.206
Alongside other changes in the economy, this meant that the decade saw a reduction in the
number of workless households, but this did not flow through into lower child poverty. 207
The measures described above made it impossible that the government would achieve the
targets set in the Child Poverty Act 2010. The Cameron government legislated in 2016 to
replace the previous child poverty targets with indicators which did not include income
poverty.208 Reflecting this repositioning, the ‘Child Poverty Commission’ established by the
Child Poverty Act 2010 was renamed first the ‘Social Mobility & Child Poverty Commission’
in 2012 and then (from 2016) the ‘Social Mobility Commission’.209
Child poverty trends

All children (0-18)

The overall trends in child poverty are set out in full elsewhere in other papers by CASE.210
To summarise, relative poverty for children fell under Labour, although the target of halving
the rate was not met. Relative child poverty was a third lower in 2010-11 than it had been in
1996-97, (18 per cent of children, down from 27 per cent of children). Most of this
reduction had been achieved by 2004-05, with tax and benefit reforms playing a significant
role.211 Under the Coalition and Conservatives, child poverty on this measure fell slightly
further to 17 per cent in 2013-14, then rose again, reaching 20 per cent in 2018-19. 212

205
Kerris Cooper and John Hills, The Conservative Governments’ Record on Social Security : Policies, Spending
and Outcomes, May 2015 to Pre-COVID 2020 (Centre for the Analysis of Social Exclusion, LSE, 2021).
206
William Elming and others, An Assessment of the Potential Compensation Provided by the New ‘National
Living Wage’ for the Personal Tax and Benefit Measures Announced for Implementation in the Current
Parliament (IFS, 2015).
207
Cooper and Hills.
208
Welfare Reform and Work Act 2016. The indicators are published at
https://www.gov.uk/government/publications/workless-households-and-educational-attainment-statutory-
indicators-2020
209
For a further discussion of its role, reports and recommendations, see Polly Vizard, The Conservative
Governments’ Record on Social Policy from May 2015 to Pre-COVID 2020: Policies, Spending and Outcomes
(Centre for the Analysis of Social Exclusion, LSE, 2021).p201-211.
210
Hills, Labour’s Record on Cash Transfers, Poverty, Inequality and the Lifecycle 1997 - 2010; Hills, The
Coalition’s Record on Cash Transfers, Poverty and Inequality 2010-2015: Social Policy in a Cold Climate Working
Paper 11; Cooper and Hills; Polly Vizard, Polina Obolenskaya, and Kritika Treebhoohun, Going Backwards? The
Slowdown, Stalling and Reversal of Progress in Reducing Child Poverty in Britain during the Second Decade of
the 21 St Century, and the Groups of Children That Were Affected (Centre for Analysis of Social Exclusion
(CASE), 2023).
211
Brewer and others. See esp Table 5.1.
212
Department of Work and Pensions, Households Below Average Income 2018/19 (DWP, 2020). Table 4.1tr.

85
Did older and younger children in poverty gain equally?
Figure CP 1 below shows how the relative child poverty trends look when analysed by age.
At the beginning of Labour’s term, child poverty rates were much higher for 0- to 10-year-
olds than for 11- to 18-year-olds: the reduction that followed was larger for 0- to 10-year-
olds. Both age groups had higher relative child poverty rates in 2018/19 than in 2010/11.
Figure CP1: Relative poverty: proportion of children living in households on incomes
below 60 per cent of the contemporary median: 1996-97 to 2018-19

0 to 10 year olds Young people aged 11 to 18

40% 40%
36.0%

35% 30.0% 35% 30.2%

27.5% 26.9% 29.3%
30% 30%

25% 25%

20% 20%

15% 15%

10% 10%

5% 5%

0% 0%

Source: Relative poverty measure, 60% of contemporary median, after housing costs. Calculated by Kitty Stewart from DWP,
Households Below Average Income, 2020.

Relative and absolute child poverty

The trends in relative measures of child poverty are rather bumpy because they are affected
both by the circumstances of those on low incomes and by movements in average incomes
across the economy. Trends in earnings explain some of the fluctuations in the graphs
above.
A different way of looking at poverty is against a fixed real income threshold - 60 per cent of
median income in a reference year, converted to the real terms equivalent in other years.
The graphs in Figure CP2 below take this approach, showing a substantial reduction in child
poverty during the period until 2004, then a levelling off. By 2010-11, the proportion of the
under-11 population whose incomes were below this absolute poverty line was 22
percentage points lower than in 1996-97. For 11–18-year-olds the reduction was 17
percentage points. Both age groups saw slight reductions in absolute poverty (one
percentage point) in the period from 2010-11 to 2018-19.

86
Figure CP 2: Absolute child poverty: proportion of children living in households below 60%
of a fixed (2010-11) median in constant prices. 1996-97 to 2018-19

0 to 10 year olds Young people aged 11 to 18

60% 60%
50.0%
50% 50% 44.3%

40% 40%
29.7% 27.5% 26.8% 29.3% 26.9% 25.9%
30% 30%

20% 20%

10% 10%

0% 0%

2004-05

2013-14
1996-97
1997-98
1998-99
1999-00
2000-01
2001-02
2002-03
2003-04

2005-06
2006-07
2007-08
2008-09
2009-10
2010-11
2011-12
2012-13

2014-15
2015-16
2016-17
2017-18
2018-19
2018-19
1996-97
1997-98
1998-99
1999-00
2000-01
2001-02
2002-03
2003-04
2004-05
2005-06
2006-07
2007-08
2008-09
2009-10
2010-11
2011-12
2012-13
2013-14
2014-15
2015-16
2016-17
2017-18

Source: Fixed income measure ( 60% of 2010-11 median held constant), after housing costs. Calculated by Kitty Stewart
from DWP, Households Below Average Income, 2020.

International comparisons
Over this two-decade period, the UK’s child poverty rankings first improved relative to other
countries, then declined. A study by Bradshaw and Main analysed the 23 country
Luxembourg Income Study between the mid-1990s and 2010 and found that only eight of
the countries covered had reduced child poverty, with the UK seeing the largest
reduction.213 More recently, OECD figures found that between 2008 and 2013 the United
Kingdom had the largest reduction in relative child poverty of over 30 countries studied, but
between 2013 and 2018 had the third largest increase.214

Cross-cutting youth policy

The section that follows discusses the extent of joined-up working arrangements on young
people’s issues under Labour, the Coalition and the Conservatives.
‘Youth policy’ and ‘cross-cutting youth policy’ are perhaps unfamiliar terms in the English
policy context. Governments do not always articulate in one place their policies in relation
to children or teenagers, or manage them as a single system. But whether or not a
government thinks of it that way, the sum of what government does, as it affects young
people, is its youth policy, and that is the picture this report tries to paint.

213
Jonathan Bradshaw and Gill Main, ‘Child Poverty and Deprivation’, in The Well-Being of Children in the UK,
ed. by Jonathan Bradshaw (Policy Press, 2016), pp. 31–69.
214
OECD, Table CO2.2 Child Poverty, 2021, <https://www.oecd.org/els/CO_2_2_Child_Poverty.pdf> [accessed
19 January 2023].

87
The problems and opportunities which joined-up working seeks to address have several
elements, set out in the box below. Although the problems and opportunities are described
here in relation to policy on adolescents, they apply equally strongly to policy on children.215

Why do joined-up youth policies and services matter?

- Adolescence is a time of transition during which young people mature, acquire the
skills for independence, and adopt habits that can influence their health and
welfare long into the future.216 It benefits society as well as young people
themselves if this transition goes well. Conversely, it costs society, and young
people themselves, if children and teenagers become stuck in problems that
prevent them thriving in education, damage their physical or mental health, or
limit their future prospects. 217

- Some young people will solve these problems for themselves, or with help from
their family. But young people with problems can’t always call on these resources
and may struggle to navigate their way around the system and advocate for what
they need.

- Many of the problems that cause difficulty for teenagers can be prevented or
mitigated if their early signs, or important risk factors, are addressed appropriately
and at the right time. Shifting from reactive to preventive spending has the
potential to improve value for money.218 This makes a strong case for moving
resources upstream, spotting early signs of risk, and providing effective support at
the right time.

- Collaboration and coordination are important foundations for prevention. This is

partly because prevention efforts tend to work best if they address multiple risk
and protective factors, so it helps if services work together.219 It is partly about
access, with a ‘no wrong door’ approach helping young people and their families
access help, whichever door they first approach.220 And it is partly because of the
costs of prevention: it may be easier to persuade public services to club together,
or to do more to minimise the future workload of another service, if they have
shared objectives, and have developed experience of joint commissioning or
pooled budgets. 221

215
Naomi Eisenstadt and Carey Oppenheim, Parents, Poverty and the State (Policy Press, 2019).
216
Jane Waldfogel, What Children Need (Harvard University Press, 2006).
217
H. Chowdry and P Fitzsimons, The Cost of Late Intervention: EIF Analysis 2016 (London: Early Intervention
Foundation, 2016).
218
National Audit Office, Early Action: Landscape Review (NAO, 2013).
219
Caroline A. Jackson and others, ‘An Overview of Prevention of Multiple Risk Behaviour in Adolescence and
Young Adulthood’, Journal of Public Health, 34.S.1 (2012), 31–40; Daniel R. Hale, Natasha Fitzgerald-Yau, and
Russell Mark Viner, ‘A Systematic Review of Effective Interventions for Reducing Multiple Health Risk
Behaviors in Adolescence’, American Journal of Public Health, 104.5 (2014), 19–42; Advisory Council on the
Misuse of Drugs, Prevention of Drug and Alcohol Dependence, 2015.
220
Social Exclusion Task Force, Think Family : Improving the Life Chances of Families at Risk (SETF, 2008).
221
Janet Walker and Cam Donaldson, Intervening to Improve Outcomes for Vulnerable Young People: A Review
of the Evidence (Department For Education, 2010).

88
 - Collaboration and coordination are particularly important in the case of the most
vulnerable who are often ‘off the radar’ entirely, or in only infrequent contact with
statutory services. This is itself is a red flag for all kinds of serious problems.
Identifying and addressing the needs of children who are missing from school or
otherwise missing out on services and protection requires strategic, proactive, and
properly-resourced effort. Addressing this issue is bound to take sustained
investment, coordination, and leadership nationally and locally.222

- These points have been framed principally in terms of local services, but
coordination at national level is just as important. If coordination is lacking, it is
easier for departments to introduce policies in one area of young people’s lives
that may have a disproportionate negative impact in another aspect, to introduce
initiatives that duplicate each other, or to leave gaps that the most vulnerable will
fall through. Without collaboration, departments will not be able to see the full
picture of the adverse outcomes facing adolescents, or to coordinate their
objectives, accountability, governance and funding arrangements to help make
things better.223 If it is no-one’s job to lead this collaboration, it may not happen,
and there is a high risk of children and young people’s needs being overlooked.224

Labour policies: 1997 to 2010

Prevention
Labour’s policies to deal with income poverty sat alongside a set of initiatives designed to
address broader drivers of disadvantage. A 1999 cross-government plan, ‘Opportunity for
All’, criticised past governments for short-term and piecemeal approaches that did little to
prevent problems occurring in the future. It listed a range of policies across government
where it intended to invest more on prevention – including policies on early years
education, schools, post-16 options, worklessness, and family support. 225
Many preventive policies were specific to particular outcomes, such as crime, teenage
pregnancy, drug use, or school exclusion and absence. Later chapters will describe these
initiatives. But Labour also put resources into broader prevention programmes designed to
address the risk of multiple adverse outcomes. By 2000, a structure for this had emerged -
Sure Start for the under-fives; investment under the Children’s Fund for the 5-13 age group,
and ‘Connexions’ for 13-19s. This structure emerged from a Treasury led cross-cutting
review on ‘Young People at Risk’ whose results were announced in the 2000 Spending
Review.226 227

222
Commission on Young Lives; Children’s Commissioner, Where Are England’s Children? Interim Findings from
the Attendance Audit (CCO, 2022).
223
House of Commons Committee of Public Accounts; National Audit Office, Support for Vulnerable
Adolescents.
224
Commission on Young Lives.
225
Department for Social Security.
226
HM Treasury, Prudent for a Purpose : Building Opportunity and Security for All.
227
As Director of the Social Exclusion Unit, I was a member of the officials’ group working on this review.

89
Of these programmes, the most directly relevant to the age group we are studying was
Connexions, which by 2004 was staffed by over 7,700 Personal Advisers and more than
2,400 other front line delivery staff. It dealt not just with educational progression but also a
broad range of issues faced by vulnerable young people. Chapter 1 discussed the operation
of Connexions, and the evidence of its impact in more detail.
Coordination and leadership
The Young People at Risk section of the 2000 spending review also stressed the need for
prevention to be underpinned by new ways of working that crossed traditional institutional
boundaries. A report from the Social Exclusion Unit in 2000 had argued that policy making
and service delivery for young people was fragmented at national and local level, and that
new objectives and structures were required.228 229 The Spending Review took up this issue
announcing that the new resources for Connexions and some children’s programmes would
be overseen by a new Cabinet Committee on Children and Young People's Services, and
administered by a new Unit.230 This unit became the Children and Young People’s Unit and
supported coordination of work on youth policy across Whitehall.
In 2003, a larger change took place, when the June reshuffle moved into the Department for
Education and Skills (DfES) responsibility for children’s social services, family policy, teenage
pregnancy, family law, and the Children and Family Court Advisory and Support Service. The
Department acquired a new Ministerial role, and Margaret Hodge became the first Minister
for Children, Young People and Families. As part of this responsibility shift, Ofsted took over
inspection of children’s services and childcare to provide a more unified focus.
Later in 2003 the Every Child Matters Green Paper called for new commitment to multi-
agency work to protect children, built around common outcomes, shared assessment
criteria, integrated service delivery, and joint commissioning and budget pooling. The five
Every Child Matters outcomes were: being healthy; staying safe; enjoying and achieving;
making a positive contribution; and economic well-being. The Green Paper announced that
in each local authority a new post of Director of Children's Services should be accountable
for education and children's social services and signalled the intention to require local
authorities to integrate key services for children and young people through Children’s
Trusts.231
Other key measures included:

• the introduction of a ‘common assessment framework’ for children who needed

additional help, to promote joint working and reduce the number of assessments
required by children who had multiple needs.
• the designation of a ‘lead professional’ to act as a single point of contact for a child
and their family when multiple services are involved and an integrated response is
required.

228
Social Exclusion Unit, Report of Policy Action Team 12: Young People (Stationery Office, 2000).
229
As Director of the Social Exclusion Unit, I chaired the Policy Action team which recommended this.
230
HM Treasury, Prudent for a Purpose : Building Opportunity and Security for All.
231
HM Government, Every Child Matters (HMG, 2003).

90
 • the powers to create for a Children’s Information Sharing Index (later known as
ContactPoint). This implemented a recommendation of the Laming Inquiry.232 Its aim
was to help services identify a child with whom they had contact and establish
whether the child was getting the universal services (education, primary health care)
to which they were entitled.
A 2010 assessment found that most parents/carers interviewed were positive about the
common assessment framework and valued the lead professional role. Professionals
reported that the assessment framework had increased awareness of the range of services
available and brought professionals together in new contexts, but there were still
differences in culture and information-sharing, as well as workload challenges. The
evaluation of the early Children’s Trusts found signs of progress, with most areas jointly
commissioning children services, increased pooling of finances, and more multi-agency
working, concluding that Children’s Trusts had enabled major changes to services in areas
where there was local enthusiasm, but in some cases the remit of Children’s Trusts was too
broad to overcome entrenched divisions.233 The ContactPoint information sharing system
was launched in 2009, after piloting. The concept divided opinion: some thought it vital,
others a civil liberties or security concern, and it was cancelled by the Coalition in 2010.234
Implementation of the Every Child Matters programme was underway when, in 2007,
Gordon Brown decided to rename the Department for Education and Skills as the
Department for Children, Schools and Families. The department published a cross-
government Children’s Plan in 2007: this ranged widely over issues the government planned
to invest in, including support for parenting, facilities for disabled children, child safety,
improving special educational needs provision, support for behaviour in schools, improving
alternative provision, and investing in youth facilities.235 The plan committed to publishing
annual updates on progress and did so in 2008 and 2009.236
Local authority spending on young people
It is not a simple matter to track total spending on young people which may be undertaken
under many headings, both national and local. The largest budget items, education and
benefits for children have already been discussed. The best estimate of total local authority
spending on youth comes from the Institute of Fiscal Studies and Office of the Children's
Commissioner who combined several different methods to estimate that local authority
spending on children and young people’s services (excluding education) doubled in real
terms over the 2000s, growing from around £4.8 billion in 2000/01 to £9.7 billion in
2009/10, with the expansion of Sure Start accounting for £1.2 billion of this additional 4.9

232
Lord Herbert Laming, The Victoria Climbie Inquiry (Stationery Office, 2003).
233
Max O. Bachmann and others, ‘Integrating Children’s Services in England: National Evaluation of Children’s
Trusts’, Child: Care, Health and Development, 35.2 (2009), 257–65.
234
Manjit Gheera, The ContactPoint Database (House of Commons Library, 2011).
235
Department for Children Schools and Families, The Children’s Plan: Building Brighter Futures (DCSF, 2007).
236
Department for Children Schools and Families, The Children’s Plan One Year On: A Progress Report (DCSF,
2008); Department for Children Schools and Families, The Children’s Plan Two Years On: A Progress Report
(DCSF, 2009).

91
billion.237 This spending included statutory services for child protection and looked after
children, spending on youth justice, but also spending on the Youth Service (local
authorities’ spending on educational and recreational leisure-time activities for young
people).
Coalition and conservative approaches to cross-cutting youth policies: 2010 to 2019
Prevention and coordination
On coming into government, the Coalition renamed the Department for Children, Schools
and Families, to become the Department for Education once more. Prevention funding was
not protected within the renamed department’s priorities: the Connexions service largely
disappeared early in the Coalition’s term. Its funding for personal advisor support was
merged into the local authority Early Intervention Grant, but the grant was then cut by 10.9
per cent for 2011/12.238 In 2013/14 the ring-fence for early intervention was removed
entirely so that the funding became like any other part of revenue support grant.239
The collaboration requirements for local areas were becoming increasingly confusing. On
the one hand, the government instituted the Troubled Families Programme with a remit to
promote joined-up services to families with problems such as criminal behaviour,
unemployment, and children not attending school.240 This programme was extended on
several occasions then renamed as the Supporting Families programme. Evaluations of the
programme are discussed in more detail in Chapter 2 (in relation to crime) and in Chapter 5
(in relation to absence).
On the other hand, the government took several steps that deprioritised local collaboration.
Academisation had this effect anyway, as did the trend for academies to band together in
larger groups which were not always geographically close. But the Coalition’s 2011
Education Act also repealed the duty on schools to participate in Behaviour and Attendance
Partnerships, and the government attempted to remove the statutory requirement for
schools to cooperate with Children’s Trusts, but could not get the provisions through the
House of Lords.241 In 2017 when the Conservative government replaced Local Children’s
Safeguarding Boards with new arrangements, the new partnerships were no longer required
to include representation from schools.242
Overall expenditure
It was noted above that total local authority spending on children services grew significantly
during Labour's period in office, but it fell back again significantly post 2010 under the
pressure of the Coalition’s deficit reduction programme. The IFS and Children's
Commissioner analysis referred to above found that local authorities’ spending on children

237
Elaine Kelly and others, Public Spending on Children in England: 2000 to 2020 (Office of the Children’s
Commissioner, 2018).
238
Tristram Hooley and A.G. Watts, Careers Work with Young People : Collapse or Transition? An Analysis of
Current Developments in Careers Education and Guidance for Young People in England, 2011.
239
Carl Purcell, The Politics of Children’s Services Reform: Re-Examining Two Decades of Policy Change (Policy
Press, 2020). P135.
240
https://www.gov.uk/government/organisations/office-for-civil-society
241
Purcell. P131.
242
Children and Social Work Act 2017

92
services fell by 20 per cent in real terms between 2009/10 and 2019/20. Within that,
preventive services fell by a larger proportion, as spending was reoriented towards statutory
and crisis services such as child protection and looked-after children.243
Other studies conducted on a different basis found a similar story. In 2021 a study
commissioned by five children’s charities found that local authority spending on early
intervention had decreased by 48 per cent between 2010/11 and 2019/20.244 Research for
the YMCA found a real terms reduction of 71 per cent in youth service spending between
2010/11 and 2018/19 with the largest fall in 2011/12.245
Responsibilities at national level
As these spending pressures played out at local level, oversight and coordination of children
and youth policy at national level has taken a back seat. The Children and Young People’s
Directorate bore much of the burden of DfE’s departmental savings target in 2010, and the
Coalition’s first Children’s Minister complained in hindsight that it had been difficult for the
children and families agenda to get attention within the department compared with schools
reform.246
Youth responsibilities at national level are now very fragmented: in 2013, responsibility for
youth services left DfE for the Cabinet Office and then in 2016 moved to the Department for
Digital Culture Media and Sport. 247 Sponsorship of the Social Mobility Commission moved
from the Department of Education to the Cabinet Office in 2021-22. The Troubled Families /
Supporting Families Programme is led from the Department for Levelling Up, Housing and
Communities.
Lack of strategic coordination in relation to vulnerable adolescents was highlighted by a
critical National Audit Office report in 2022. Its key findings are set out in the box below.

Support for vulnerable adolescents – NAO report key findings248

Several government departments have lead policy responsibilities that address the
challenges facing vulnerable adolescents.249 […] The 2021 Spending Review announced £2
billion of additional spending, on various additional programmes which support families,
vulnerable adolescents and children, in addition to the ongoing funding for universal
services. This money was allocated to seven departments.

243
Elaine Kelly and others. p37
244
Max Williams and Jon Franklin, Children and Young People’s Services : Spending 2010-11 to 2019-20 (Pro
Bono Economics, 2021).
245
YMCA, Out of Service: Local Authority Expenditure on Youth Services in England and Wales (Local
Government Association, 2020).
246
Tim Loughton MP quoted in Purcell. p129.
247
https://www.gov.uk/government/news/cabinet-office-to-take-on-responsibility-for-cross-government-
youth-policy
248
National Audit Office, Support for Vulnerable Adolescents.
249
Department for Education, Home Office, Department for Levelling Up, Housing and Communities,
Department of Health and Social Care, Home Office, Ministry of Justice, Department for Work and Pensions,
Department for Digital Culture Media and Sport.

93
 Departments do not treat vulnerable adolescents as one group with a single, specific
cross-government policy programme. […] Departments use mechanisms such as 16 multi-
departmental boards to manage projects which cut across different department
responsibilities. […] There has been no overall analysis of the policies and programmes
that support vulnerable adolescents to identify gaps and overlaps in support and inform
policy development.

Local bodies implementing policies led by different departments are navigating complex
requirements which can lead to confusion at the local level. […] The multiple
programmes from government often impact the same local authority teams, who
themselves are then in some cases working with the same young people. […] The short-
term nature of funding makes it difficult to sustain projects and allow for long-term
planning.

In the November 2022 Public Accounts Committee hearing which discussed this report, the
DfE Permanent Secretary said that officials had set up a Vulnerable Children and Families
Programme Board during Covid and had now developed this into a strategy board. 250

Discussion
As this chapter demonstrates, responsibility for youth policy is often poorly defined and can
struggle to exist in the gaps between larger established departments. Many consequences
follow from this. A subject that has no home and no champion has no real power, no-one to
fight for spending budgets, and no select committee to challenge strategy and performance.
Lack of coordination can be a problem both in times of plenty and in times of famine.
Despite plentiful spending, Labour took time to settle coordination arrangements for policy
on children and young people. The profusion of initiatives and funds that they brought into
being must raise the question whether the same could have been achieved in a more
streamlined way with less money. By the time Labour left office, they had not managed to
embed a settled range of programmes for children and young people that were widely
known and understood, and that would hold their place in a more austere spending climate.
After 2010, recently invented and little-known programmes for young people were highly
vulnerable to budget cuts, and the consequences were left to play out in what was now a
more devolved and fragmented system.
The hope of Coalition ministers, apparently, was that their objectives for children would be
met through schools, that the five Every Child Matters outcomes were ‘what every teacher
will want to do’, and that a ‘massive bureaucratic superstructure’ was not needed to police
this.251 However, in the context of the incentive structures and funding pressures in schools
described in Chapter 1, and with significant number of children not attending school
because of absence, exclusion, or off-rolling (Chapters 4 and 5) this was unlikely to be a
realistic hope.

250
Public Accounts Committee.
251
Michael Gove to Education Select Committee 2010, quoted in Purcell. P128

94
This chapter concludes the introductory consideration of the ‘cornerstone’ areas of
education, crime, and youth policy which set the context for disadvantaged young people
during the last two decades. The rest of this report now looks at how policy and outcomes
developed over the same period in relation to five more specific problems of adolescence -
school exclusion, school absence, teenage pregnancy, adolescent alcohol use, and
adolescent drug use.

95
Chapter 4 School Exclusion
Introduction
The permanent exclusion of a pupil from secondary school is a significant step, often
associated with negative outcomes for the young person, their family, and society. The
children most likely to be excluded are overwhelmingly the most vulnerable and frequently
need specialist support to address serious issues. When Labour came to office in 1997
school exclusions were high and on an upward trend. The level of exclusions then fell
significantly, by nearly two thirds, before rising again after 2013.
This chapter looks at the circumstances behind these striking trends. It first sets out some
key background facts on permanent exclusion – its scale, who is most affected, and the cost.
It then charts the development of policy under Labour, Coalition and Conservatives, sets out
the trends in exclusion rates, and discusses possible reasons for the trends. The chapter
also summarises (in Box SE1) what is known on the parallel phenomenon of off-rolling.

What is a permanent exclusion?

Permanent exclusion is a serious sanction intended to be used as a last resort where other
strategies have failed. Statutory guidance sets out the framework within which schools
must act. Shorter suspensions (known as fixed-term exclusions) are available as a sanction
for less serious cases.
The most common reason given for permanent exclusions in secondary schools in 2018/19
were:

• ‘persistent disruptive behaviour’ (36 per cent of exclusions)

• ‘other’ (19 per cent)
• physical assault of a pupil (13 per cent)
• drugs and alcohol (10 per cent).
When a child is permanently excluded, the local authority is responsible for arranging
alternative education from the sixth day of exclusion.252

How many permanent exclusions are there?

The key data on permanent exclusion for England are set out below, for the beginning and
end of the period under study, and for 2012/13 when rates were lowest. The full trends are
shown in Figure SE2 later in this section.
Table SE1: Number and rates of permanent exclusion, state secondary schools, selected
years, England253
1997/98 2012/13 2018/19
Permanent exclusions (number) 10,190 3,903 6,753
Permanent exclusions rate (per cent) 0.33 0.12 0.20

252
Department for Education, Children Missing Education: Statutory Guidance for Local Authorities (DfE, 2016).
253
DfE, Permanent and fixed exclusions in England, 2018/19 and previous years’ releases.

96
Groups over-represented in permanent exclusions
The prevalence of permanent exclusion is not evenly distributed across the secondary
school population, as Figure SE1 shows. Nearly three times as many boys as girls are
permanently excluded, and the exclusion rate peaks in year 10.
Children eligible for free school meals, those with special educational needs, and ‘children in
need’ are all heavily over-represented in permanent exclusions, and young people who face
all three of these issues represent 11 per cent of all permanent exclusions.254 255 An Ofsted
review in 2017 found that children and young people identified as needing special needs
support but who did not have an Education and Health Care Plan were particularly
prominent in exclusions, and were especially vulnerable to exclusion in mainstream
secondary schools.256
Figure SE1: Groups over-represented in permanent exclusion, state-funded secondary
schools in England, 2018/19

0.8
0.7 0.65
0.6
0.6 0.51
0.5
Per cent

0.38
0.4 0.33 0.31
0.3
0.3 0.21
0.2
0.2 0.14 0.15 0.15
0.11 0.11
0.1 0.02
0

Source: DfE, Permanent and fixed period exclusions in England

Multiple vulnerabilities
Many forms of adolescent disadvantage are associated with elevated rates of permanent
exclusion. These include low attainment and school absence, 257 prior mental health
problems,258 and a wider range of issues such as bereavement, traumatic life events, family

254
Edward Timpson, Timpson Review of School Exclusions (DfE, 2019).
255
‘Children in need’ is a term used to describe children in need of help and protection, who are assessed and
supported through children’s social care. Over the course of a year, it is estimated that around 6 percent of all
children in England will be in need at some point.
256
Ofsted, Local Area SEND Inspections: One Year On. (Ofsted, 2017).
257
Steve Strand and John Fletcher, A Quantitative Analysis of Exclusions from English Secondary Schools
(University of Oxford, 2014).
258
T. Ford and others, ‘The Relationship between Exclusion from School and Mental Health: A Secondary
Analysis of the British Child and Adolescent Mental Health Surveys 2004 and 2007’, Psychological Medicine,
48.4 (2018), 629–41.

97
breakdown, poverty, living in disadvantaged communities, and discrimination, the effects of
which may be multiplicative. 259
A large study linking education and health data sets for more than 400 000 pupils in Wales
draws attention to over-representation of young people with mental health problems
amongst school excludees (and absentees). It found that children and young people
diagnosed with a neurodevelopmental disorder or mental disorder, or with a record of self-
harm, are much more likely to be excluded from school than their peers, even after
adjusting for age, sex, and deprivation. Individuals with more than one condition were even
more likely to miss school through absence or exclusion with each extra disorder
exacerbating the likelihood. But having SEN status reduced the likelihood of being absent or
excluded, highlighting the positive effect of recognition, diagnosis and intervention.260
Ethnicity
There are also significant disparities in exclusion rates by ethnicity, (shown in Figure SE5
later in this chapter) with the highest exclusion rates experienced by pupils from Gypsy
Roma, Irish Traveller, Black Caribbean, and Mixed White and Black Caribbean heritage.
Research on this topic has found a concerning list of contributory issues including lack of
training but also racism, stereotyping, low expectations, differential treatment, and pupils
feeling under-valued and disrespected.261 A study conducted in 2014 with parents of
excluded pupils found that the majority of parents felt that race, class, gender or SEN played
a role in their child’s exclusion.262

The costs of school exclusion

Young people who have been permanently excluded from school are at high risk of a range
of negative outcomes and the costs of exclusion to the taxpayer are significant – estimated
at £2.1 billion for every cohort in terms of education, health, benefits, and criminal justice
costs.263
Most young people who have been excluded go on to be educated in alternative provision:
this is a sector which is under high pressure, some of which is unregistered and therefore
uninspected, and where persistent absence levels are high.264 The figures on educational
attainment in alternative provision are striking: only 1 per cent of students who complete
their GCSEs in alternative provision achieve 5 good GCSEs including English and Maths,

259
Ted Cole, Mental Health Difficulties and Children at Risk of Exclusion from Schools in England (University of
Oxford, 2015); Kiran Gill, Harry Quilter-Pinner, and Danny Swift, Making the Difference Breaking the Link
Between School Exclusion and Social Exclusion (London: Institute for Public Policy Research, 2017).
260
Ann John and others, ‘Association of School Absence and Exclusion with Recorded Neurodevelopmental
Disorders, Mental Disorders, or Self-Harm: A Nationwide, Retrospective, Electronic Cohort Study of Children
and Young People in Wales, UK’, The Lancet Psychiatry, 9.1 (2022), 23–34.
261
Berni Graham and others, School Exclusion: A Literature Review on the Continued Disproportionate
Exclusion of Certain Children (DfE, 2019); Feyisa Demie, ‘The Experience of Black Caribbean Pupils in School
Exclusion in England’, Educational Review, 73.1 (2019), 55–70.
262
C. Kulz, Mapping the Exclusion Process: Inequality, Justice and the Business of Education (Communities
Empowerment Network, 2015).
263
Gill, Quilter-Pinner, and Swift.
264
House of Commons Education Committee, Forgotten Children: Alternative Provision and the Scandal of Ever
Increasing Exclusions (House of Commons, 2018).

98
compared with a national average of 64.5 per cent.265 The average cost of a full-time
placement in alternative provision for one academic year was £18,000 in 2017/18. 266
Research has found that, compared with similar students who had not been excluded,
permanent exclusion in secondary school increases the risk of becoming NEET at the age of
19/20, and the risk of unemployment and lower wages at the age of 25/26. 267 Experience of
exclusion from school is associated with an increase in mental health difficulties.268 Young
people who have been excluded are commonly targeted by County Lines drug gangs, as
described in Chapter 8. Young people who have been permanently excluded are heavily
over-represented in the criminal justice system. Recent research by the Department for
Education and Ministry of Justice found that around a fifth of young people who had been
permanently excluded from school had also been sentenced or cautioned for a serious
violence offence: for three quarters of these young people the violence offence was at least
a year after the exclusion. 269
It is not possible to say that these adverse outcomes are all attributable to the impact of
exclusion. There will often be risk factors in young people’s lives that may contribute both
to the exclusion and the emergence of later problems. However, there is little disagreement
that the experience of exclusion is one in which young people’s lives can go badly off track.
A Home Office-commissioned study in 2001 found that permanent exclusion tended to
trigger a complex chain of developments events that made young people more vulnerable,
involving loss of time structures, recasting of identity, changed family relationships, and
adverse changes in peer group. It found little evidence that exclusion led to desistance, with
only 5 per cent of the study who had recorded offences before the exclusion having no
further offences afterwards.270
The costs of exclusions are part of a bigger picture of cost connected with pupil behaviour.
Developing and maintaining good behaviour management in a school takes resource and
sustained attention. Conversely, disruptive behaviour in school leads to lost learning for
pupils and added stress for teachers. Ofsted research has found that poor behaviour is a
considerable source of low occupational wellbeing amongst teachers.271

What works in reducing exclusions?

Evidence from international studies shows that there are interventions that show results in
reducing the level of disciplinary exclusions. A systematic review of randomised controlled

265
Dannie Mason and others, Pinball Kids (Royal Society of Arts, 2020).
266
Isos partnership, Alternative Provision Market Analysis (DfE, 2018).
267
Joan E. Madia and others, ‘Long-Term Labour Market and Economic Consequences of School Exclusions in
England: Evidence from Two Counterfactual Approaches’, British Journal of Educational Psychology, 92.3
(2022), 801–16.
268
T. Ford and others.
269
Department for Education and Ministry of Justice, Education, Children’s Social Care and Offending;
Descriptive Statistics (HMG, 2022). This research relates to young people who had a Key Stage 4 academic year
of 2012/13, 2013/14 or 2014/15.
270
David Berridge and others, The Independent Effects of Permanent Exclusion from School on the Offending
Careers of Young People (Home Office, 2001).
271
Ofsted, Teacher Well-Being at Work in Schools and Further Education Providers.

99
trials from the US and UK brought this evidence together, looking at a range of interventions
most of which attempted either to improve the academic, social and behavioural skills and
mental health of pupils, or to develop the skills of teachers and make systemic changes
across the whole school to promote good behaviour, learning and safety. The research
found four broad approaches with promising and significant results: enhancement of
academic skills, counselling, mentoring/monitoring, and skills training for teachers.272
More evaluations of UK programmes are referenced in later sections of this chapter, and
there are also several accessible summaries of good practice in the UK context.273

Policies and spending programmes

Labour: 1997–2010
When Labour came into Government in 1997 school exclusions were high and had been
rising. A 1998 report by the Social Exclusion Unit found that too many children were being
excluded ‘for relatively minor reasons, or because they needed help they didn't get’. 274 275
A range of new policies were implemented in the wake of the report, covering both the legal
framework for exclusion, and prevention - the development of approaches to avert and
address behaviour problems. There were more resources for work with children at risk of
exclusion, measures to tackle school disaffection with more imaginative approaches to the
curriculum, and new statutory guidance on exclusions which emphasised the need for
prevention and made it clear that exclusion should be a last resort. Targets were set for
local education authorities to reduce exclusions by a third within three years. Investment in
a new area-based programme, ‘Excellence in Cities’, was a key part of the government’s
effort on exclusion. This initiative, whose budget was £139 million in 2000/01 rising to some
£300 million by 2002/03, was intended to tackle low attainment and poor attendance as
well as behaviour and exclusion.276 For that reason, it also features in Chapter 1 and
Chapter 5 of this report.
Permanent exclusions fell sharply between 1997 and 1999 but the targets were
controversial. By 1999, the Secondary Headteachers Association was complaining that local
authorities were implementing the policy with excessive zeal, and that justified exclusions
were being overturned on technicalities.

272
Sara Valdebenito and others, ‘What Can We Do to Reduce Disciplinary School Exclusion? A Systematic
Review and Meta-Analysis’, Journal of Experimental Criminology, 15.3 (2019), 253–87.
273
Susan Hallam and Lynne Rogers, Improving Behaviour and Attendance at School (Open University Press,
2008); Carl Parsons, Strategic Alternatives to Exclusion from School (Trentham, 2011); Louise Gazeley and
others, Reducing Inequalities in School Exclusion: Learning from Good Practice (University of Sussex, 2013); Ted
Cole and others, ‘Factors Associated with High and Low Levels of School Exclusions: Comparing the English and
Wider UK Experience’, Emotional and Behavioural Difficulties, 24.4 (2019), 374–90; Alice Tawell and Gillean
McCluskey, ‘Utilising Bacchi’s “What’s the Problem Represented to Be” (WPR) Approach to Analyse National
School Exclusion Policy in England and Scotland: A Worked Example’, International Journal of Research and
Method in Education, 45.2 (2022), 137–49.
274
Social Exclusion Unit, Truancy and School Exclusion (Stationery Office, 1998).
275
As Director of the Social Exclusion Unit at the time, I was involved in the development of this report.
276
Machin, McNally, and Meghir.

100
In 2001 the Education Secretary, David Blunkett, announced that he would not be setting
further targets. However, the government continued to work at reducing exclusion through
prevention and support, not least because the involvement of excluded pupils in street
crime kept the issue on the agenda. The Department for Education and Skills set up the
Behaviour Improvement Programme in 2002, which tackled exclusion, behaviour and
absence, and provided £331 million for it between July 2002 and 2005-06.277 Building on
the lessons of Excellence in Cities, it funded a menu of initiatives mainly within schools,
including learning support units (in-school units which provided extra support out of the
classroom for children who needed this) and other initiatives such as key workers and
behaviour support teams, adapting the curriculum, increasing support for parents, and new
partnership programmes with local police. 278 These policies were aimed at tackling some of
the underlying issues that cause any or all of exclusion, absence, disengagement, and low
attainment. Chapter 5 sets out more detail of what the Behaviour Improvement
Programme funded: it is set out there because work on absence and behaviour were so
closely linked.
Excellence in Cities and the Behaviour Improvement Programme showed positive results in
their evaluations and the government sought to encourage all schools to adopt the good
practice that was emerging.279 The Key Stage 3 strategy, designed to raise the achievement
of 11- to 14-year-olds, began to focus on improving behaviour and attendance from
2003/04, providing training and practical advice, and encouraging behaviour management
to be seen as an integral part of teaching.280 It led to the recruitment of 236 behaviour and
attendance consultants in local authorities by December 2004. 281 A national practitioners
review group led by headteacher Sir Alan Steer set a strong professional lead, declaring that
the quality of learning, teaching, and behaviour in schools were inseparable, and the
responsibility of all staff, and that while there was no single solution to the problem of poor
behaviour, all schools had the potential to raise standards if they were consistent in
implementing good practice.282 A national drive to improve unsatisfactory behaviour
continued in the DfES/DCSF for the rest of the decade, with direct funding for regional
support teams and the local authority consultants mentioned above. Describing this work to
a Select Committee in 2010, the lead official reported a reduction in the number of schools
which had inadequate behaviour from 72 in 2005 to 18 in 2009, listing a range of initiatives
that had been implemented, including work with school senior teams to establish clear and
simple behaviour codes, in-school training, sharing of good practice between departments

277
National Audit Office, Improving School Attendance in England (NAO, 2005).
278
Susan Hallam, Frances Castle, and Lynne Rogers, Research and Evaluation of the Behaviour Improvement
Programme. (DfES, 2005).
279
Kendall and others; Hallam, Castle, and Rogers.
280
Ofsted, The Key Stage 3 Strategy : Evaluation of the Second Year (Ofsted, 2005).
281
National Audit Office, Improving School Attendance in England.
282
Sir Alan Steer, Learning Behaviour: The Report of the Practitioners’ Group on School Behaviour and
Discipline (DfES, 2005).

101
in schools, and the development of a National Programme for Specialist Leaders of
Behaviour and Attendance which had 7,000 participants as of 2010. 283
Developments in other policy areas were relevant to the effort to reduce exclusion. As
Chapter 1 sets out, several features of special educational needs policy incentivised schools
to identify pupils with special needs and to support them to remain in mainstream
schooling. The proportion of pupils qualifying for ‘school action’ and ‘school action plus’
support rose from 13 per cent of secondary pupils to a peak of 19.7 per cent in 2010.
Increases in school spending per pupil, the increased targeting towards schools in deprived
areas, and initiatives to make a wider range of vocational subjects available at Key Stage 4
all contributed to schools' ability to support and motivate pupils at risk of exclusion. Other
relevant initiatives include increases in the financial support for families with children, and
the creation of Connexions, as a new youth support service to support 13- to 19-year-olds
through the transition to work or post 16 learning (covered in Chapter 1), and policies and
spending programmes to tackle drug and alcohol use by young people (covered in Chapters
7 and 8).
Coalition: 2010–2015

Policy on school exclusion took a very different direction under the Coalition government.
The Coalition took the view that ‘heads and teachers want to improve behaviour and teach
in a calm, orderly environment, but are too often constrained by regulations which inhibit
them from maintaining order’.284 In future, therefore, exclusion appeals would go to review
panels which would not be able to require a school to reinstate a pupil.285 If a governing
body was directed by a review panel to reconsider its decision and did not wish to, the
school could still implement its original decision, subject to an additional payment to the
local authority of £4,000. 286 The new policy came into effect in September 2012.
At the same time the emphasis on prevention declined. New exclusions guidance
superseded the previous (2008) version, replacing extensive material on prevention with a
few paragraphs about the need to consider support and vulnerabilities, but no practical
detail.287 The Coalition’s 2011 Education Act also repealed the duty on schools to
participate in Behaviour and Attendance Partnerships: Sir Alan Steer who had led the
previous government’s practitioner group on behaviour and discipline described this as ‘a
major error’ since ‘schools operating without a sense of responsibility to other schools, or

283
House of Commons Education Committee, Behaviour and Discipline in Schools : Oral and Written
Evidence. (HoC, 2011).
284
House of Commons Education Committee, Behaviour and Discipline in Schools : Oral and Written Evidence.
285
Department for Education, The Importance of Teaching.
286
This payment would be additional to the long-standing requirement for transfer of a capitation amount
when a pupil moves to another school.
287
Department for Education, Exclusion from Maintained Schools, Academies and Pupil Referral Units in
England (DfE, 2012).

102
the needs of the wider community, can become destructive forces, denying parental rights
and damaging the education of children’.288
There were also significant developments in special educational needs policy under the
Coalition. From 2013/14 the DfE introduced a new rule that schools had to find the first
£6,000 of support for each pupil with special educational needs. A 2019 report from the
National Audit Office noted that this threshold was widely perceived as making mainstream
schools reluctant to admit or keep pupils with costly special educational needs. Their report
recommended that the government should review the incentives in the funding and
accountability systems and make changes that support mainstream schools to be more
inclusive.289
The wider funding background for schools was challenging. As Chapter 1 sets out, during the
Coalition, per-pupil funding began to fall in real terms and pupil-adult ratios began to rise.
The external support available to help schools to manage pupils with problems declined
over this period, due to local authority budget reductions, the move to traded services, and
the impact of rapid academisation. There were reductions in local authority services such as
educational welfare, educational psychology, children’s centres, and mentoring for at-risk
young people.290 A 2014 DfE-funded study of change in local education systems during this
period found that support for vulnerable pupils was the area where the new landscape was
working least well, with 44 per cent of school leaders saying there was not the provision in
their school and across the local area to ensure vulnerable children received a high quality
education.291 For services outside education, the picture was also one of retrenchment in
Connexions, policing, youth services, and drug and alcohol services (see Chapters 1, 2, 3, 7
and 8).
With the peak age of exclusion being year 10, changes to GCSEs were potentially also of
great significance to young people at risk of exclusion. As described in Chapter 1, from
summer 2014, a large number of vocational qualifications were removed from eligibility
towards performance measures, and a cap was placed on the total number of vocational
qualifications that could count. The last of these measures had a particular impact on
disadvantaged pupils for whom these qualifications tended to be more significant.292
Conservatives: 2015 onwards
If the Coalition’s aim was to improve behaviour in schools, the results were not encouraging.
In 2014 the Chief Inspector’s Annual Report reported that there had been a substantial fall
in the number of secondary schools in which behaviour was judged good or outstanding
compared with those inspected in the previous year – down from 28 per cent to 21 per

288
Sir Alan Steer, ‘Education Bill: Memorandum E 16’, 2011
<https://publications.parliament.uk/pa/cm201011/cmselect/cmpublic/education/memo/e16.htm> [accessed
13 February 2023].
289
National Audit Office, Support for Pupils with Special Educational Needs and Disabilities in England.
290
Cole and others.
291
Sandals and Bryant.
292
Simon Burgess and Dave Thomson, ‘The Impact of the Wolf Reforms on Education Outcomes for Lower-
Attaining Pupils’, British Educational Research Journal, 45.3 (2019), 592–621.

103
cent.293 After the 2015 election, the incoming Conservative government asked its
behaviour advisor, Tom Bennett, to draw up plans to help teachers deal with low-level
disruption in classrooms: his report recommended that all initial teacher training should
include ‘a broad and practical introduction to the understanding and craft of behaviour
management’ and that this should be mandatory.294 The government welcomed the report
but said it did not wish to make the requirement mandatory. 295 But in 2019, it published a
summary of Bennett’s advice as a toolkit, supporting a revised framework for initial teacher
training.296
Otherwise, the new government in 2015 continued with very similar education and youth
policies. Political attention continued to focus on academisation and the creation of free
schools. Progress 8 became the key indicator of secondary school performance: this gave
double weight to English and Maths, but very little weight to some of the subjects which
were most valued by pupils at risk of exclusion. 297 As Chapter 1 describes, funding
reductions continued, and teacher numbers continued to fall. Several critical reports
focused on inadequacies and underfunding of the new special educational needs system. 298
Outside education, as described in Chapters 3, 7 and 8 of this report, spending reductions
continued across a range of services affecting young people, including the youth service,
drugs, and alcohol treatment. Pressures on Child and Adolescent Mental Health services
also rose as referrals increased by 26 per cent between 2013 and 2018. In 2018, between a
fifth and a third of children referred to specialist mental health services were not accepted
for treatment. 299 300
Rising permanent exclusion rates began to attract political attention, given impetus by
growing public concern about knife crime, and the phenomenon of ‘County Lines’ drug
gangs recruiting young people who were not in mainstream schools.301 A 2018 report from
the House of Commons Education Select Committee expressed severe concern about the
rise in exclusion rates and the quality of much alternative provision.302
The government commissioned Conservative MP and former Children’s Minister Edward
Timpson to review exclusion policy. His report, published in 2019, found that there was

293
Her Majesty’s Chief Inspector of Education Children’s Services and Skills, Annual Report 2013/14 (Ofsted,
2014).
294
Tom Bennett, Developing Behaviour Management Content for Initial Teacher Training (DfE, 2016).
295
Rt Hon Nicky Morgan MP, Government Response to Carter, Bennett and Holder Reviews of Initial Teacher
Training (DfE, 2016).
296
https://www.gov.uk/government/publications/initial-teacher-training-itt-core-content-framework/the-
trainee-teacher-behavioural-toolkit-a-summary
297
Cole and others.
298
National Audit Office, Support for Pupils with Special Educational Needs and Disabilities in England; House
of Commons Education Committee, Special Educational Needs and Disabilities.
299
Whitney Crenna-Jennings and Jo Hutchinson, Access to Children and Young People’s Mental Health Services
(Education Policy institute, 2018).
300
Mental health policy is discussed in more detail in Chapter 5.
301
National Crime Agency, County Lines Drug Supply, Vulnerability and Harm (NCA, 2019).
302
House of Commons Education Committee, Forgotten Children: Alternative Provision and the Scandal of Ever
Increasing Exclusions, (House of Commons, 2018).

104
more to do to ensure that every exclusion was ‘lawful, reasonable and fair’, and highlighted
the perverse incentive that schools could improve their performance and finances through
exclusion, but did not bear the cost of the provision these children then attend, or
responsibility for their outcomes. 303 The report’s recommendations covered a wide range
of issues: setting high expectations for every child; giving schools the capacity to deliver;
incentives and monitoring; and ensuring no child misses out on education.
The government’s response to the Timpson review fell short of full commitment.304 Some
recommendations have been progressed, including some additional investment in
alternative provision, and expanded guidance on exclusions and behaviour management. 305
But many of the recommendations have not been fully implemented.306 For example:

• Timpson recommended that DfE should extend its funding for equality and diversity
hubs, a well-regarded initiative to increase diversity of senior leadership teams in
England’s schools, but it was cancelled in the 2020 spending review. 307

• The recommendation that schools be made responsible for the children they
exclude, and that funding should support schools in retaining children at risk of
exclusion was incorporated in wider work on alternative provision and special
educational needs. However, four years later, the government’s final Improvement
Plan merely promised, without deadline, to ‘develop new approaches [..] that
prioritise preventative work and reintegration of pupils back into mainstream
schools. 308 309

• The recommendation that pupil moves out of schools should be systematically

tracked was potentially going to be dealt with by measures in the Schools Bill 2022.
However, this Bill was then dropped. There is no clarity about when legislation will
be brought forward again. 310

303
Timpson.
304
Department for Education, The Timpson Review of School Exclusion: Government Response (DfE, 2019).
305
Department for Education, Revised Behaviour in Schools Guidance and Suspension and Permanent Exclusion
Guidance: Government Response to Consultation (DfE, 2022); Department for Education, Behaviour in Schools:
Advice for Headteachers and School Staff (DfE, 2022).
306
‘Timpson Review of School Exclusion: Westminster Hall Debate, Thursday 16 September 2021’ (Hansard,
2021); Aaron Kulakiewicz and Rob Long, The Implementation of the Recommendations of the Timpson Review
of School Exclusion (House of Commons Library, 2021); IntegratED, ‘Timpson Tracker’, Timpson Tracker
<https://www.integrated.org.uk/what-needs-to-change/timpson-tracker/> [accessed 24 March 2023].
307
Department for Education, Leadership Equality and Diversity Fund 2018/19: Programme Analysis (DfE,
2020); James Carr, ‘“Disgraceful”: DfE Axes Diversity Hubs Funding’, Schools Week, November 2020.
308
Department for Education, Provision for Children and Young People with Special Educational Needs and
Disabilities, and for Those Who Need Alternative Provision : How the Financial Arrangements Work. Call for
Evidence (DfE, 2019).
309
HM Government, Right Support, Right Place, Right Time (HMG, 2023).
310
Department for Education, Children Not in School: Schools Bill Factsheet (DfE, 2022); House of Commons
Education Committee, ‘Transcript of Oral Evidence: 7 December’ (HoC, 2022).

105
Outcomes
Permanent exclusion rate
We now turn to the data on trends in school exclusion. As Figure SE2 below illustrates, the
rate of permanent exclusions in secondary schools fell significantly, in two phases. It fell by a
third in the two years after 1997/98, then edged up again. From 2003/04 to 2012/13 it
halved. Since then, it has risen again.
Figure SE2: Trends in permanent exclusion rate in state-funded secondary schools in
England

0.35

0.3 0.33%

0.25 6,753
pupils
0.2
Per cent

0.15 0.20%
0.1
0.12%
0.05

Source: DfE, Permanent and Fixed Period Exclusions in England

Age distribution
Data on the age distribution of permanent exclusions shows that secondary schools, and
within that, the GCSE years have been strongly affected by rises since 2012/13. (Table SE2)
Table SE2: Percentage change in permanent exclusion rate by national curriculum stage,
state funded schools in England, change between 2012/13 and 2018/19
% change
Primary schools +46
year 7 +66
year 8 +61
year 9 +62
year 10 +72
year 11 +80

106
Inequalities
Data on inequalities in school exclusion is in most cases only available from 2006/07
onwards. It shows that inequalities were reduced then rose again.
Free school meals and special educational needs
The free school meals gap in exclusions narrowed significantly until 2012/13 but widened
again thereafter. The same happened for exclusions of pupils with special needs. Both these
trends are shown in Figure SE3.
For young people with a special needs statement (later known as an Education and Health
Care Plan) the increase after 2012/13 is smaller than for young people with lower levels of
special needs. Throughout this period, guidance has emphasised that head teachers should
‘as far as possible, avoid excluding permanently any pupil with a statement of SEN or a
looked after child’.311
Figure SE3: Trends in permanent exclusion rates by free school meal and special
educational needs status, state-funded secondary schools in England
Free school meal status Special educational needs status

1 1

0.9 0.9

0.8 0.8

0.7 0.7

0.6 0.6
Per cent

0.5 0.5

0.4 0.4

0.3 0.3

0.2 0.2

0.1 0.1

0 0
1997/98
1998/99
1999/00

2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
2017/18
2018/19
1997/98

2012/13
1998/99
1999/00
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12

2013/14
2014/15
2015/16
2016/17
2017/18
2018/19

2000/01

No SEN
Eligible for FSM
Not eligible for FSM Statement or EHC
SEN without statement
Source: DfE, Permanent and fixed period exclusions in England

Department for Education, Exclusion from Maintained Schools, Academies and Pupil Referral Units in
311

England. (DFE, 2012). Similar wording was in the 2017 guidance.

107
Regions
Regional rates of permanent exclusion are shown in Figure SE4. The gap between regions
changed little in the first decade but was at its lowest between 2011 and 2013. Increases
since 2013/14 have been particularly pronounced in the North East and the North West. As
of 2019, permanent exclusion rates in the North East were the highest in two decades and
double the national average.

Figure SE4: Trends in permanent exclusions in state-funded secondary schools in England,

by region

0.45

0.4

0.35

0.3

0.25
Per cent

0.2

0.15

0.1

0.05

North East North West West Midlands South West

Yorks Humb E Mids E Eng I Lond
O Lond SE

Source: DfE Permanent and fixed-period exclusions in England

Ethnicity
The largest reductions of any ethnicity between 2006/07 and 2012/13 were recorded by
Black Caribbean, Mixed White and Black Caribbean, and Gypsy Roma ethnicities - three of
the four most over-represented groups. These three groups then lost ground again in
subsequent rises. (Figure SE5). Permanent exclusions of white pupils, although low
compared with other ethnicities, returned to their 2006/07 level by 2018/19. Exclusion rates
for Travellers of Irish heritage ended the period higher than they began (with large
fluctuations year to year).

108
Figure SE5: Permanent exclusion rates in state-funded secondary schools in England, by
ethnicity, selected years
1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

2006/07 2012/13 2018/19

Source: DfE Permanent and fixed-period exclusions in England

Gender
Figure SE6 shows that exclusions fell for both boys and girls until 2012/13 and the gender
gap narrowed slightly. Thereafter both boys and girls’ exclusions rose again, and the gap
widened again.
Figure SE6: Trends in permanent exclusion rates in state-funded secondary schools in
England, by gender
0.4
0.35
0.3
0.25
Per cent

0.2
0.15
0.1
0.05
0

Girls Boys
Source: DfE, Permanent and Fixed Period Exclusions in England

109
Reasons for exclusion
Data on the reasons for permanent exclusions exists back as far as 2006/07. The main
components of the fall and rise are set out in below. In both the fall of exclusion and its
subsequent rise, changes in exclusion for ‘persistent disruption’ and for ‘other’ were the
main components. Reductions in exclusion for assault were the third largest component of
the reduction between 2006/07 and 2012/13, and drug and alcohol related exclusions were
the third largest contributor to the rise after 2012/13.
Table SE3: Breakdown of change in permanent exclusions over time by reason recorded
Level in Change 06/07 Change 12/13 Level in
2006/07 to 12/13 to 18/19 2018/19
Persistent disruption 2,360 - 1,143 + 1,196 2,413

Other 1,257 - 606 + 635 1,286

Physical assault against pupil 1,160 - 517 + 266 909

Physical assault against teacher 675 - 415 + 168 428

Drugs or alcohol 396 - 38 + 329 687

Trends in behaviour
A great deal of information is available about permanent exclusion, but we have far less of a
grasp on underlying behaviour trends – the issue to which exclusion is supposed to be
responding. According to DfE, in 2012, there is no accepted measure of behaviour levels in
school and how they have changed over time.312 Ofsted inspection data on behaviour
judgements has from time to time been drawn together in the Chief Inspector’s annual
report. This suggests that there was an improvement in behaviour during the period when
exclusions were falling. The proportion of secondary schools rated inadequate for behaviour
fell from just under one in ten in 2003/04 to 3 per cent in 2005/06 and 2 per cent in
2010/11.313
In 2012, the Chief Inspector's annual report observed that ‘Schools are mainly calm, well-
managed places, providing a good environment in which poor behaviour is much less of an
issue’. But within two years the Annual Report described ‘a substantial fall in the number of
secondary schools in which behaviour was judged good or outstanding compared with those

312
Department for Education, Pupil Behaviour in Schools in England (DfE, 2012).
313
Her Majesty’s Chief Inspector of Schools, Annual Report 2003/04 (Ofsted, 2005); Her Majesty’s Chief
Inspector of Schools, Annual Report 2005/06 (Ofsted, 2006); Her Majesty’s Chief Inspector of Education
Children’s Services and Skills, Annual Report 2010/11 (Ofsted, 2011).

110
inspected in the previous year’ (the fall was from 28 per cent of inspected schools to 21 per
cent). 314
The appeals process and its outcome

DfE also publish data on the caseload and outcome of the independent review process since
2012/13. The percentage where the exclusion is upheld has fallen over time as can be seen
in Figure SE7. Only a minority of cases where the panel recommends or directs
reconsideration lead to pupil reinstatement – 74 cases out of 263 in 2018/19.
Figure SE7: Trends in appeals against permanent exclusion and their outcome, England,
all schools

700
661
640

600
560

500
461
Number of appeals

395 393
400
363
351 347
320
296
300
262 267 263
236
214
200 172

125
106
95
100 84 79 74
45
20 21 21 22

0
2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 2018/19

Reviews lodged Exclusions upheld Reconsideration recommended or directed Reinstated

Source: DfE, Permanent exclusions and suspensions in England

314
Her Majesty’s Chief Inspector of Education Children’s Services and Skills, Annual Report 2011/12: Schools
(Ofsted, 2012); Her Majesty’s Chief Inspector of Education Children’s Services and Skills, Annual Report
2013/14. (Subsequent reports have not commented further on this metric.)

111
Comparisons within the United Kingdom
Wales
The permanent exclusion rate in secondary schools in Wales has been lower than England
throughout the period discussed. It sustained its low level for longer than England but then
rose after 2015/16. (Figure SE8)
Figure SE8: Trends in permanent exclusion rate, England and Wales compared
0.35

0.30

0.25
0.20
0.20
Per cent

0.15 0.13

0.10

0.05

0.00

England Wales

Sources: DfE, Permanent & fixed period exclusions in England; Welsh Government. Permanent & fixed-term
exclusions from schools

Scotland
Scottish policy and practice on permanent exclusions is very different from England.
Permanent exclusions were always low in Scotland with a total of just 292 across primary
and secondary combined in 2002/03. This fell to 87 by 2008/09, and after further falls has
been in single figures since 2014/15. To quote from the commentary in the 2018/19
Summary Statistics for Schools in Scotland:
Over 99 per cent of all exclusions were for a fixed period of time […] and pupils are
expected to return to their original school when the exclusion period is completed. In
a small number of cases, three in 2018/19, an excluded pupil is ‘removed from the
register’. When this occurs, a pupil does not return to their original school and will be
educated at another school or in some other educational provision. 315
Scotland’s approach is discussed in more detail in a 2019 article by Gillean McCluskey and
others: this highlights as key success factors ‘the effectiveness of a strategic emphasis on
prevention, of national/local co-design and planning, and of maintaining focus on the
complexity of some young people’s lives and the often deep levels of disadvantage’. 316

Scottish Government, Summary Statistics for Schools in Scotland (Scottish Government, 2019).
315

Gillean McCluskey and others, ‘Exclusion from School in Scotland and across the UK: Contrasts and
316

Questions’, British Educational Research Journal, 45.6 (2019), 1140–59.

112
Plausible explanations of the trends in England

The two-decade picture

The outcome trends above show striking changes: first, a sustained fall in permanent
exclusions experienced across regions and groups, then a marked rise after 2012. During the
period of reduction, the improvement was most pronounced for certain over-represented
ethnic minorities, those with special educational needs, and those on free school meals.
After 2012, these groups saw larger increases in permanent exclusion.
The literature on this subject tends to explain both the improvement in exclusion rates, and
their subsequent deterioration, by reference to a set of mutually reinforcing systemic
changes. The system in question has two linked elements:

• exclusion itself (the legal and accountability processes that govern exclusion and the
extent to which they encourage or discourage the decision to exclude)

• prevention (the presence or absence of support systems to support good behaviour

and manage issues in ways that avoid a permanent exclusion).
A 2019 study that covered the entire two decade period included both these elements in its
analysis, describing the late 1990s to 2010 as a period of effective practice, combining
national direction and investment, local leadership and services, and whole-school
approaches that addressed the difficulties of at-risk children, tailored the curriculum, and
deployed extra interventions through teaching assistants, SEN teachers and other
professionals. This was contrasted with the Coalition and Conservative period of
government, which brought adverse changes to review panels, reduced funding for schools
and local authorities, reduced local collaboration, and introduced Progress 8 accountability
measures and the requirement for schools to fund the first £6,000 of a child’s special
educational needs support. The authors summarised this by saying that since 2010, inclusive
practice had been ‘increasingly forgotten or ignored or […] financially impracticable’.317
The fall in exclusions
Analysis of the fall in exclusion rates which began under Labour and continued until 2012/13
tends to converge on the impact of new approaches to prevention. Contemporary
evaluations of programmes from the early 2000s, such as Excellence in Cities and the
Behaviour Improvement Programme, found a wide range of evidence to suggest that they
were effective in improving behaviour, and thereby allowed schools to reduce exclusions.

• Ofsted reviewed the operation of The Excellence in Cities programme: it found that
in secondary schools the use of both learning mentors and learning support units
were making a significant impact on behaviour, and that three quarters of secondary
schools involved in the programme had reduced their permanent exclusions by more
than the national average.318 Qualitative research found that many teachers thought

317
Cole and others.
318
Ofsted, Excellence in Cities and Education Action Zones: Management and Impact (Ofsted, 2003).

113
 the programme had brought about positive changes in pupils’ attitudes, motivation,
behaviour and attendance with the introduction of learning mentors and learning
support units, again being thought to have contributed.319

• Evaluation of The Behaviour Improvement Programme (BIP) found that secondary

schools involved in BIP Phase 2 showed a statistically significant reduction in
permanent exclusions, with qualitative evidence that the programme had fostered
more application of good practice such as a positive school ethos, raising pupils’ self-
esteem, clearer behaviour policies implemented more consistently, and increased
access to multi-agency support for children whose behaviour was being affected by
issues outside the school. 320

• Learning Support Units in schools, widely rolled out as part of the approach to
behaviour, were the subject of an Ofsted special inspection in 2005. All twelve units
inspected were found to be successful in improving the behaviour and attendance of
their pupils.321 By 2005 there were 1,500 learning support units operating in English
schools322
It is worth noting that these innovations were strongly championed by leaders within the
profession, as described earlier in the chapter. This support probably rested in part on the
switch to an investment approach, replacing the earlier dependence on local targets: local
targets for reducing exclusions were widely seen as too crude, and were abandoned by
David Blunkett in 2001.
Wider developments in education policy
It also appears that wider changes in education will have played a reinforcing role, in a way
that is hard to disentangle:

• Increased per pupil funding for secondary schools, its increased targeting on poverty,
increased identification of special educational needs, and the increased flexibility to
study vocational options at Key Stage 4 may all have played a role in supporting
better achievement and engagement for pupils at risk of exclusion.

• The teaching assistant workforce grew from just under 8,000 in 1997 to nearly
40,000 in 2010 and provided a significant additional resource to support teachers in
the classroom. Research on the deployment of teaching assistants found that their
presence in class had a positive impact on behaviour, with pupils spending less time
off task, teachers having to spend less time on behaviour management, and more
adult time spent teaching.323

319
Kendall and others.
320
Hallam, Castle, and Rogers.
321
Ofsted, Evaluation of the Impact of Learning Support Units (Ofsted, 2006).
322
DfE, Higher Standards , Better Schools For All: More Choice for Parents and Pupils. (London: HMSO, 2005).
323
Peter Blatchford, Paul Bassett, and Penelope Brown, ‘The Effect of Support Staff on Pupil Engagement and
Individual Attention’, British Educational Research Journal, 35.5 (2009), 661–86.

114
 • Over this period initial teacher training also appears to have improved in ways
closely relevant to behaviour and exclusion. Surveys of newly qualified teachers
(NQTs) one year into the profession showed substantial increases in new teachers’
satisfaction that they had been well prepared to manage behaviour (60 per cent of
NQTs saying their training had been good or very good in 2003, rising to 83 per cent
in 2013). Over the same period satisfaction in relation to training to teach learners
from all ethnic backgrounds rose from 32 to 66 per cent and for training on special
educational needs it rose from 46 per cent to 74 per cent.324

• Local collaboration also played a significant role in how exclusions and behaviour
were managed. In a 2009 study comparing low and high excluding local authorities,
Parsons identified features of local leadership associated with lower exclusion levels.
They were: shared commitment across schools and local authority members and
officers; broadening the school by making more diverse provision in schools; a
commitment to organise managed moves so that school clusters can share
responsibilities; seeing alternative provision as finding or making a place for every
child; joining up the dots to make multi-agency work effective; and winning hearts
and minds to gain support for responding to all children's needs. 325

Social policy and the wider risk factors

Finally, in the broader context of this report we should note the potential impact of
developments outside the world of education policy, such as policies to reduce crime
(Chapter 2) and drug and alcohol use (Chapters 7 and 8) as well as preventive policies for
young people (Chapter 3). These chapters describe reduced self-reported offending, lower
levels of fighting, and reductions in drug and alcohol use by adolescents, all of which are
factors that would have reduced some of the triggers for exclusion.
The rise in exclusions
Explorations of the rise in exclusions since 2013 again tend to focus on mutually reinforcing
systemic changes, including four key elements - changes to the legal process for exclusions,
reduced funding for preventive support, perverse incentives, and changes outside the
school system. Some of the key studies are summarised below:
• Cole and others attribute the rise in exclusions in England since 2012 to a wide range
of educational policy changes - the dropping of proactive efforts to reduce exclusion,
the removal of review panels’ ability to require reinstatement, dissipation of the
focus on the social and emotional aspects of learning, inadequate funding for
support work, and insufficient attention to the interaction of polices on behaviour,
special educational needs, and mental health. 326

324
Department for Education, Newly Qualified Teachers : Annual Survey 2013.
325
Summarised by its author, Carl Parsons in: House of Commons Education Committee, Education Committee
Behaviour and Discipline in Schools : Oral and Written Evidence.
326
Cole and others.

115
 • The IPPR research referred to earlier cites many of the same factors, including the
effect of funding cuts on pastoral support in schools and preventative services out of
school. This report also highlights the impact of curriculum changes, shortfalls in
mental health services, the impact of child poverty and rising numbers of children
classified as children in need.327

• Qualitative research conducted with Local Education Officers in 2017/18 quoted

multiple examples of financial pressures causing schools to exclude rather than add
in extra support for a child with difficulties. This research highlighted a perceived
erosion of support and legitimacy for local authorities’ work to reduce exclusion, and
a financial vicious cycle as more exclusions and moves to special schools shifted costs
onto local authorities and exhausted the ‘High Needs Block’ from which preventive
support ought to be financed.328

• Pressures and perverse incentives within the special educational needs system and
the accountability framework have been noted in many studies. 329 The Education
Select Committee was told by many of those it consulted that schools might be
deliberately failing to identify a child as having special needs to save money, or
excluding pupils to respond to the incentives of Progress 8.330

• An RSA report in 2020 similarly attributed the rise in exclusions to funding and
workforce reductions, GCSE changes, accountability measures, greater school
autonomy, the rise in zero tolerance policies, as well as wider societal drivers such as
growing poverty and mental health problems.331

This is a long list but its content fits plausibly with the profile of change seen in the national
data. The consistent turning points across regions and many sub-categories of exclusion
indicate that practice changed suddenly, nationwide, in or around 2012/13. This suggests a
forceful cause of change such as worsening financial constraints or changed performance
pressures. As we saw in Chapter 1, from 2010 onwards multiple changes in education policy
and funding affected both schools and, within schools, the most vulnerable pupils. Several
other education indicators, such as pupils not attaining Level 2 and pupil absence follow a
similar trajectory with a turning point around 2013 and may be responding to similar forces.

327
Gill, Quilter-Pinner, and Swift.
328
Ian Thompson, Alice Tawell, and Harry Daniels, ‘Conflicts in Professional Concern and the Exclusion of Pupils
with SEMH in England’, Emotional and Behavioural Difficulties, 26.1 (2021), 31–45.
329
Harry Daniels, Ian Thompson, and Alice Tawell, ‘After Warnock: The Effects of Perverse Incentives in Policies
in England for Students With Special Educational Needs’, Frontiers in Education, 4.April (2019), 1–12; National
Audit Office, Support for Pupils with Special Educational Needs and Disabilities in England.
330
House of Commons Education Committee, Forgotten Children: Alternative Provision and the Scandal of Ever
Increasing Exclusions. Paragraph 21.
331
Dannie Mason and others, Pinball Kids (Royal Society of Arts, 2020).

116
These may in turn have been reinforced by other factors external to schools including the
post 2014 resurgence in adolescent drug use described in Chapter 8.

Discussion
School exclusion is a policy area that saw enormous change under the Labour government,
with a significant infusion of extra resource and attention directed towards improving
behaviour and minimising exclusion. After 2010, resources for children most at risk of
exclusion became much scarcer, and it became easier for head teachers to exclude. The
outcomes reflect these policy swings, as well as developments in wider schools and youth
policy. The most disadvantaged pupils were over-represented in the reduction in exclusions
and also over-represented in the increase when exclusions rose again.
A high level of exclusions might be defended by some who feel that the desire for inclusion
should not always prevail. And it is perfectly possible to make the argument that pupils with
severe problems need to be educated in a different setting with high quality specialist
support. But that is not the reality of how the system has been operating. The safeguards to
prevent a child being excluded for inappropriate reasons are now very weak. And many
children off-rolled or excluded from school are being consigned to educational
environments that are under severe pressure, and in some cases to no education at all. To
add to all these negative outcomes, there is no evidence to suggest that behaviour in
schools is any better as a result of these changes.
The ideal, of course, is to be able to support children with problems, and maintain a calm
and safe environment for learning. To achieve this takes time and resource not just in
schools but also in a range of other services that schools need to call on. If this is not
provided, powerful incentives tend to push children out of school. Given the evidence
suggesting that vulnerable pupils can be supported within mainstream schools, if schools
have appropriate resources and policies, the failure to make this possible is a great waste,
with the potential to lead to significant avoidable costs for the pupil and society over a long
period of time.

Post-script: Off-rolling
This chapter has focused on legal exclusions that are formally recorded as such. However,
there is clear evidence that pupils are also sometimes pressured out of schools in informal
ways.
The box below sets out more information on the phenomenon of ‘off-rolling’, the term
which now tends to be used for this practice. In some cases, this may involve a move to
another type of school, in circumstances where the child and their parents might not have
wished to move. In other cases, it may lead to parents saying they will home-educate their
child.

117
 Box SE 1: Off-rolling

Informal and illegal exclusions are not a wholly new problem. In a 2013 report the then
Children’s Commissioner described illegal exclusions as ‘an elephant in the room’. 332

The scale of off-rolling

By its nature off-rolling is unrecorded so it is impossible to know its scale. A YouGov study
involving survey and interview research in 2019 found that 24 per cent of responding
teachers said that off-rolling had occurred in their current or previous school. In senior
leadership teams, 61 per cent of respondents thought it was on the rise.333

Several studies have attempted to infer the scale of off-rolling by looking at pupil
movement data. For instance, Ofsted published an analysis of the figures for pupils
leaving the state system between year 10 and 11 in two successive years (January 2016 to
January 2017 and January 2017 to January 2018). In both cases, approaching 20,000
pupils left a state funded secondary school during this important twelve-month period. In
the first year, 810 schools had lost 5 or more pupils and 5 per cent of their cohort. In the
second year 940 schools had done so. This level of pupil movement during the GCSE years
is surprising given that this is a time when most families seek to avoid moves if they can.

In both the years examined around half of the pupils who had left their school did not
reappear in another state school. Ofsted’s analysis found that for children in the GCSE
years there was a higher proportion of children disappearing from the state system
entirely than for younger cohorts (51 per cent compared with 29 per cent). 334 335 Some
degree of movement out of the sector can be explained by issues such as migration to
another country, or a move to a school that has not been recorded (e.g. an independent
school). But it also seems likely that some children are not receiving an appropriate
education after leaving their original school.

Home-schooling may be part of the picture. Off-rolling is often masked by what appears
to be a parental decision to home-school their child. The number of children being home-
schooled has risen sharply in recent years, even before the pandemic. The Office of the
Schools Adjudicator’s 2018/19 report found that the total number of children local
authorities reported as being electively home-educated (all ages) had increased by 14.7
per cent in one year. Three successive annual reports by the Adjudicator recorded
concern from local authorities about children ending up in elective home education as a
result of inappropriate pressure and in situations where parents are not in a position to

332
Children’s Commissioner, Always Someone Else’s Problem (CCO, 2013).
333
YouGov, Exploring the Issue of Off-Rolling (Ofsted, 2019).
334
Jason Bradbury, ‘Off-Rolling: An Update on Recent Analysis’, 6 September 2019 <https://
educationinspection.blog.gov.uk/2019/09/06/off-rolling-an-update-on-recent-analysis/>.
335
For further analysis, see also: Warwick Mansell, ‘The Strange Case of the Vanishing GCSE Pupils’, The
Guardian, 21 January 2014; Warwick Mansell, ‘Where Did All the GCSE Pupils Go – and Why Has No One
Noticed?’, The Guardian, 21 March 2017; Philip Nye and Dave Thomson, Who’s Left 2019, Part Two: How Do
You Lose 6,700 Pupils? (FFT Education Datalab, 2019).

118
 home-educate successfully. 336 The numbers home-educated have risen further since the
pandemic, as has concern about a variety of related welfare and educational issues.337

The process and reasons for off-rolling

Many factors, some of them overlapping, are thought to lie behind off-rolling.

• In the YouGov research quoted earlier most respondents thought behavioural

reasons lay behind off-rolling, but ‘maintaining a high position in a league table’
was respondents’ second most common understanding of the reason for a pupil
being off-rolled.
• Ofsted inspectors reported that the instigation of formal procedures to fine or
prosecute parents for child’s poor attendance was on occasion followed by a
parental decision to home-educate their child.338 The Children's Commissioner has
also found that some removals from rolls were a response to the threat of
attendance prosecution or the risk of the child being formally excluded.339
• One final cause, identified by the Office of the Children's Commissioner in a 2013
report, is a structural gap. In the words of their report, with the partial exception
of Ofsted, they found no statutory body, either local or national, proactively
seeking to address illegal exclusions. They also observed that there were limited
consequences for a school that broke the law - no clear appeals mechanism or
financial penalty. 340
The policy response

Concern about the numbers of children out of school and their education was one of the
reasons behind the Timpson Review’s recommendation that pupil moves between
schools should be systematically tracked. This was to be taken forward in the Schools Bill
2022. However, as noted above, this Bill was withdrawn later that year and the next step
is unclear.

In relation to off-rolling, Ofsted’s new framework for inspection now makes it very clear
that allowing off-rolling is inconsistent with good leadership.341. Since 2021, their
guidance to inspectors says that they will explicitly use the term ‘off-rolling’ in inspection
reports where the practice is found, since some schools had issued press releases stating
that off-rolling was not found because Ofsted had not used the precise term.342

336
S Scott, Office of the Schools Adjudicator Annual Report September 2018 to August 2019 (OSA, 2020).
337
Centre for Social Justice, Out of Sight and Out of Mind: Shining a Spotlight on Home Education in England
(CSJ, 2022).
338
Her Majesty’s Chief Inspector of Education Children’s Services and Skills, Annual Report 2018/19.
339
Children’s Commissioner, Falling through the Gaps in Education (CCO, 2017).
340
Children’s Commissioner, Always Someone Else’s Problem.
341
https://www.gov.uk/government/publications/education-inspection-framework/education-inspection-
framework
342
https://www.gov.uk/government/publications/schools-and-early-education-inspection-update-academic-
year-2021-to-2022/schools-and-early-education-inspection-update-september-2021#reporting-on-off-rolling

119
Chapter 5 School Absence

Introduction

Absence from school is a serious problem with damaging and long-term consequences for
children’s education, health, and welfare. But absence is an issue which attracted
surprisingly little attention in education debates over the last decade, until rates of non-
attendance rates soared in the wake of the pandemic. Many people, including many
politicians, are unaware of the scale of absence, the multitude of factors that lie behind it,
or the size of the reduction in absence levels that took place between 2000 and 2013.
This chapter summarises findings on secondary school absence to explore what happened
during that reduction, and since. The chapter begins with definitions and headline data,
and some analysis of the causes and costs of children missing school. It then discusses
absence policies, outcomes, and plausible explanations for the trends.

Key data and definitions

The term absence covers pupils missing school for any reason, with or without permission.
There are three main ways of measuring it, with data published regularly. 343
• The overall absence rate measures the loss of schooling across the whole school
population, including both pupils who miss little or no schooling, and those who miss
large amounts. In 2018/19, the overall amount of schooling missed rate in state-
funded secondary schools stood at 5.5 per cent of sessions. In 1997/98 it had been 9
per cent.
• Government statistics also measure the number of young people who miss large
amounts of school. Persistent absence measures the proportion of pupils missing at
least 10 per cent of their schooling. As of 2018/19, for state funded secondary
schools, persistent absence stood at 13.7 per cent. In 2006/07 (the first date when
we can measure the amount of individual pupils’ absence) the equivalent figure was
24.9 per cent.
• Severe absence refers to the number of young people who miss more than 50 per
cent of their schooling. As of 2018/19, the severe absence rate was 1.3 per cent for
state-funded secondary schools, having been 1.6 per cent in 2006/07.

This chapter mainly considers data on absence up to 2018/19, but briefly outlines the
statistics for 2021/22 (see Table SA1). Post-pandemic, overall absence, persistent absence,
and severe absence are all substantially higher.
Other terms are sometimes used to describe absence, notably ‘truancy’. People sometimes
use truancy as a synonym for all absence, or just for unauthorised absence, or just for
343
DfE, Pupil absence in schools in England

120
absence that is not known to parents. Because of this ambiguity, the term will not be used
in this report, except where it is the term used in a quoted source.

Why children miss school

Absence can be triggered by multiple features of young people’s lives at home, in school
and in their lives outside. Historically, the most frequent reason given by children for not
going to school regularly is illness (physical or mental). After that, the most commonly
cited issues in a 2019 survey were: being bullied or having been bullied (9 per cent), lack of
support for health/disability, boredom (both 8 per cent) struggling to keep up (7 per cent),
and transport problems (6 per cent). 344 But a range of other problems in and out of school
may contribute. Figure SA1 illustrates the diversity of these reasons.345
Figure SA1: Reasons why young people may be absent from school

Groups over-represented in absence

School absence may become a problem for any child, but some groups are heavily over-
represented in the absence statistics. DfE statistics show that the national persistent
absence rate was 13.7 per cent in 2018/19, but for young people on free school meals, the
persistent absence rate was nearly 30 per cent, and for those who have special educational

344
Department for Education, Omnibus Survey of Pupils and Their Parents or Carers: Wave 3 (DfE, 2019).
345
Sources for this Figure are listed at the end of the chapter.

121
needs, the persistent absence rate was over 23 per cent.346 There are similar disparities in
severe absence rates. (Table SA1)
There is a strong age dimension to absence: persistent absence rates peak in the GCSE
years, with 15.9 per cent of year 10 pupils persistently absent in 2018/19. There are also
considerable ethnic disparities, with particularly high rates of absence experienced by
Travellers of Irish heritage and Gypsy Roma pupils, for whom persistent absence rates were
over 50 per cent in 2018/19. 347
Children with mental disorders are much more likely than other children to have had time
off school. In 2004, 17 per cent of those with emotional disorders and 14 of those with
conduct disorders had been away from school for over 15 days in the previous term,
compared with just 4 per cent of children without a mental disorder.348 A large study linking
education and health data sets for more than 400 000 pupils in Wales found that individuals
with more than one neurodevelopmental disorder or mental disorder were more likely to
miss school, with each extra disorder exacerbating the likelihood. But having SEN status
reduced the likelihood of being absent, highlighting the positive effect of recognition,
diagnosis and intervention.349
Alcohol use is strongly associated with school absence. A 2010 analysis of the first LSYPE
Cohort found that at ages 14-15 young people who drank on most days had over 4 times the
odds compared to other young people of increasing their truancy, and at ages 15-16 this
had increased to over 10 times the odds.350

The costs of school absence

The absence of a child from school poses a range of short- and long-term concerns. The first
issue must always be to establish that the child is safe, is not in danger, or being exploited,
and has not run away. Beyond that, a range of other concerns arise, given the association
between school absence and other negative outcomes. For example:
• Absence is linked to reduced educational attainment. Controlling for a range of other
pupil characteristics, persistent absenteeism over the final Key Stage 4 year has been
found to have a strong relationship with GSCE attainment. 351

• US evidence has found that truancy can act as an accelerating factor for young
people’s drug use.352 School surveys in England consistently find that pupils who

346
These groups overlap: as of 2019, in state-funded secondary schools, a quarter of pupils recorded as having
special educational needs were also eligible for free school meals, and 22 per cent of pupils eligible for free
school meals also had identified special needs. (DfE, Special educational needs in England: 2019)
347
DfE, Pupil absence in schools in England
348
Hazel Green and others, Mental Health of Children and Young People in Great Britain, 2004 (ONS, 2005).
349
John and others.
350
Rosie Green and Andy Ross, Young People’s Alcohol Consumption and Its Relationship to Other Outcomes
and Behaviour (DfE, 2010).
351
Education Standards Analysis and Research Division, A Profile of Pupil Absence in England (DfE, 2011).
352
Kimberly L. Henry, Kelly E. Knight, and Terence P. Thornberry, ‘School Disengagement as a Predictor of
Dropout, Delinquency, and Problem Substance Use During Adolescence and Early Adulthood’, Journal of Youth
and Adolescence, 41.2 (2012), 156–66.

122
 have ever truanted are more likely to have drunk alcohol in the last week, and to
have used drugs in the last month.353 As noted above, researchers have looked at
what the first LSYPE Cohort can tell us about links between alcohol and truancy.
They found that drinking alcohol was a stronger predictor of truancy than truancy
was of drinking alcohol. Nonetheless, truancy was a predictor of trying alcohol
among young people who had not previously tried it at ages 14 and 15.354
• School absenteeism is associated with an increased risk of self-harm, and of suicidal
ideation.355
• Multiple studies show an association between absence and teenage pregnancy.356
• As noted in Chapter 4, persistent absentees appear to be over-represented in the
phenomenon of ‘off-rolling’. 357
• Young people with a history of persistent absence in secondary school are heavily
overrepresented amongst 18-year-old long term NEETs.358
• Research by the Ministry of Justice and Department for Education found that 9 per
cent of young people who had ever been persistent absentees had been cautioned
or sentenced for an offence by age 17, and 2 per cent were also cautioned or
sentenced for a serious violence offence. Where there was a serious violence
offence, for 96 per cent of children the persistent absence preceded the violent
offence.359

Absence imposes costs on teachers, who need to find catch-up time for children who have
missed lessons. Absence is a source of stress for parents too, with concern about fines and
possible prosecution compounding the parent’s worry about the child’s absence and the
issues that may underpin it.360
All these factors mean that absence has high costs to society as a whole. Aggregating these
is difficult, but a 2007 study put the ten-year cost of the 198,000 children (then) missing
more than 20 per cent of their schooling in UK primary and secondary schools at £8.8
billion.361

353
Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
England 2018 (NHS Digital, 2018). Table 5.26 and 8.10.
354
Green and Ross.
355
Sophie Epstein and others, ‘School Absenteeism as a Risk Factor for Self-Harm and Suicidal Ideation in
Children and Adolescents: A Systematic Review and Meta-Analysis’, European Child and Adolescent Psychiatry,
1, 2019, 1–20.
356
Claire Crawford, Jonathan Cribb, and Elaine Kelly, Teenage Pregnancy in England (Institute for Fiscal Studies,
2013); Alison C.S. Hosie, ‘“I Hated Everything About School”: An Examination of the Relationship between
Dislike of School, Teenage Pregnancy and Educational Disengagement’, Social Policy and Society, 6.3 (2007),
333–47.
357
Jo Hutchinson and Whitney Crenna-Jennings, Unexplained Pupil Exits: Further Analysis by Multi-Academy
Trust and Local Authority (EPI, 2019).
Department for Education, Characteristics of Young People Who Are Long-Term NEET (DfE, 2018).
358

Department for Education and Ministry of Justice. The study population was in Key Stage 4 in the years
359
2012/13, 2013/14 or 2014/15.
360
Rona Epstein, Geraldine Brown, and Sarah O’Flynn, Prosecuting Parents for Truancy : Who Pays the Price?
(Coventry University and Roehampton University, 2019).
361
Martin Brookes, Emilie Goodall, and Lucy Heady, Misspent Youth: The Costs of Truancy and Exclusion (New
Philanthropy Capital, 2007).
123
Absence policies and spending programmes
Labour: 1997–2010
The Labour government in 1997 identified school absence as a priority because of the
impact on both attainment and crime. A 1998 Social Exclusion Unit report analysed the
causes and proposed a programme of action including improvements to the management
of pupil absence (for example replacing paper record-keeping with electronic registration,
and introducing first-day calling of parents/carers of absent pupils) as well as preventive
action to tackle the underlying causes of absence, and a national target to reduce
unauthorised absence by a third.362 The Department for Education took on this target and
funded extensive programmes of spending and support over the following years in
partnership with schools, local authorities, educational welfare services, and the police.
A notable example of these preventive programmes was Excellence in Cities (EiC) which
provided resources for learning mentors to support pupils with academic or personal issues,
and also funded Learning Support Units to provide short-term teaching and support
programmes within schools for children who needed this. Beginning in 1999 with 24 local
authority areas including all twelve Inner London boroughs, EiC grew to cover a third of
secondary schools in England by 2003/04. Its successor, the Behaviour Improvement
Programme (BIP) funded a menu of measures to improve behaviour and attendance and
covered the majority of local authorities by 2005/06.363 Figure SA2 below shows the options
chosen by BIP Phase 1 areas, and how they addressed some of the underlying causes of
absence (and exclusion) such as educational difficulties, lack of interest in school, family
issues, and bullying and crime in schools.
Figure SA2: Behaviour Improvement Programme initiatives by frequency and theme

Frequency Theme
Police in schools 53 Bullying / fear of
Anti- bullying programmes 91 crime

Registration, truancy sweeps, attendance officers 71 Attendance

management
Learning support units 65
Key workers for young people 41 Pupil support

Bored,
Alternative curriculum development 65 disengaged

Extended schools: study support, sports, arts,… 59 Support for

Supporting parents with children at risk 41 families

0 20 40 60 80 100
(% of local areas adopting)
Source: Hallam, Castle and Rogers, Research and Evaluation of the Behaviour Improvement Programme Programme,
2005

362
Social Exclusion Unit, Truancy and School Exclusion. As Director of the Social Exclusion Unit, I was involved
in the development of this report.
363
National Audit Office, Improving School Attendance in England.

124
dŚĞŽǀĞƌĂůůĚŝƌĞĐƚŝŽŶŽĨƐĐŚŽŽůƐƉŽůŝĐLJĚĞƐĐƌŝďĞĚŝŶŚĂƉƚĞƌƐϭĂŶĚϰǁĂƐĂůƐŽƌĞůĞǀĂŶƚƚŽ
ĚĞǀĞůŽƉŵĞŶƚƐŝŶĂďƐĞŶĐĞ͗ŝŶĐƌĞĂƐĞƐŝŶĨƵŶĚŝŶŐĞŶĂďůĞĚƐĐŚŽŽůƐƚŽĞŵƉůŽLJŵŽƌĞƐƚĂĨĨĂŶĚ
ƉƌŽǀŝĚĞŵŽƌĞƐƵƉƉŽƌƚ͖ĨůĞdžŝďŝůŝƚLJƚŽŝŶĐůƵĚĞŵŽƌĞǀŽĐĂƚŝŽŶĂůŽƉƚŝŽŶƐĂƚ<ĞLJ^ƚĂŐĞϰŚĞůƉĞĚ
ƌĞƚĂŝŶĂŶĚŵŽƚŝǀĂƚĞƐŽŵĞƉƵƉŝůƐǁŚŽŵŝŐŚƚŽƚŚĞƌǁŝƐĞŚĂǀĞďĞĐŽŵĞĚŝƐĞŶŐĂŐĞĚ͖ĂŶĚƚŚĞ
ŝŶƚƌŽĚƵĐƚŝŽŶŽĨƚŚĞĚƵĐĂƚŝŽŶ DĂŝŶƚĞŶĂŶĐĞůůŽǁĂŶĐĞƉƌŽǀŝĚĞĚĂƌŽƵƚĞƚŽĐŽŶƚŝŶƵĞĚƐƚƵĚLJ
ĨŽƌƉƵƉŝůƐǁŚŽŵŝŐŚƚŽƚŚĞƌǁŝƐĞŶŽƚŚĂǀĞďĞĞŶĂďůĞƚŽĂĨĨŽƌĚƚŽĐŽŶƚŝŶƵĞŝŶĞĚƵĐĂƚŝŽŶĂĨƚĞƌ
ϭϲ͘ŚĂƉƚĞƌϮ͕ϯ͕ϳĂŶĚϴĐŽǀĞƌƉŽůŝĐLJĚĞǀĞůŽƉŵĞŶƚƐƵŶĚĞƌƚŚĞ>ĂďŽƵƌŐŽǀĞƌŶŵĞŶƚǁŝƚŚĂ
ďĞĂƌŝŶŐŽŶŽƚŚĞƌƌŝƐŬĨĂĐƚŽƌƐĨŽƌĂďƐĞŶĐĞͲLJŽƵƚŚŽĨĨĞŶĚŝŶŐ͕ĐŚŝůĚƉŽǀĞƌƚLJ͕ĂĚŽůĞƐĐĞŶƚ
ĚƌŝŶŬŝŶŐ͕ĂŶĚĂĚŽůĞƐĐĞŶƚĚƌƵŐƵƐĞ͘
>ĂďŽƵƌ͛ƐĨŝƌƐƚƚĂƌŐĞƚ͕ƚŽƌĞĚƵĐĞƵŶĂƵƚŚŽƌŝƐĞĚĂďƐĞŶĐĞďLJĂƚŚŝƌĚ͕ǁĂƐŶŽƚŵĞƚ͘ƵƚƚĂƌŐĞƚƐ
ĐŽŶƚŝŶƵĞĚƚŽďĞƐĞƚ;ŶŽǁĨƌĂŵĞĚŝŶƚĞƌŵƐŽĨŽǀĞƌĂůůĂďƐĞŶĐĞ͕ƚŽĂǀŽŝĚĂƉĞƌǀĞƌƐĞŝŶĐĞŶƚŝǀĞ
ƚŽĂƵƚŚŽƌŝƐĞĂďƐĞŶĐĞͿĂŶĚŽǀĞƌĂůůƐĞĐŽŶĚĂƌLJƐĐŚŽŽůĂďƐĞŶĐĞďĞŐĂŶĂƐƚĞĂĚLJĨĂůů͘LJϮϬϬϲ͕
ƚŚĞĞƉĂƌƚŵĞŶƚĨŽƌĚƵĐĂƚŝŽŶĞŵƉůŽLJĞĚƌĞŐŝŽŶĂůĂĚǀŝƐĞƌƐƚŽĨŽĐƵƐŽŶƐĐŚŽŽůƐǁŝƚŚƚŚĞ
ŚŝŐŚĞƐƚůĞǀĞůƐŽĨĂďƐĞŶĐĞ͕ƐƵƉƉŽƌƚŝŶŐƚŚĞŵƚŽƉƵƚŝŶƉůĂĐĞĂĐƚŝŽŶƉůĂŶƐĨŽƌŝŶĚŝǀŝĚƵĂů
ĂďƐĞŶƚĞĞƐŝŶƚŚĞƐĞƐĐŚŽŽůƐ͕ŝŶĐůƵĚŝŶŐĂĚĞĚŝĐĂƚĞĚĨŝĞůĚŽĨĨŝĐĞƌ͕ƉĂƌĞŶƚĂůůLJĂŐƌĞĞĚŐŽĂůƐ͕ĂŶĚ
ĐŽŽƌĚŝŶĂƚĞĚŵƵůƚŝͲĂŐĞŶĐLJƐƵƉƉŽƌƚ͘ϯϲϰ&ƌŽŵϮϬϬϲŽŶǁĂƌĚƐ͕ƐƚĂƚŝƐƚŝĐƐǁĞƌĞĐŽůůĞĐƚĞĚŝŶĂǁĂLJ
ƚŚĂƚĂůůŽǁĞĚƚŚĞŶƵŵďĞƌŽĨƉĞƌƐŝƐƚĞŶƚĂďƐĞŶƚĞĞƐƚŽďĞƚƌĂĐŬĞĚ͘
ůŽŶŐƐŝĚĞŵĞĂƐƵƌĞƐŝŶĂŶĚĂƌŽƵŶĚƐĐŚŽŽůƐ͕ƚŚƌŽƵŐŚŽƵƚƚŚĞŝƌƚŝŵĞŝŶŽĨĨŝĐĞ>ĂďŽƵƌĂůƐŽ
ĞŵƉŚĂƐŝƐĞĚƚŚĞƌĞƐƉŽŶƐŝďŝůŝƚLJŽĨƉĂƌĞŶƚƐĨŽƌƚŚĞŝƌĐŚŝůĚƌĞŶ͛ƐƐĐŚŽŽůĂƚƚĞŶĚĂŶĐĞ͘dŚĞLJ
ůĞŐŝƐůĂƚĞĚĨŽƌŶĞǁƐĂŶĐƚŝŽŶƐ;&ŝdžĞĚWĞŶĂůƚLJEŽƚŝĐĞƐǁŚĞƌĞĂƉĂƌĞŶƚŽƌĐĂƌĞƌǁĂƐƵŶǁŝůůŝŶŐƚŽ
ŝŵƉƌŽǀĞƚŚĞŝƌĐŚŝůĚ͛ƐĂƚƚĞŶĚĂŶĐĞ͕ĂŶĚŝŵƉƌŝƐŽŶŵĞŶƚĂƐĂƉŽƐƐŝďůĞƐĂŶĐƚŝŽŶŝŶĐĂƐĞƐǁŚĞƌĞ
ƉĂƌĞŶƚƐŬŶŽǁƚŚĞŝƌĐŚŝůĚŝƐŶŽƚĂƚƚĞŶĚŝŶŐƐĐŚŽŽůĂŶĚƚĂŬĞŶŽƌĞĂƐŽŶĂďůĞƐƚĞƉƐƚŽƐĞĐƵƌĞ
ƚŚĞŝƌĂƚƚĞŶĚĂŶĐĞͿ͘ƵƚƚŚĞLJĂůƐŽŝŶƚƌŽĚƵĐĞĚƚŚĞ;ŵĂŶĚĂƚŽƌLJͿWĂƌĞŶƚŝŶŐKƌĚĞƌ͕ǁŚŝĐŚ
ƌĞƋƵŝƌĞĚƉĂƌĞŶƚƐƚŽĂƚƚĞŶĚĐůĂƐƐĞƐƚŽŚĞůƉƚŚĞŵŝŵƉƌŽǀĞƚŚĞŝƌĐŚŝůĚ͛ƐĂƚƚĞŶĚĂŶĐĞĂŶĚƚŚĞ
;ǀŽůƵŶƚĂƌLJͿWĂƌĞŶƚŝŶŐŽŶƚƌĂĐƚƐĞƚƚŝŶŐŽƵƚǁŚĂƚƐĐŚŽŽůƐĂŶĚƉĂƌĞŶƚƐǁŽƵůĚĚŽƚŽŝŵƉƌŽǀĞ
ĂƚƚĞŶĚĂŶĐĞ͘ϯϲϱ^ƵƌǀĞLJĂŶĚĐĂƐĞƐƚƵĚLJƌĞƐĞĂƌĐŚŝŶϮϬϭϬǁŚŝĐŚůŽŽŬĞĚĂƚWĂƌĞŶƚŝŶŐŽŶƚƌĂĐƚƐ͕
WĞŶĂůƚLJEŽƚŝĐĞƐ͕ĂŶĚWĂƌĞŶƚŝŶŐKƌĚĞƌƐ͕ĂƐǁĞůůĂƐĨĂƐƚƚƌĂĐŬĂƉƉƌŽĂĐŚĞƐƚŽƉƌŽƐĞĐƵƚŝŽŶ͕
ĨŽƵŶĚĞǀŝĚĞŶĐĞƚŽƐƵŐŐĞƐƚƚŚĂƚƚŚĞLJǁĞƌĞĐŽŶƚƌŝďƵƚŝŶŐƚŽŝŵƉƌŽǀĞŵĞŶƚƐŝŶĂƚƚĞŶĚĂŶĐĞĨŽƌ
ƉƵƉŝůƐ͕ǁŚĞŶƵƐĞĚĞĂƌůLJŽƌĂƐƉĂƌƚŽĨĂǁŝĚĞƌŝŶƚĞŐƌĂƚĞĚůŽĐĂůĂƵƚŚŽƌŝƚLJĂƉƉƌŽĂĐŚƚŽ
ĂĚĚƌĞƐƐŝŶŐŝƌƌĞŐƵůĂƌĂƚƚĞŶĚĂŶĐĞĂŶĚƉĞƌƐŝƐƚĞŶƚĂďƐĞŶĐĞ͘ϯϲϲϮϬϬϰƐƚƵĚLJŽĨƉĂƌĞŶƚŝŶŐ
ƉƌŽŐƌĂŵŵĞƐĨŽƵŶĚŝŶƚĞƌĞƐƚŝŶŐƌĞƐƵůƚƐ͗ƐLJƐƚĞŵƐĨŽƌƉƌŽǀŝĚŝŶŐƉĂƌĞŶƚŝŶŐƉƌŽŐƌĂŵŵĞƐŝŶůŽĐĂů
ĂƵƚŚŽƌŝƚŝĞƐǁĞƌĞĨƌĂŐŝůĞ͕ƉƌŽǀŝƐŝŽŶǁĂƐŐĞŶĞƌĂůůLJŝŶƐŚŽƌƚƐƵƉƉůLJĂŶĚĨƵŶĚŝŶŐǁĂƐŝŶƐĞĐƵƌĞ͘
,ŽǁĞǀĞƌ͕ϵϳƉĞƌĐĞŶƚŽĨƌĞƐƉŽŶĚŝŶŐƉĂƌĞŶƚƐƌĞƉŽƌƚĞĚƚŚĂƚƚŚĞƉƌŽŐƌĂŵŵĞƐǁĞƌĞĞŶũŽLJĂďůĞ
ĂŶĚŚĞůƉĨƵů͕ĂŶĚƚŚĞƌĞǁĂƐĞǀŝĚĞŶĐĞŽĨŝŵƉƌŽǀĞĚďĞŚĂǀŝŽƵƌĂŶĚĂƚƚĞŶĚĂŶĐĞĂƚƐĐŚŽŽůĂƐĂ
ƌĞƐƵůƚŽĨƉĂƌĞŶƚƐ͛ĂƚƚĞŶĚĂŶĐĞĂƚƉĂƌĞŶƚŝŶŐƉƌŽŐƌĂŵŵĞƐ͘ϯϲϳ

ϯϲϰ
ĞƉĂƌƚŵĞŶƚĨŽƌĚƵĐĂƚŝŽŶĂŶĚ^ŬŝůůƐ͕Departmental Report 2006;Ĩ^͕ϮϬϬϲͿ͘
ϯϲϱ
WĂƌĞŶƚŝŶŐKƌĚĞƌƐǁĞƌĞŝŶƚƌŽĚƵĐĞĚďLJƚŚĞƌŝŵĞĂŶĚŝƐŽƌĚĞƌĐƚϭϵϵϴ͘dŚĞŝŶĐƌĞĂƐĞĚƐĂŶĐƚŝŽŶŽĨ
ŝŵƉƌŝƐŽŶŵĞŶƚǁĂƐŝŶƚŚĞƌŝŵŝŶĂů:ƵƐƚŝĐĞĂŶĚŽƵƌƚ^ĞƌǀŝĐĞƐĐƚϮϬϬϬ͘WĞŶĂůƚLJEŽƚŝĐĞƐ;ĨŝŶĞƐͿĂŶĚƚŚĞ
WĂƌĞŶƚŝŶŐŽŶƚƌĂĐƚǁĞƌĞŝŶƚŚĞŶƚŝͲ^ŽĐŝĂůĞŚĂǀŝŽƵƌĐƚϮϬϬϯ͘
ϯϲϲ
<ĂƚŚƌLJŶƌŽǁƚŚĞƌĂŶĚ^ĂůůLJ<ĞŶĚĂůů͕Investigating the Use of Parental Responsibility Measures for School
Attendance and Behaviour;Ĩ͕ϮϬϭϬͿ͘
ϯϲϳ
^ƵƐĂŶ,ĂůůĂŵ͕>LJŶŶĞZŽŐĞƌƐ͕ĂŶĚ:ĂĐƋƵĞůĞŶĞ^ŚĂǁ͕Improving Children’s Behaviour and Attendance through
the Use of Parenting Programmes: An Examination of Good Practice;Ĩ^͕ϮϬϬϰͿ͘

ϭϮϱ
Coalition: 2010–2015
In 2010 the new government inherited a trend of falling absence. It commissioned a report
on attendance from its Expert Advisor on Behaviour and implemented its recommendations
to increase the level of attendance fines and remove headteachers’ discretion to authorise
family holiday during term.368 The government also tightened the definition of persistent
absentees to cover pupils missing 10 per cent of school, and included absence from school
as one of three criteria for inclusion in the Troubled Families Programme. The number of
fines began a steady climb, as Figure SA3 below illustrates. By the Coalition’s last year, the
number of penalty notices issued was six times the level of 2009/10. By 2018/19, it was
thirteen times the 2009/10 level.
Figure SA3: Trends in number of penalty notices issued for attendance, England, all school
types

400,000
333,388
350,000
300,000
250,000
200,000 151,125
150,000
100,000
50,000 25,657

0
2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 2018/19

Source: DfE, Parental Responsibility Measures for Attendance, 2018 to 2019

By contrast, during this period, there was little support for what schools and other services
needed to do to tackle the causes of absence. The duty for schools to participate in
behaviour and attendance partnerships was repealed in 2011, and spending reductions and
policy changes sapped some of the resources and flexibilities that had been most important
in reducing absence. As described in Chapter 1, per pupil funding in secondary school began
a period of real terms decline, falling by 3 per cent in real terms between 2011/12 and
2015/16, and pupil teacher ratios began to worsen. From 2013 onwards, schools faced
budgetary disincentives to identify pupils as having special educational needs. Changes to
GCSEs altered both the curriculum and the assessment method during years 10 and 11, the
peak period for school absence, and reduced access to vocational qualifications, and the
Education Maintenance Allowance was abolished and replaced with a less generous
scheme.
The support for schools from other key external services – the police, Connexions, youth
services, mental health services, and drug and alcohol services – all suffered cuts. Benefits
for low-income families fell behind inflation. Education welfare budgets fell by 39 per cent in

368
Charlie Taylor, Improving Attendance at School (DfE, 2012).

126
five years.369 Ofsted inspections remained a possible safeguard against high levels of
absence, but these inspections became less frequent for many schools as a result of the
scaling back of inspection requirements described in Chapter 1.
Conservatives: 2015 onwards
As the outcomes section of this chapter will show, both severe absence and the overall
absence rates of disadvantaged pupils had begun to rise again by 2015. However, this did
not trigger any major policy attention. The Troubled Families programme continued, and
still included absence, though now as one of six criteria.370 Evaluation of the programme
found that there was no consistent evidence of significant or systematic impact on school
attendance.371 372
Pressures in schools funding were by now making it very hard to resource the support for
children with poor attendance. Secondary school teacher numbers were 20,000 lower in
2018 than in 2010, and the squeeze on funding in the decade after 2009–10 meant a 13 per
cent real terms cut for spending per pupil in the fifth of secondary schools with most free
school meals pupils (against an 8 per cent fall for the best-off fifth). 373 374 Implementation
of the Progress 8 accountability measure for schools left little space for vocational
qualifications and caused some schools to reduce the provision for arts, design technology,
ICT, PE and religious studies, affecting curriculum choice and interest for many learners. 375
Plans to strengthen mental health provision in schools and colleges (see Box SA1 below)
were a welcome initiative that began to roll out in 2018/19 but only reached a quarter of
learners by 2022.
A recent (2022) report by the Children's Commissioner paints a disturbing picture of the
local management of attendance as it has operated in recent years. It found that:

• autonomous schools frequently now use incompatible attendance management

systems which get in the way of reporting and exchange of information.
• there is limited partnership working between Multi Academy Trusts and Local
Authorities
• roles and responsibilities are ambiguous, and problems get ‘passed on’ to other
services.
• local authority attendance teams frequently have unmanageable caseloads. In one
authority, over 15,000 children were persistently absent in autumn 2021, which

369
School Home Support, ‘Research Highlights Significant Cuts to the Education Welfare Service’, 2015
<https://www.schoolhomesupport.org.uk/news/research-highlights-significant-cuts-to-the-education-welfare-
service/> .
370
Philip Loft, The Troubled Families Programme (House of Commons Library, 2020).
371
Ministry of Housing Communities and Local Government.
372
Day and others.
373
Britton, Farquharson, and Sibieta.
374
Kate Ogden, David Phillips, Luke Sibieta, and Ben Zaranko, ‘Is Public Service Spending Aligned with the
“Levelling up” Agenda?’, 18 March 2022 <https://ifs.org.uk/articles/public-service-spending-aligned-levelling-
agenda>.
375
Cooper Gibson Research.
127
 would equate to a caseload of over 3000 children per member of staff in the
inclusion and attendance team. 376
Some additional policies on attendance have been announced since 2019 which should be
mentioned briefly here. In September 2021, the Department of Education announced that
£30 million would be invested into a programme of SAFE (‘Support, Attend, Fulfil, Exceed’)
taskforces to be rolled out in 10 serious violence hotspots areas from early 2022, as a 3-year
initiative to be led by local schools and to protect young people at risk of truancy or being
permanently excluded. 377 The ten areas were announced in December that year, with a
new national attendance alliance also formed at that point to ‘supercharge efforts to
improve school attendance’.378 379

Mental health policies

Policy on child and adolescent mental health (CAMHS) is relevant to many of the subjects
discussed in this report, but has a particular relevance to school absence. For many years
referred to as a ‘Cinderella service’, CAMHS has developed considerably over the last
twenty years, but evidence suggests that there remains considerable unmet need. Box SA1
below summarises some of the key developments in policy, spending and demand.

376
Children’s Commissioner, Where Are England’s Children? Interim Findings from the Attendance Audit.
377
https://www.gov.uk/government/news/targeted-support-for-vulnerable-young-people-in-serious-violence-
hotspots
378
https://www.gov.uk/government/news/support-for-vulnerable-young-people-in-serious-violence-hotspots
379
https://www.gov.uk/government/news/education-secretary-launches-new-attendance-alliance

128
Box SA1: Child and Adolescent Mental Health Policy

Labour: 1997-2010
Under Labour, child and adolescent mental health services (CAMHS) received additional
investment in early spending reviews.380 Spending on CAMHS increased substantially
from £284 million in 2002/03 to a projected £565 million in 2008/09.381 The government
set targets from 2003, requiring all child and adolescent mental health services to provide
comprehensive services including mental health promotion and early intervention by
2006, and pledging to increase child and adolescent mental health services by at least 10
per cent each year across the service according to agreed local priorities.382 Mental
health featured as one of the five outcomes in the Every Child Matters Green Paper and
the NHS articulated good practice expectations in a 2004 National Service Framework:
both these documents positioned CAMHS as ‘a broad concept embracing all services that
contribute to the mental health care of children and young people, whether provided by
health, education, social services or other agencies’ and thus a shared responsibility
between Primary Care Trusts and local authorities, with an emphasis on the role of local
agencies in collaborating on prevention and information sharing.383

A 2008 review of progress found evidence of some improvements. Between 2005 and
2007, there had been a 24 per cent fall in the number of children and young people
waiting to be seen, an increase in the number of local authorities reporting fully
comprehensive provision for children and young people with complex needs (from 23 per
cent of areas to 53 per cent) and an increase in the provision of round the clock on-call
CAMHS services (from 44 of areas to 56 per cent). But the review also highlighted
unacceptable variations between areas, and found that children, young people, and
families were still waiting too long.384 The government accepted the review’s
recommendations, promising to put in place a National Support Programme to drive
improvements at national and local levels, announcing further investment in both
universal and specialist services, including the Targeted Mental Health in Schools
programme, a £60 million project which ran between 2008 and 2011, eventually reaching
over 2,500 schools across all local authorities in England.385

Coalition and Conservatives: 2010 to 2019

The Coalition government published several strategy documents designed to raise the
profile of mental health, initially framed in terms of mental health for people of all
ages.386 This included some specific programmes for young people, such as £60 million to
improve access to psychological therapies for young people. But the overall funding
picture was much more challenging. Spending on CAMHS by Primary Care Trusts fell in
real terms between 2010 and 2012.387 And other preventive support funded by schools
and local authorities was subject to the funding pressures described in Chapters 1 and 2.

By 2015, data was showing that there were increases in referrals and waiting times, with
providers reporting increased complexity and severity of presenting problems. 388 A
review by the Children’s Commissioner in 2016 found that on average, 28 per cent of
children and young people referred to CAMHS were not allocated a service, rising to 75
per cent in one region. In one CAMHS in the West Midlands the average waiting time was
200 days. 389

129
 dŚĞϮϬϭϱƵĚŐĞƚ;ƵŶĚĞƌƚŚĞŽĂůŝƚŝŽŶͿĂůůŽĐĂƚĞĚάϭ͘ϮϱďŝůůŝŽŶĨŽƌŝŶǀĞƐƚŵĞŶƚŝŶĐŚŝůĚƌĞŶ͛Ɛ
ŵĞŶƚĂůŚĞĂůƚŚŽǀĞƌƚŚĞĨŽůůŽǁŝŶŐĨŝǀĞLJĞĂƌƐ͘ŽŶƐĞƌǀĂƚŝǀĞ'ƌĞĞŶWĂƉĞƌƉƵďůŝƐŚĞĚŝŶ
ϮϬϭϳŝŶĐůƵĚĞĚƐƉĞĐŝĨŝĐƉƌŽƉŽƐĂůƐĨŽƌƐĐŚŽŽůƐĂŶĚĐŽůůĞŐĞƐƚŽŚĂǀĞĂĚĞƐŝŐŶĂƚĞĚƐĞŶŝŽƌůĞĂĚ
ĨŽƌŵĞŶƚĂůŚĞĂůƚŚ͕ƚŽĨƵŶĚŶĞǁŵĞŶƚĂůŚĞĂůƚŚƐƵƉƉŽƌƚƚĞĂŵƐ͕ĂŶĚƚŽƉŝůŽƚĂĨŽƵƌͲǁĞĞŬ
ǁĂŝƚŝŶŐƚŝŵĞĨŽƌĂĐĐĞƐƐƚŽƐƉĞĐŝĂůŝƐƚĐŚŝůĚƌĞŶĂŶĚLJŽƵŶŐƉĞŽƉůĞ͛ƐŵĞŶƚĂůŚĞĂůƚŚ
ƐĞƌǀŝĐĞƐ͘ϯϵϬ

dŚĞ'ƌĞĞŶWĂƉĞƌǁĂƐŐĞŶĞƌĂůůLJǁĞůĐŽŵĞĚďƵƚŶŽƚƐĞĞŶĂƐĂƉĂŶĂĐĞĂ͘EĂƚŝŽŶĂůƵĚŝƚ
KĨĨŝĐĞƌĞƉŽƌƚĨŽƵŶĚƚŚĂƚƚŚĞE,^ůĂĐŬĞĚůĞǀĞƌƐĂŶĚĚĂƚĂƚŽďĞĐĞƌƚĂŝŶƚŚĂƚƚŚĞŵŽŶĞLJǁĂƐ
ďĞŝŶŐƐƉĞŶƚŽŶLJŽƵŶŐƉĞŽƉůĞ͛ƐŵĞŶƚĂůŚĞĂůƚŚ͕ĂŶĚƚŚĂƚƚŚĞgŽǀĞƌŶŵĞŶƚĚŝĚŶŽƚŚĂǀĞ
ĐŽƐƚĞĚƉůĂŶƐĨŽƌŝƚƐŽďũĞĐƚŝǀĞƐ͕ŽƌƚŚĞĂďŝůŝƚLJƚŽĚƌŝǀĞƉƌŽŐƌĞƐƐŽŶŝƚƐƚĂƌŐĞƚƐ͘ϯϵϭŶĂůLJƐŝƐ
ďLJƚŚĞĚƵĐĂƚŝŽŶWŽůŝĐLJ/ŶƐƚŝƚƵƚĞŝŶϮϬϭϴ;ďĂƐĞĚŽŶ&K/ƌĞƋƵĞƐƚƐͿĨŽƵŶĚƐŽŵĞƉƌŽŐƌĞƐƐ
ƐŝŶĐĞϮϬϭϮŝŶƌĞĚƵĐŝŶŐƚŚĞůŽŶŐĞƐƚƚƌĞĂƚŵĞŶƚǁĂŝƚŝŶŐƚŝŵĞƐ͕ďƵƚĂƐŝŐŶŝĨŝĐĂŶƚƉŽƐƚĐŽĚĞ
ůŽƚƚĞƌLJ͘ZĞĨĞƌƌĂůƐƚŽƐƉĞĐŝĂůŝƐƚD,^ŚĂĚŝŶĐƌĞĂƐĞĚďLJϮϲƉĞƌĐĞŶƚŝŶĨŝǀĞLJĞĂƌƐĂŶĚ
ďĞƚǁĞĞŶĂĨŝĨƚŚĂŶĚĂƋƵĂƌƚĞƌŽĨĐŚŝůĚƌĞŶƌĞĨĞƌƌĞĚǁĞƌĞďĞŝŶŐƌĞũĞĐƚĞĚĂƐŝŶĂƉƉƌŽƉƌŝĂƚĞ
ĨŽƌƐƉĞĐŝĂůŝƐƚƚƌĞĂƚŵĞŶƚ͘ϯϵϮ

dŚĞ'ƌĞĞŶWĂƉĞƌ͛ƐƐƉĞĐŝĨŝĐƉƌŽƉŽƐĂůƐĨŽƌŶĞǁƚĞĂŵƐŝŶƐĐŚŽŽůƐĂŶĚĐŽůůĞŐĞƐǁĞƌĞƉŝůŽƚĞĚ
ŝŶ͚ƚƌĂŝůďůĂnjĞƌĂƌĞĂƐ͛ďĞŐŝŶŶŝŶŐŝŶϮϬϭϴͬϭϵǁŝƚŚϱϴƚĞĂŵƐ͘dŚĞƌŽůůͲŽƵƚŽĨƚŚĞƐĞƚĞĂŵƐ
ŽǀĞƌůĂƉƉĞĚǁŝƚŚƚŚĞƉĂŶĚĞŵŝĐ͕ďƵƚďLJ^ƉƌŝŶŐϮϬϮϮƚŚĞE,^ƐĂŝĚƚŚĂƚĐŽǀĞƌĂŐĞŚĂĚ
ƌĞĂĐŚĞĚũƵƐƚŽǀĞƌĂƋƵĂƌƚĞƌŽĨƉƵƉŝůƐŝŶƐĐŚŽŽůƐĂŶĚ&ƵƌƚŚĞƌĚƵĐĂƚŝŽŶ͘ϯϵϯ

ĂĐŬŐƌŽƵŶĚĚĂƚĂŽŶŵĞŶƚĂůŚĞĂůƚŚŝƐƐĞƚŽƵƚŝŶƚŚĞĂƉƉĞŶĚŝdžƚŽƚŚŝƐƌĞƉŽƌƚ͕ĐŽǀĞƌŝŶŐƚŚĞ
ƉƌĞǀĂůĞŶĐĞŽĨŵĞŶƚĂůĚŝƐŽƌĚĞƌƐ͕ƐĞůĨͲŚĂƌŵĂŶĚƐƵŝĐŝĚĞ͘

ϯϴϬ
,DdƌĞĂƐƵƌLJ͕Prudent for a Purpose : Building Opportunity and Security for All͘
ϯϴϭ
ZĂĐŚĞů:ĞŶŬŝŶƐĂŶĚŽƚŚĞƌƐ͕͚ŚŝůĚĂŶĚĚŽůĞƐĐĞŶƚDĞŶƚĂů,ĞĂůƚŚ͗/ŶĨƌĂƐƚƌƵĐƚƵƌĞ͕WŽůŝĐŝĞƐĂŶĚWƌĂĐƚŝĐĞƐŝŶ
ŶŐůĂŶĚ͗dŚĞD,WƌŽũĞĐƚ͕͛Journal of Public Mental Health͕ϵ͘ϭ;ϮϬϭϬͿ͕Ϯϲʹϯϵ͘
ϯϴϮ
E,^ŶŐůĂŶĚ͕Improvement, Expansion and Reform; the next 3 Years;,͕ϮϬϬϮͿ͘
ϯϴϯ
,D'ŽǀĞƌŶŵĞŶƚ͕Every Child Matters͖ĞƉĂƌƚŵĞŶƚĨŽƌĚƵĐĂƚŝŽŶĂŶĚ^ŬŝůůƐĂŶĚĞƉĂƌƚŵĞŶƚŽĨ,ĞĂůƚŚ͕
National Service Framework for Children, Young People and Maternity Services: The Mental Health and
Psychological Wellbeing of Children and Young People;Ĩ^͕ϮϬϬϰͿ͘
ϯϴϰ
EĂƚŝŽŶĂůD,^ZĞǀŝĞǁ͕Final Report;^&ĂŶĚ,͕ϮϬϬϴͿ͘
ϯϴϱ
DtŽůƉĞƌƚ͕:ĞŝŐŚƚŽŶ͕ĂŶĚWWĂƚĂůĂLJ͕Me and My School: Findings from the National Evaluation of Targeted
Mental Health in Schools 2008-2011.;Ĩ͕ϮϬϭϭͿ͘
ϯϴϲ
ĞƉĂƌƚŵĞŶƚŽĨ,ĞĂůƚŚ͕No Health without Mental Health: Outcomes Strategy for People of All Ages;,D'͕
ϮϬϭϭͿ͘
ϯϴϳ
,ŽƵƐĞŽĨŽŵŵŽŶƐtƌŝƚƚĞŶŶƐǁĞƌŚƚƚƉƐ͗ͬͬƋƵĞƐƚŝŽŶƐͲƐƚĂƚĞŵĞŶƚƐ͘ƉĂƌůŝĂŵĞŶƚ͘ƵŬͬǁƌŝƚƚĞŶͲ
ƋƵĞƐƚŝŽŶƐͬĚĞƚĂŝůͬϮϬϭϰͲϭϮͲϭϲͬϮϭϴϴϲϱ
ϯϴϴ
ĞƉĂƌƚŵĞŶƚŽĨ,ĞĂůƚŚ͕Future in Mind;,͕ϮϬϭϱͿ͘
ϯϴϵ
ŚŝůĚƌĞŶ͛ƐŽŵŵŝƐƐŝŽŶĞƌ͕Lightning Review: Access to Child and Adolescent Mental Health Services;K͕
ϮϬϭϲͿ͘
ϯϵϬ
ĞƉĂƌƚŵĞŶƚŽĨ,ĞĂůƚŚĂŶĚĞƉĂƌƚŵĞŶƚŽĨĚƵĐĂƚŝŽŶ͕Transforming Children and Young People’s Mental
Health Provision: A Green Paper;,D'͕ϮϬϭϳͿ͘
ϯϵϭ
EĂƚŝŽŶĂůƵĚŝƚKĨĨŝĐĞ͕Improving Children and Young People’s Mental Health Services;EK͕ϮϬϭϴͿ͘
ϯϵϮ
ƌĞŶŶĂͲ:ĞŶŶŝŶŐƐĂŶĚ,ƵƚĐŚŝŶƐŽŶ͘
ϯϵϯ
ŚƚƚƉƐ͗ͬͬǁǁǁ͘ĞŶŐůĂŶĚ͘ŶŚƐ͘ƵŬͬŵĞŶƚĂůͲŚĞĂůƚŚͬĐLJƉͬƚƌĂŝůďůĂnjĞƌƐͬηͺDĞŶƚĂůͺ,ĞĂůƚŚͺ^ƵƉƉŽƌƚ

ϭϯϬ
Outcomes
Data on overall absence and persistent absence at secondary level shows a significant
improvement and narrowing inequalities until 2013/14. There was then some deterioration,
in rates of severe absence, and in the absence rates of disadvantaged groups.
Most of the data in this section covers the period until 2018/19, reflecting the main focus of
this report and the fact that the usual data release was not issued in 2019/20. The post-
pandemic position is summarised in Table SA1.

Trends in the three key measures

The three charts at Figure SA4 set out the trends in the three main measures of absence
for secondary schools. They show that:

• When Labour came into office in 1997, overall absence, i.e. the proportion of half
days missed in secondary schools was 9 per cent. After 2000/01 this measure began
an almost uninterrupted decline to reach 5.2 per cent in 2013/14.394 This was a 43
per cent fall over a period of thirteen years. Overall secondary school absence
edged up again after 2013/14 to reach 5.5 per cent in 2018/19.

• The reduction in persistent absence was even steeper. In just seven years (between
2006/07, when it was first measured, and 2013/14) there was a 45 per cent fall in
the proportion of secondary pupils missing more than 10 per cent of their
schooling.395 The rate then plateaued from the middle of the second decade. As of
2018/19, there were 400,000 persistent absentees in secondary schools.

• Severe absence (missing more than 50 per cent of schooling) also saw a large fall
from when it was first measured in 2006/07. In the seven years until 2013/14 it
halved, from 1.6 per cent of secondary pupils to 0.8 per cent, then increased again to
reach 1.3 per cent in 2018/19.

394
The rise in absence in 2005/06 was attributed by the department to outbreaks of flu and norovirus. The
increase in absence (primary and secondary) that year was seen in authorised absence, not unauthorised.
(DfES Annual Report 2007)
395
This itself was probably a reduction from earlier in the decade, but persistent absence was only measured
from 2006/07.

131
Figure SA4: Trends in overall absence, persistent absence, and severe absence, state-funded secondary schools, England

(Note - different scales on axes)

Overall absence (per cent of all sessions Proportion of pupils missing more than ten Proportion of pupils missing more than 50%
missed) per cent of their schooling of schooling

10 30% 1.8%
9.0
1.6%
9 1.6%
24.9%
25%
8
1.4% 1.3%

7
20% 1.2%
5.5
6 5.2
1.0%
13.6%
Per cent of sessions missed

13.7%
5 15% 0.8%

0.8%
4
10% 0.6%
3

2 0.4%
5%
1 0.2%

0 0% 0.0%
1997/98
1998/99
1999/00
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
2017/18
2018/19

1997/98
1998/99
1999/00
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
2017/18
2018/19

1997/98
1998/99
1999/00
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
2017/18
2018/19
Source: DCSF / DfE, Pupil absence in schools in England. (Pre 2006 data from SFR 05/2008)

132
Inequalities
Long run trends on inequalities are mostly only published for overall absence, and then only
back to the mid-2000s.
Free school meals, special educational needs
During the period of reduction, overall absence rates showed reducing gaps in relation to
free school meals status, and special educational needs status. Since 2013/14 these
inequalities have widened again Figure SA5.
Figure SA5: Trends in overall absence in state-funded secondary schools in England by free
school meals and special educational needs status

Free school meals gap The special educational needs gap

14 14

12 12

10 10
Per cent of sessoins missed

8 8

6 6

4 4

2 2

0 0
2017/18
1997/98
1998/99
1999/00
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17

2018/19
1997/98
1998/99
1999/00
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
2017/18
2018/19

eligible for free school meals Statements of SEN/ EHC plan

not eligible for free school meals No identified SEN

Source: DfE, Pupil absence in schools in England

Ethnicity
Figure SA6 shows overall absence rates for selected ethnic minorities in 2005/06,
2013/14 and 2018/19. The largest percentage point falls in the first period were for pupils
of Bangladeshi ethnicity, followed by pupils of Mixed White and Black Caribbean
ethnicity. In the second period the largest increases were for Mixed White and Black
Caribbean pupils, followed by Black Caribbean pupils.

133
Figure SA6: Overall absence in state-funded secondary schools in England: change by
ethnic group. Selected years

12

10
Per cent of all sessions missed

0
2005/06
2013/14
2018/19

2005/06
2013/14
2018/19

2005/06
2013/14
2018/19

2005/06
2013/14
2018/19

2005/06
2013/14
2018/19

2005/06
2013/14
2018/19

2005/06
2013/14
2018/19

2005/06
2013/14
2018/19

2005/06
2013/14
2018/19

2005/06
2013/14
2018/19
White and White Pakistani White and White and Black Indian Black Chinese
Black British Bangladeshi Black Asian Caribbean African
Caribbean African

Source: DfE Pupil absence in schools in England, Table 5.3

Throughout this period, the two ethnic groups with the highest absence rates were
Travellers of Irish heritage and Gypsy/Roma pupils. These groups would be off the scale of
the chart above. In 2005/06 the percentage of sessions missed for Travellers of Irish
heritage in secondary schools was 26.3 per cent, and for Gypsy/Roma pupils it was 23.2 per
cent. The 2018/19 figures were lower, but still extremely concerning - 16.5 per cent and
14.5 per cent respectively.396
Gender
Girls had very slightly higher rates of absence than boys in the mid-2000s, but the gender
difference in overall absence had disappeared by 2018/19. However, as Table SA1 shows, in
the post-pandemic data, girls’ absence has risen more than that of boys.
Years 10 and 11
During the years of falling absence the greatest reductions were in years 11 and 10 (the pre-
GCSE years, and the age where absence tends to peak). After 2013/14, absence edged up in
all year groups, but the rise in year 11 was very muted. (Figure SA7) It is possible that these
trends reflect off-rolling: investigations of off-rolling have found that pupils with absence
problems are heavily over-represented amongst those leaving state school rolls in the run-
up to year 11.397 For persistent absence, year 11s also show different trends from other
years after 2013/14 (not shown).

396
DfE, Pupil Absence SFRs for 2005/06 and 2018/19
397
Hutchinson and Crenna-Jennings; YouGov.

134
Figure SA7: Trends in overall absence in state-funded secondary schools in England by
national curriculum year

12

10
per cent of schooling missed

Yr 7 Yr 8 Yr 9 Yr 10 Yr 11

Source: DfE, Pupil absence in schools in England, successive releases

Regional patterns
Data at a regional level shows broadly the same trends and turning points as the national
data, but there are significant regional inequalities, and the position of London has
improved relative to other regions. Figure SA8 below shows that secondary absence levels
in Inner London (the black line) fell very rapidly over the three years 2001/02 to 2003/04.
Having been the worst region for secondary school absence, Inner London became one of
the best in the space of a very few years. It retained and widened this advantage, as did
Outer London (the brown line). London as a whole has had much better secondary
attendance than other regions for over a decade.
The other two regions in the top three for absence in 1997 were the North East and
Yorkshire and Humberside. They were still in the top two in 2018/19. The North East region
and Yorkshire and Humberside have also had the highest persistent absence rates through
much of the period and saw the largest increases after 2013/14.398

398
As of 2018/19, the secondary school persistent absence rate for the North East was 15.6 per cent of pupils,
for Yorkshire and Humberside it was 15.3 per cent, and for London it was 12 per cent.

135
Figure SA8: Trends in overall absence in state-funded secondary schools, by region

12

10
Per cent of schooling missed

North East North West Yorkshire and the Humber

East Midlands West Midlands East of England
Inner London Outer London South East
South West

Source: DfE, Pupil absence in schools in England

Post-pandemic update

The Covid public health crisis that began in 2020, had a major effect on school
attendance and has left a legacy of much higher levels of absence. Table SA1 shows that:

• in 2021/22, persistent absence and severe absence were both twice the level of
2018/19
• 45 per cent of secondary school pupils eligible for free school meals were persistent
absentees.
• one in 20 secondary school pupils eligible for free school meals was missing half their
education through absence.
• girls’ non-attendance had increased by much more than that of boys, and nearly 30
per cent of female secondary school pupils were persistent absentees.

136
Table SA1: The rise in overall, persistent and severe absence after the pandemic, state
funded secondary schools, England 399

2018/19 2021/22
Overall absence 5.5% 9.0%

Persistent absence
Total persistent absence rate 13.7% 27.7%
(proportion of pupils missing more than 10% of their schooling)
Persistent absence rate for pupils eligible for free school meals 28.5% 45.3%
Persistent absence rate for pupils with SEN statement or EHCP 23.7% 38.8%
Persistent absence rate for pupils with SEN Support 23.2% 39.5%
Persistent absence rate for girls 13.7% 29.2%
Persistent absence rate for boys 13.7% 26.3%
Persistent absence rate year 9 14.7% 29.5%
Persistent absence rate year 10 15.9% 30.5%
Persistent absence rate year 11 15.7% 31.9%

Severe absence
Severe absence rate 1.3% 2.7%
(proportion of pupils missing more than 50% of their schooling)
Severe absence rate for pupils eligible for free school meals 3.1% 5.7%
Severe absence rate for pupils with SEN statement or EHCP 3.8% 7.8%
Severe absence rate for pupils with SEN Support 2.9% 5.7%
Severe absence rate for girls 1.3% 3.0%
Severe absence rate for boys 1.3% 2.4%
Severe absence rate year 9 1.3% 2.7%
Severe absence rate year 10 1.6% 3.3%
Severe absence rate year 11 1.7% 3.5%

399
https://explore-education-statistics.service.gov.uk/find-statistics/pupil-absence-in-schools-in-england

137
Plausible explanations of the trends in England

Explanations of trends in absence and persistent absence in England have to account for an
unusual trajectory, steady improvement from 2001 until 2013 then a stalling from 2014,
with worsening rates for disadvantaged groups and rising levels of severe absence. In the
existing literature, there is no dedicated study examining the causes of these absence trends
over this long period. But there is good evaluation evidence on some of the initiatives
undertaken in England (see Box SA1 below). There is also a large literature on ‘what works’
aimed at practitioners.400
The fall in absence
Drawing on the literature, and the outcome data analysed above, plausible reasons for the
fall in absence between 2001 and 2013 include:
Sustained commitment and quantified outcome targets
The steps taken by the new government in 1997 to get to grips with school absence gave
the issue greater priority than it had previously received. Reducing absence was an explicit
government target in every spending review during the Labour government. This helped to
unlock resources in spending reviews, led to high visibility in the lead department
(Education) and incentivised government to introduce policies that would reduce the
absence rate.
Effective absence policies implemented at scale
Between 1997 and 2010, the government invested significant time and resource in
initiatives designed to tackle the many causes and triggers of absence. There is extensive
research evidence on their impact, much of it positive. Key points are summarised in Box
SA2 below, which covers both large area-based programmes such as Excellence in Cities and
the Behaviour Improvement Programme, and individual components that were sometimes
implemented under these programmes and sometimes separately.
Although some of these initiatives started only in a few places, the effort grew to a scale
large enough to drive the national trend. And the regional pattern of roll-out - with early
efforts concentrated in London – may also help to explain the faster start made in London
compared with other regions.

See for instance: Susan Hallam and Lynne Rogers, Improving Behaviour and Attendance at School (Open
400

University Press, 2008); Ken Reid, Managing School Attendance: Successful Intervention Strategies for
Reducing Truancy (Routledge, 2014).

138
Box SA2: Evidence on Absence Initiatives under the Labour Government: 1997 to 2010

• Electronic registration schemes: the National Audit Office found these schemes were
rated effective in improving attendance by 80 per cent of secondary head teachers who
had experienced them. 401 The full evaluation also found them to be good value for
money. 402

• Excellence in Cities was found by three different studies to have had a positive impact on
absence. 403

• Evaluation of the Behaviour Improvement Programme found a statistically significant

reduction in overall absence, as well as perceived positive changes in the status of
behaviour and pastoral issues in school, school policies and practices, school ethos, the
way that schools supported families, children’s well-being and learning, relationships with
parents, and staff stress.404

• Ofsted found positive impacts from the post 2005 Behaviour and Attendance Strategy,
with a 27 per cent reduction in the number of persistent truants in the target schools.
Specific improved practices included a higher profile for attendance in some ‘coasting’
schools and local authorities, a higher status for the Education Welfare Service in schools,
and increased multi-agency involvement with Connexions, police liaison, health advisers,
and child and adolescent mental health services.405

• Experiments with more flexible curricula and alternative providers during the GCSE years
showed some positive results in building engagement, confidence, and attitudes towards
learning amongst disengaged and low-attending students.406

• Evaluation of police-school partnerships found promising results in relation to truancy

rates. 407

Wider schools policy

These specific initiatives to reduce absence were reinforced by the broader educational
policies of the time. Labour’s schools policy (discussed in more detail in Chapter 1) is
normally assessed in relation to the impact on improved attainment and post-16
participation. The likely impact on attendance is perhaps less appreciated. However, the

401
National Audit Office, Improving School Attendance in England.
402
Geoff Lindsay and others, Evaluation of Capital Modernisation Funding for Electronic Registration in
Selected Secondary Schools (DfES, 2006).
403
Ofsted, Excellence in Cities and Education Action Zones: Management and Impact; Machin, McNally, and
Meghir; National Audit Office, Improving School Attendance in England.
404
Hallam, Castle, and Rogers.
405
Ofsted, Attendance in Secondary Schools (Ofsted, 2007).
406
Sarah Golden, Lisa O Donnell, and Peter Rudd, Evaluation of Increased Flexibility for 14 to 16 Year Olds
Programme : The Second Year (DfES, 2005); Susan Hallam and others, ‘Pupils’ Perceptions of an Alternative
Curriculum: Skill Force’, Research Papers in Education, 22.1 (2007), 95–111.
407
Roger Bowles, Maria Garcia Reyes, and Rima Pradiptyo, Safer Schools Partnerships (Youth Justice Board,
2005).

139
increase in schools funding and workforce, changes to the eligibility of vocational
qualifications at Key Stage 4, and the introduction of the Education Maintenance Allowance
also had the capacity to address absence drivers such as young people falling behind, being
bored, and not expecting to continue in school after 16.
Social policy and the wider risk factors
The reduction in absence may also reflect reductions in some of the risk factors for absence
such as alcohol and drug use. As noted in other chapters of this report, between 2001 and
2014, the proportion of 14- and 15-year-olds who had drunk alcohol in the last week fell by
two thirds; between 2003 and 2011 the proportion of 11–15-year-old pupils who had taken
(any) drugs in the last month halved; and the number of proven offences committed by 10–
17-year-olds reduced by more than half between 2004 and 2013. All of these trends will
have reduced the proportion of secondary school pupils who might be considered at risk of
becoming absentees.
The rise in absence after 2013/14

For the first two years of the Coalition, absence figures continued to improve, but the
numbers began to edge up again after 2013/14 for the most vulnerable groups. Off-rolling
may have masked the true scale of the problem. The reasons for the stalling in progress may
well be the mirror image of factors influencing the earlier improvement.
Reduced priority
The incoming government in 2010 had a significant change agenda for education and it was
perhaps inevitable that some issues went down the pecking order. Absence was one of
these, and this may have seemed reasonable as the numbers were still falling. The
government set itself no targets to reduce absence, resourcing for educational welfare fell,
and the government’s policy approach to absence was small scale, concentrated on parental
and family factors. Government policy does not seem to have tracked the issue of rising
absence among disadvantaged groups, or addressed its multiple potential drivers.
Ineffective absence policies
Evidence does not suggest that the Coalition and Conservatives’ absence policies were
effective. The Troubled Families Programme has been evaluated twice and found to have
had no sustained effect on absence. 408 The ban on term-time holidays has not been subject
to evaluation. However, outcome data does not suggest it reduced absence overall. As
Figure SA9 below illustrates, days missed due to holiday in secondary schools rose by 30 per
cent in the year between the policy change being announced (in 2012) and coming into
effect (2013/14). After implementation, days lost to family holidays fell slightly, but by
2017/18 the number of sessions missed due to holiday was back to pre-policy-change levels,
despite now being unauthorised, and despite significant use of fines for non-compliance.

408
Ministry of Housing Communities and Local Government; Day and others.

140
Figure SA9: Trends in absence due to holidays in state-funded secondary schools

6,000,000

5,000,000 New holiday policy New holiday policy

announced implemented
Total half days missed

4,000,000

3,000,000

2,000,000

1,000,000

Authorised holiday Authorised extended holiday Unauthorised holiday

Source: DfE, Pupil absence in schools in England

Changes to schools policy

The absence data shows a break in trend in 2013/14 when a long period of falling absence
came to an end. This timing aligns with a large set of changes to secondary schooling which
appear to have had a significant impact on the vulnerable groups most at risk of
absenteeism. These changes, which are described in more detail in Chapter 1, include:
rising pupil-teacher ratios (from 2013); altering SEN funding in a way that incentivised
schools not to identify pupils as having special educational needs (from 2013); and excluding
large numbers of vocational qualifications from GCSE equivalence (took effect in summer
2014).
Under the Conservatives this direction of policy continued: school funding cuts were
concentrated on the most deprived local authorities; the SEN system became increasingly
unsustainable; and the new Progress 8 performance measure further tilted the curriculum
away from the vocational and towards the academic. These policy decisions undercut
schools’ ability to provide extra support for pupils who were falling behind in lessons; and
made it harder for some pupils to study the subjects that motivated them. The changes
coincided with a significant weakening of local authorities’ ability to support attendance,
and should have been expected to add to the upward pressures on absence, impacting
pupils with special educational needs in particular, which is precisely what happened.
Social policy and the wider risk factors
Data from other chapters of this report gives more detail on how policies changed in
relation to some of the other adolescent issues which can be triggers for absence. Under the
Coalition and the Conservatives, support services for young people with drug and alcohol
problems were cut, benefits for low-income families were uprated by less than inflation,

141
support services such as Connexions were effectively abolished, and demand for mental
health services grew faster than supply. Adverse trends in these areas increased the
proportion of secondary school pupils who might be considered at risk of becoming
absentees.

Discussion

The analysis above shows that the striking trends in absence mirror dramatic changes in
government policy relating to absence and its causes. Absence fell after a period of well-
resourced, well evaluated, and sustained national effort, backed up by well-resourced
schools and youth policies which were focused on prevention. When these policies changed,
and the support services, flexibilities, and funding streams which helped to reduce absence
withered away, absence started to nudge up again, especially for the most vulnerable.
Given the importance of school absence, it is hard to understand why it receives so little
attention in public debate about schools policy. This seems likely to change now, for the
saddest of reasons, as the Covid pandemic has played havoc with school attendance, and
left a legacy of disrupted routines, lost learning, and increased mental health problems for
many young people. The attention being given by the Children’s Commissioner to absent
and missing children is very welcome and should cast an important spotlight on the number
of children who are out of school and the reasons for their situation. 409
Recent survey data (collected in January 2022 when schools were open and attendance
required) found that anxiety or mental health was now a common reason for non-
attendance, being given as a reason by 22 per cent of secondary pupils who had missed
some schooling in the previous two weeks, and 30 per cent of those who had not attended
school at all over that period.410 Support for mental health will clearly have to be a major
part of the response to the current levels of absence. But there will never be just one reason
for all pupil absence, and all the other factors that have historically caused absence will
continue to play a role. To take just one example, a different survey, conducted in summer
2022 and looking at teenagers’ experience and perceptions of violence, found that 14 per
cent of respondents had missed school in the previous twelve months in order to protect
themselves from violence or make themselves feel safe.411 These striking figures remind us
of the diversity of reasons that may lead a child to miss school, and the importance of a
broad-based approach to tackling the causes of non-attendance.
As we face the significant scale of absence from schools, there are important lessons to be
learnt from past success in reducing absence. Above all, we are unlikely to bring absence
down from its current high levels without a sustained, adequately resourced national effort,
which works effectively with schools, local authorities and other partners, addresses the

409
https://www.childrenscommissioner.gov.uk/report/voices-of-englands-missing-children/
410
Sarah Hingley and others, Parent, Pupil and Learner Panel Recruitment Wave 1: November 2021 to January
2022 (DfE, 2022).
411
Youth Endowment Fund, Children, Violence and Vulnerability (YEF, 2022).

142
multiple causes of absence, and focuses in particular on the most vulnerable groups.
Starting where we are, this is a considerable challenge.

Sources for Figure SA1

Children’s Commissioner, Attendance Is Everyone's Business (CCO, 2023)

Colechin, Jane, Combating Truancy (Family Lives/Parentline, 2011)

Finning, Katie, Obioha C. Ukoumunne, Tamsin Ford, Emilia Danielson-Waters, Liz Shaw,
Ingrid Romero De Jager, and others, ‘Review: The Association between Anxiety and Poor
Attendance at School – a Systematic Review’, Child and Adolescent Mental Health, 24.3
(2019), 205–16

Finning, Katie, Polly Waite, Kate Harvey, Darren Moore, Becky Davis, and Tamsin Ford,
‘Secondary School Practitioners’ Beliefs about Risk Factors for School Attendance
Problems: A Qualitative Study’, Emotional and Behavioural Difficulties, 25.1 (2020), 15–28

Goodall, Emilie, School’s out? Truancy and Exclusion: A Guide for Funders (New
Philanthropy Capital, 2005)

Gubbels, Jeanne, Claudia E. van der Put, and Mark Assink, ‘Risk Factors for School
Absenteeism and Dropout: A Meta-Analytic Review’, Journal of Youth and Adolescence,
48.9 (2019), 1637–67

Heyne, David, Malin Gren-Landell, Glenn Melvin, and Carolyn Gentle-Genitty,

‘Differentiation Between School Attendance Problems: Why and How?’, Cognitive and
Behavioral Practice, 26.1 (2019), 8–34

Malcolm, Heather, Valerie Wilson, Julia Davidson, and Susan Kirk, Absence from School : A
Study of Its Causes and Effects in Seven LEAs (Department for Education and Skills, 2003)

Reid, Ken, ‘The Causes of Non-Attendance: An Empirical Study’, Educational Review, 60.4
(2008), 345–57

Wimmer, Mary, ‘Why Kids Refuse to Go to School … and What Schools Can Do about It.’,
Education Digest, 74.3 (2008), 32–37

143
Chapter 6 Teenage Pregnancy
In the late 1990s the level of teenage pregnancy in the UK was a matter of great public
concern. While many other Western European countries had seen falling teenage birth rates
in the 1980s and 1990s, the UK had not. Over the following two decades, the under-18
conception rate in England fell by two thirds, and the under-18 birth rate fell by nearly three
quarters. Steep reductions in conception rates occurred for teenagers of all ages after 2007.
This chapter looks at the circumstances behind these striking trends. It first sets out some
key background facts – the number of teenage conceptions and births, who is most
affected, and why teenage conceptions matter. It then charts the development of policy
under Labour, Coalition and Conservatives, sets out the trends in conception and birth rates,
and discusses possible reasons for the changing trends.

Key data

Overall numbers
In 2019, there were just over 14,000 conceptions to under-18 girls in England.412 More than
half ended in abortion, and there were 6,350 births. Compared with 1998, the conception
rate to under-18 girls had fallen by two thirds, and the under-18 birth rate by three
quarters. Table TP1 shows these figures in more detail.
Table TP1: Conceptions, births, and abortions to girls under 18: England, 1998 and 2019 413

1998 2019
Conceptions (number) 41,089 14,019
Conceptions (rate per 1,000) 46.6 15.7

Births (number) 23,667 6,351

Births (rate per 1,000) 26.9 7.1
Abortions (number) 17,422 7,668
Abortions as % of conceptions 42.4 54.7

Age of mother
The breakdown by age of mother at conception is shown in Table TP2 overleaf. The age
breakdown is only published for England and Wales combined.

This data covers any pregnancy conceived to a young woman under 18 whether it ends in abortion or
412
maternity.
413
Office for National Statistics, Conceptions in England and Wales: 2019 (ONS, 2021).

144
Table TP2: Births and conceptions by mother’s age, 1998 and 2019, England and Wales

Mother’s age Conceptions Births

(absolute numbers and % of <18 total)
1998 2019 1998 2019
<14 423 (1.0%) 110 (0.7%) 170 37
14 1,988 (4.5%) 614 (4.1%) 821 225
15 6,041 (13.7%) 1,706 (11.5%) 3,032 646
16 13,802 (31.3%) 4,198 (28.3%) 8,078 1,951
17 21,865 (49.6%) 8,229 (55.4%) 13,503 3,913
All < 18 44,119 (100%) 14,857 (100%) 25,604 6,772

Age of father (live births)

The available data from birth registrations shows that for births to women under 20, the
father was also aged under 20 in 35 per cent of registrations. For a further 48 per cent of
under-20 registrations, the father was aged between 20 and 24.414

Groups over-represented in teenage conceptions

In the UK, as across the world, teenage pregnancy is more common amongst girls who have
grown up in poverty, those who are disengaged from education, and those who lack access
to sex education and contraception. Research in England and the UK has explored aspects
of this.

• Teenage conceptions and births are more common in deprived areas, and eligibility
for free school meals has been found to be independently associated with teenage
pregnancy in IFS research.415
• Multiple studies show an association between school absenteeism and teenage
pregnancy. Some also suggest connections with exclusion from school. 416
• Dislike of school is strongly associated with early sexual activity, unprotected sex,
and pregnancy.417
• Declining attainment during school years is associated with teenage conception and
birth, as is expecting or wanting to leave school at the first opportunity. 418

414
Office for National Statistics, Births by Parents’ Characteristics, England and Wales (ONS, 2019).
415
Office for National Statistics, Conceptions in England and Wales: 2018 (ONS, 2020); Crawford, Cribb, and
Kelly.
416
Crawford, Cribb, and Kelly; Hosie; Department for Education and Skills, Teenage Pregnancy Next Steps:
Guidance for Local Authorities and Primary Care Trusts on Effective Delivery of Local Strategies (DfE, 2006);
Mike Robling, Evaluating the Family Nurse Partnership in England: The Building Blocks Trial (Cardiff University,
2015).
417
C. Bonell and others, ‘The Effect of Dislike of School on Risk of Teenage Pregnancy: Testing of Hypothesis
Using Longitudinal Data from a Randomized Trial of Sex Education’, Journal of Epidemiology and Community
Health, 59.3 (2005), 223–30.
418
Kathleen E. Kiernan, ‘Becoming a Young Parent : A Longitudinal Study of Associated Factors’, The British
Journal of Sociology, 48.3 (1997), 406–28; Crawford, Cribb, and Kelly; Angela Harden and others, ‘Teenage

145
Adverse childhood experiences (ACEs) are also associated with higher prevalence of teenage
pregnancy. Research in England has found that men and women exposed to four or more
ACEs were four times more likely to have had or caused an unintended teenage pregnancy
than those with no ACEs (controlling for socio-demographic factors).419
Several factors are known to be associated with the age of first sexual activity, and with the
use or non-use of contraception.

• Evidence suggests that school-based sex education delays the onset of sexual activity
and increases contraceptive use among sexually active young people. 420
• A 2015 study found that citing school as the main source of information had a
significant association with older age at first sex. 421
• Early regular alcohol consumption is associated with early onset of sexual activity,
and several different studies have found associations between adolescent alcohol
use and unprotected sex. 422
• Research into why young people don’t use contraception or don’t use it effectively
has found multiple reasons including not being aware of contraceptive choices, not
knowing how to access services, concern about confidentiality, sex being unplanned,
not taking the pill consistently, and lack of confidence in discussing condoms.423

The cost of teenage conceptions

Policy and public debate about teenage pregnancy has often had moral overtones in the
past, and it remains a controversial subject.424 Discussion of the ‘costs’ of teenage
pregnancy needs care, and needs to recognise that some teenage conceptions are planned,

Pregnancy and Social Disadvantage: Systematic Review Integrating Controlled Trials and Qualitative Studies’,
BMJ (Online), 339.7731 (2009), 1182–85; Suzanne Cater and Lester Coleman, Planned Teenage Pregnancy
(Joseph Rowntree Fundation, 2006); Dylan Kneale and others, ‘Distribution and Determinants of Risk of
Teenage Motherhood in Three British Longitudinal Studies: Implications for Targeted Prevention
Interventions’, Journal of Epidemiology and Community Health, 67.1 (2013), 48–55.
419
Katharine Ford and others, Adverse Childhood Experiences (ACEs) in Hertfordshire, Luton and
Northamptonshire (Centre for Public Health, Liverpool John Moores University, 2016).
420
Douglas B. Kirby, B. A. Laris, and Lori A. Rolleri, ‘Sex and HIV Education Programs: Their Impact on Sexual
Behaviors of Young People Throughout the World’, Journal of Adolescent Health, 40.3 (2007), 206–17.
421
Wendy Macdowall and others, ‘Associations between Source of Information about Sex and Sexual Health
Outcomes in Britain: Findings from the Third National Survey of Sexual Attitudes and Lifestyles (Natsal-3)’, BMJ
Open, 5.3 (2015).
422
Mark A Bellis and others, Contributions of Alcohol Use to Teenage Pregnancy- An Initial Examination of
Geographical and Evidence Based Associations (North West Public Health Observatory, 2009); Lester M.
Coleman and Suzanne M. Cater, ‘A Qualitative Study of the Relationship between Alcohol Consumption and
Risky Sex in Adolescents’, Archives of Sexual Behavior, 34.6 (2005), 649–61; Royal College of Physicians, Alcohol
and Sex : A Cocktail for Poor Sexual Health (RCP, 2011); Mari Imamura and others, ‘Factors Associated with
Teenage Pregnancy in the European Union Countries: A Systematic Review’, European Journal of Public Health,
17.6 (2007), 630–36.
423
Lesley Hoggart and Joan Phillips, ‘Teenage Pregnancies That End in Abortion: What Can They Tell Us about
Contraceptive Risk-Taking?’, Journal of Family Planning and Reproductive Health Care, 37.2 (2011), 97–102;
Sally Brown and Kate Guthrie, ‘Why Don’t Teenagers Use Contraception? A Qualitative Interview Study’,
European Journal of Contraception and Reproductive Health Care, 15.3 (2010), 197–204.
424
Lisa Arai, Teenage Pregnancy: The Making and Unmaking of a Problem (Policy Press, 2009).

146
and that in qualitative research some young parents report positive experiences and a
strong sense of achievement and identity in their new role.425
But for many young women, a teenage conception is unwanted. More than half of under-18
conceptions now end in abortion (over 60 per cent for under-16s).426 In 2019, 7.6 per cent
of abortions for 16- and 17-year-olds were for girls who were not having an abortion for the
first time.427
Research finds associations between teenage births and poorer outcomes for both mothers
and their children. This does not mean that the poorer outcomes are the direct result of the
mother’s age at birth. The high levels of disadvantage that many teenage mothers have
experienced are also part of the picture.428 With that caveat, some examples are given
below.

• Evidence worldwide finds that teenage pregnancy is associated with increased

obstetric risks.429 There is debate about how far poorer outcomes relate to age, and
how much to the other circumstances of teenage mothers’ lives. Research in England
in 2010 found that 14- 17 year old mothers had a significantly higher risk of pre-term
and very pre-term birth than adult women, even after adjusting for deprivation,
body mass index and ethnicity. 430
• Family conflict, vulnerability, lack of support, and poverty are all widely reported
amongst young mothers, as are poor mental health and high rates of post-natal
depression.431
• Several research studies in this country and the US have suggested that teenage
mothers are more likely to suffer certain forms of disadvantage after controlling for
the impact of background factors in childhood.432 433 One study summarised these
extra risks as arising both from negative material consequences of teenage
motherhood (e.g. greater risk of living in a workless family) and negative

425
Elizabeth McDermott and Hilary Graham, ‘Resilient Young Mothering: Social Inequalities, Late Modernity
and the “problem” of “Teenage” Motherhood’, Journal of Youth Studies, 8.1 (2005), 59–79; Cater and
Coleman.
426
Office for National Statistics, Conceptions in England and Wales, 2019
427
Department of Health & Social Care, Abortion Statistics, England and Wales: 2019 (ONS, 2020). Table 4b.
428
Alison Hadley, Teenage Pregnancy and Young Parenthood: Effective Policy and Practice (Routledge, 2018).
429
Fergus P. McCarthy, Una O’Brien, and Louise C. Kenny, ‘The Management of Teenage Pregnancy’, BMJ
(Online), 349.October (2014), 1–6.
430
Ali S. Khashan, Philip N. Baker, and Louise C. Kenny, ‘Preterm Birth and Reduced Birthweight in First and
Second Teenage Pregnancies: A Register-Based Cohort Study’, BMC Pregnancy and Childbirth, 10.36 (2010).
Sue Maguire and Luke Martinelli, The Next Chapter: Young People and Parenthood (Institute for Policy
431
Research, University of Bath, 2017); Vanessa Reid and Mikki Meadows-Oliver, ‘Postpartum Depression in
Adolescent Mothers: An Integrative Review of the Literature’, Journal of Pediatric Health Care, 21.5 (2007),
289–98.
432
Jason M. Fletcher and Barbara L. Wolfe, ‘Education and Labor Market Consequences of Teenage
Childbearing’, Journal of Human Resources, 44.2 (2009), 303–25; Arnaud Chevalier and Tarja K. Viitanen, ‘The
Long-Run Labour Market Consequences of Teenage Motherhood in Britain’, Journal of Population Economics,
16.2 (2003), 323–43.
433
John Hobcraft and Kathleen Kiernan, ‘Childhood Poverty, Early Motherhood and Adult Social Exclusion’,
British Journal of Sociology, 52.3 (2001), 495–517.

147
 consequences for emotional wellbeing (e.g. dissatisfaction with neighbourhood, and
lower levels of social support from family and friends).434
• Research into long term outcomes for children born in New Zealand in 1970 found
that, when followed up at age 15 and 21, children born to teenage mothers had
higher odds of having left school early, being unemployed, becoming parents early
themselves, and being a violent offender: maternal characteristics and the
circumstances of the family after the child's birth had independent and additive
effects.435
• Research in the UK using the Millennium Cohort examined cognitive scores at age 5,
finding that many differences between children of teenage mothers and those of
older mothers disappeared after adjusting for background factors, but a delay of
about five months in average verbal skills persisted.436

Policies and spending programmes

Labour: 1997–2010
When Labour came to office in 1997, the level of teenage pregnancy in the UK was a matter
of great public concern. While many other Western European countries had seen falling
teenage birth rates in the 1980s and 1990s, the UK had not. The Social Exclusion Unit was
commissioned to look in depth at teenage pregnancy in England.437 Their 1999 report
identified causes including deprivation and poor school experience, inadequate sex
education, and limited understanding of the challenges of parenthood. The report also
highlighted mixed messages in society, with one part of the adult world bombarding
teenagers with sexually explicit material, while parents and public institutions were ‘at best
embarrassed and at worst silent, hoping that if sex isn’t talked about, it won’t happen’. 438
The teenage pregnancy strategy
Following the report, the government put in place a dedicated England-wide ten-year
Teenage Pregnancy strategy, led by a national unit in the Department of Health. (In 2004
this was moved to the Department for Education and Skills to sit alongside other
programmes for young people.) Action included:

• a drive to improve access to contraception and good quality sex and relationships
education
• a national media campaign (‘Sex. Are you thinking about it enough’?) to encourage
young people to delay sex until they were ready and to use contraception when they
became sexually active

434
Ann Berrington and others, Consequences of Teenage Parenthood: Pathways Which Minimise the Long
Term Negative Impacts of Teenage Childbearing (University of Southhampton, 2005).
435
Sara Jaffee and others, ‘Why Are Children Born to Teen Mothers at Risk for Adverse Outcomes in Young
Adulthood? Results from a 20-Year Longitudinal Study’, Development and Psychopathology, 13.2 (2001), 377–
97.
436
Julia Morinis, Claire Carson, and Maria A. Quigley, ‘Effect of Teenage Motherhood on Cognitive Outcomes in
Children: A Population-Based Cohort Study’, Archives of Disease in Childhood, 98.12 (2013), 959–64.
437
As Director of the Social Exclusion Unit, I was involved in the development of this report.

438
Social Exclusion Unit, Teenage Pregnancy. (Stationery Office, 1999).

148
 • encouragement to parents to discuss sex and relationships with their children (‘Time
to talk’)
• preventive work with at-risk groups
• a special Sure Start Plus programme to support teenagers with a baby or toddler
with the issues they faced including health, housing, parenting, education,
employment and childcare
• centrally funded coordinators in every local authority to devise tailored local action,
and regional coordinators to share good practice and support national-local links. 439
Funding of £60 million was identified for the first three years of the strategy. A target was
set to halve the rate of under-18 conceptions by 2010. This strategy applied to England only
but similar policy initiatives were launched in the following years in Scotland and Wales. 440
Joined up action and wider polices.
A range of wider national policies discussed in other chapters of this report were relevant to
the issue of teenage pregnancy, including the increase in education resourcing, the focus on
reducing school absence and exclusions, support for staying on at school after 16 through
the Education Maintenance Allowance, the establishment of the Connexions Service to
support 13 - 19 year olds, and the range of policies designed to reduce young people’s
alcohol use.
Partnership working at local, regional and national level was an important element of the
teenage pregnancy strategy and local areas were expected to form partnership boards to
support this. Many areas established co-located services providing sex and relationships
education, contraceptive advice and support for young parents alongside other services
such as Connexions and young people’s drug and alcohol services.441
A sharper focus on contraception and sex-education
The Teenage Pregnancy Strategy was closely monitored, and a first evaluation was
conducted in 2005. This found an improvement in local joint working, and a modest increase
in the perceived quality of sex and relationships education. But contraception use had not
improved, young people were not getting sufficient access to reliable methods of
contraception such as long-acting reversible contraception (LARCs), and confusion remained
among young people about the confidentiality of contraception services. 442
The target to halve the rate of under-18 conceptions was maintained by the government.
The Department of Health focused more support and attention on local areas where
conception rates were not falling, emphasising the need for a whole systems approach

439
Alison Hadley, Roger Ingham, and Venkatraman Chandra-Mouli, ‘Implementing the United Kingdom’s Ten-
Year Teenage Pregnancy Strategy for England (1999-2010): How Was This Done and What Did It Achieve?’,
Reproductive Health, 13.1 (2016), 1–11.
440
In Wales, a Sexual Health strategy for Wales was launched in 2000 with the aims of reducing rates of
teenage conceptions, sexually transmitted infections and ensuring access to good quality sexual health advice.
The Scottish government published a strategy document in 2003, Enhancing sexual wellbeing in Scotland.
441
Hadley.
442
K Wellings and others, Teenage Pregnancy Strategy Evaluation - Final Report (London School of Hygiene and
Tropical Medicine, 2005).

149
rather than focusing narrowly on one service, or a few groups or areas. Ministers had face-
to-face meetings with local areas to sustain progress and identify next steps.443 The
Department also intensified efforts to improve access to contraception and promote
awareness and use of more reliable methods. The Department of Health published guidance
on how to make health services more welcoming to young people.444 By 2007, around 30
per cent of secondary schools and three quarters of FE colleges had an on-site health
service, providing advice on relationships and a range of sexual health services. 445 The
Department of Health secured £33 million additional funding in its budget settlement for
2008-11 to make further improvements in access to contraception.446
Take-up of more reliable contraception methods was promoted through National Institute
for Health and Care Excellence guidance in 2005 promoting the use of LARCs. 447 LARCs,
which include implants, inter-uterine devices and injections, are highly effective in
preventing pregnancy and do not depend on the user remembering to use them or
obtaining regular repeat prescriptions. Further guidance in 2007 drew attention to the case
for highlighting LARCs to vulnerable young people aged under 18. 448 From 2009/10 an
incentive scheme was introduced to encourage primary care physicians to inform women of
all ages about LARC options.449 These steps were bolstered in 2009 by an additional media
campaign aiming to raise use and awareness of contraception (‘Sex. Worth talking about’).
Sex and relationships education remained a controversial subject throughout the 2000s and
into the 2010s. A Review in 2008 reported that both Ofsted and several surveys of young
people were critical of provision and recommended that Personal Social and Health
Education be made part of the curriculum.450 The government proposed legislation to
require that all children receive at least one year of sex and relationship education, but this
was lost during the last stages of parliamentary debate before the 2010 election.451
Coalition and Conservatives: 2010 to 2019
After 2010, Coalition policy was to encourage local areas to maintain their efforts to reduce
teenage pregnancy further, as part of wider sexual health and public health priorities. There
was no dramatic change in policy, but a reduction in support and performance management
from government. National support teams ended, as did the national media campaigns, and
the Teenage Pregnancy Unit closed in 2012. A Teenage Pregnancy Advisor was appointed
to the newly established Public Health England, but the requirement for local partnership

443
Hadley.
444
Department of Health, You’re Welcome: Making Health Services Young People Friendly (DH, 2007).
445
Department for Children Schools and Families and Department of Health, Teenage Pregnancy Strategy :
Beyond 2010 (DCSF, 2010).
446
Hadley, Ingham, and Chandra-Mouli.
447
National Institute for Health and Care Excellence, Long-Acting Reversible Contraception (NICE, 2005).
448
National Institute for Health and Care Excellence, Sexually Transmitted Infections and Under-18
Conceptions: Prevention (NICE, 2007).
449
Richard Ma and others, ‘Impact of a Pay-for-Performance Scheme for Long-Acting Reversible Contraceptive
(LARC) Advice on Contraceptive Uptake and Abortion in British Primary Care : An Interrupted Time Series
Study’, PLoS Medicine, 17(9) (2020), 1–18.
450
External Steering Group, Review of Sex and Relationship Education in Schools (DCSF, 2008).
451
R. Long, Relationships and Sex Education in Schools (England) (House of Commons Library, 2019).

150
boards stopped. Responsibility for some sexual health services was transferred to local
authorities: local authority spending on sexual health increased in real terms between
2013/14 and 2014/15, but then fell back. By 2017/18 total sexual health services spending
by local authorities was 8.9 per cent lower in cash terms than in 2013/14 and 88 per cent of
authorities had decreased their sexual health spending.452 453
There have been multiple warnings that fragmentation of services, poor monitoring, and
inadequate funding were leading to declining access to LARCs. 454 In 2019 the government
recognised that the picture was ‘concerning’ and said the issue would be addressed in a
Sexual and Reproductive Health Strategy.455 In 2022 the government’s Women’s Health
Strategy said plans on sexual and reproductive health would be set out by the end of the
year, but this deadline was not met.456 There is evidence of continued problems with
access to LARCs: the 2021 statistics showed that provision had not yet returned to pre-
pandemic levels.457
In 2017, sex and relationships education was put on a statutory footing. The Children and
Social Work Act 2017 requires all English primary schools to teach age-appropriate
relationships education, and all secondary schools in England to teach age-appropriate
relationships and sex education. This extended the reach of sex education in several ways. It
meant that academies and free schools were now obliged to teach sex and relationship
education. And it limited the right for parents to withdraw their children from sex education
by giving children a right to opt in, three terms before they reach 16. New guidance on
content was also issued. 458
Many other changes to education and youth policies under the Coalition and Conservatives
were of relevance to young women at risk of teenage pregnancy: these are set out in more
detail in Chapters 1, 3, 5 and 7, covering respectively education, youth support, school
absence and adolescent alcohol use.

452
Tom Powell, Carl Baker, and Michael O’Donnell, Opposition Day Debate: Health and Local Public Health Cuts
(House of Commons Library, 2019).
Ruth Robertson, Sexual Health Services and the Importance of Prevention (King’s Fund, 2018).
453

House of Commons Health Committee, Public Health Post-2013 (HoC, 2016); David Buck, The English Local
454
Government Public Health Reforms: An Independent Assessment (King’s Fund, 2020); House of Commons
Health and Social Care Committee: Sexual Health (HoC, 2019); Royal College of General Practitioners, Sexual
and Reproductive Health: Time to Act (RCGP, 2017).
Department of Health and Social Care, Government Response to the Health and Social Care Select
455

Committee Report on Sexual Health (HMG, 2019).

456
Department of Health and Social Care, Women’s Health Strategy for England (DHSC, 2022).
457
Faculty of Sexual and Reproductive Healthcare, ‘Statement on Latest UKHSA and OHID Sexual and
Reproductive Health Profile Statistics’, 2023 <https://www.fsrh.org/news/fsrh-statement-on-latest-ukhsa-and-
ohid-sexual-and-reproductive/> [accessed 16 April 2023].
458
The guidance can be found here: https://www.gov.uk/government/publications/relationships-education-
relationships-and-sex-education-rse-and-health-education/relationships-and-sex-education-rse-secondary

151
Outcomes
Conceptions and births
Between 1998 and 2019 the conception rate for under-18s fell by 66 per cent. A sharp
acceleration in the rate of reduction can be seen after 2007. (Figure TP1) In the five years
from 2008 to 2013 the conception rate fell by more than 15 percentage points, then fell by
nearly 8 percentage points in the five years to 2018.
The birth rate fell by more as under-18s became increasingly likely to choose to have an
abortion if they became pregnant. The proportion of under-18 conceptions ending in
abortion rose between 1998 and 2007, then stabilised until 2012, then rose again. Over the
whole period the proportion of conceptions ending in abortion rose from just over 42 per to
just under 55 per cent. Overall, however, as Table TP1 earlier showed, the number of
abortions to girls under 18 more than halved.
Figure TP1: Trends in under-18 conception and birth rate: England

50
46.6
45
National target
41.4
to halve <18
40
conception rate
35 met in 2014
Rate per 1,000

30 26.9
25
20.5
20
15.7
15

10 7.1

Conception rate per 1,000 Maternity rate per 1,000

Source: ONS Conceptions Statistics England and Wales 2019, Table 6

Age
Reductions in conception rates occurred across all age groups, and all age groups show a
steep decline after 2007. (Figure TP2: England and Wales figures) The proportion of under-
18 conceptions that were to under-16s fell slightly.459 The improvement trend for the
youngest age groups appears to have stalled between 2017 and 2019.

459
Table TP2, England and Wales figures

152
Figure TP2: Conception rates per 1000 by mother’s age at conception: England and Wales, 1998 to 2019 (separate y-axis for each age group)

Under 14s 14 year olds 15 year olds 16 year olds 17 year olds
1.4 7 25 50 80

45
70
1.2 6
20 40
60
1.0 5 35
Rate per 1,000

50
15 30
0.8 4
25 40

0.6 3
10 20
30

0.4 15
2
20
5 10
0.2 1 10
5

0.0 0 0 0 0
1998
2001
2004
2007
2010
2013
2016
2019

1998
2001
2004
2007
2010
2013
2016
2019

1998
2001
2004
2007
2010
2013
2016
2019
2013
1998
2001
2004
2007
2010
2013
2016
2019

1998
2001
2004
2007
2010

2016
2019
Source: ONS Conceptions Statistics England and Wales 2019, Table 1b

153
Inequalities
The deprivation gap
Teenage conceptions have fallen in both deprived and non-deprived areas, but the sharp fall
in deprived areas after 2007 initiated a marked narrowing of the deprivation gap. This fell
from just under 29 per percentage points in 2007 to below 15 points in 2013. Between
2013 and 2019 the gap fell slightly further. (Figure TP 3). Despite the
narrowing of the gap over time, significant inequalities remained between areas, and within
them, with the majority of councils having at least one ward with a very high teenage
conception rate.460
Figure TP3: Trends in under-18 conception rates in the most and least deprived local
authorities in England

70
63.4

60

50
Rate per 1,000 women

28.9
ppt
40 gap
31.6
30
14.6 22.4
ppt
20 gap
10.2
10

Most deprived decile Least deprived decile

Source: ONS Conceptions statistics, and author's calculations using Index of Multiple deprivation

Regions
Regional gaps in teenage conception rates narrowed substantially from the early 2000s
onwards, initially mainly driven by the steep fall in London. Indeed, the Inner London
trajectory is very striking. Although the sharp downward trend in the national under-18
conception rates began in 2008, the rate in inner London began to fall much earlier, from
2002, and fell further. London teenage conception rates fell 79 per cent between 1998 and
2019 (and had halved by 2011). Figure TP4 shows that, while Inner London had the highest
under-18 conception rate from 1997 to 2007, it was replaced by the North East after 2008.

460
Local Government Association and Public Health England, Good Progress but More to Do: Teenage
Pregnancy and Young Parents (LGA, 2018).

154
Figure TP4: Trends in under-18 conception rates by English region

80

Inner London
70

60

50
Rate per 1,000

40

30

20

10

NE NW Yrks H East Mids

West Mids East Inner London Outer London
SW SE

Source: ONS Conceptions Statistics England and Wales 2019, Table 6

Ethnicity
The published conception statistics do not record ethnicity. Other data suggests that
relativities appear to have changed over time. In the 1980s and 1990s it was the case that
Caribbean, Pakistani and Bangladeshi women were more likely to have been teenage
mothers than white women, but Indian women were less likely.461 Teenage Pregnancy Unit
analysis in 2006 found that all Asian ethnic groups had a lower than average incidence of
teenage motherhood and that the ethnic groups with teenage motherhood rates above
average were now young women of ‘Mixed White and Black Caribbean’, ‘Other Black’, ‘Black
Caribbean’ and ‘White British’ ethnicity.462 More recent research (based on small numbers)
finds a similar pattern.463

461
R. Berthoud, ‘Teenage Births to Ethnic Minority Women.’, Population Trends, 104 (2001), 12–17.
462
Department for Children Schools and Familes and Department of Health, Teenage Parents Next Steps:
Guidance for Local Authorities and Primary Care Trusts (DCSF, 2007).
463
Crawford, Cribb, and Kelly.

155
Post-pandemic data
The main focus of this report is on the period ending 2019. However data are now available
on under-18 conceptions for the two following years, both of which were affected by the
pandemic. They show that the under-18 conception rate fell from 15.8 conceptions per
thousand in 2019 to 13.1 per thousand in 2020. There was then a tiny rise in 2021 to 13.2
per thousand.464

International comparisons
Conceptions: Scotland and Wales
Scotland and Wales have also seen large falls in under-18 conception rates over the last two
decades. The gaps between the three nations have narrowed since 1998. (Figure TP5) Wales
still has the highest rates, but by less, and England no longer exceeds Scotland’s rate.
England and Scotland saw a steep decline after 2007, and Wales after 2008.
Figure TP5: Trends in under-18 conception rates in England, Scotland and Wales

60

50
Rate per 1,000 women

40

30

20

10

England Wales Scotland

Source: ONS for England and Wales, Scottish Teenage Pregnancy statistics

Teenage births: other OECD countries

Broader international comparisons are not available for teenage conception rates. But
there is data for births to 15–19-year-olds, shown in Figure TP6. This shows, first, the UK in
comparison with other high-income countries which also had high teenage birth rates in
1998 - the United States, the UK, New Zealand, Canada, and Australia. The 15-19 birth rate
has fallen in all these countries: but the patterns differ. The fall was largest in the US,
followed by the UK.465 After 2002, Canada saw its reduction trend slow, while the trend in

464
Office for National Statistics, Conceptions in England and Wales, 2021 (ONS, 2023).
465
For more background on US experience see Rachel H. Scott, Kaye Wellings, and Laura Lindberg, ‘Adolescent
Sexual Activity, Contraceptive Use, and Pregnancy in Britain and the U.S.: A Multidecade Comparison’, Journal

156
Australia plateaued, and in New Zealand birth rates started to rise again. This Figure also
shows the UK alongside other Western European countries: the gap has narrowed, but even
so, the UK still has far higher teenage birth rates than its European neighbours.

Figure TP6: Trends in live births to 15–19-year-olds: international comparisons

UK and other high income countries UK and other European countries
with high rates
60
60

50
50

40 40
Rate per 1,000 women
Pate per 1,000 women

30 30

20 20

10 10

0 0

United States United Kingdom United Kingdom Netherlands

New Zealand Australia Germany France

Canada Italy

Source: United Nations Population Division, World Population Prospects

of Adolescent Health, 66.5 (2020), 582–88; Claire D Brindis and others, ‘Perspectives on Adolescent Pregnancy
Prevention Strategies in the United States: Looking Back, Looking Forward’, Adolescent Health, Medicine and
Therapeutics, Volume 11 (2020), 135–45.

157
Plausible explanations of the trends in England
Since the scale of the fall in teenage conceptions in England began to emerge, there have
been several attempts to understand the reasons behind it. Researchers differ in the
emphasis given to sex and relationships education and access to contraception, on the one
hand, and wider social drivers such as rising educational participation and changes in
adolescent lifestyles. Key studies are summarised below.

Previous research about the fall in teenage conceptions

Education and contraception: Kaye Wellings and colleagues drew on conception data and
survey results to argue for an explanation involving both rising educational attainment
and increased use of effective contraception. They found that over a decade young
women became slightly more likely to use the pill or condoms at first sex and to have got
their sex education mainly at school, and there were signs of an increase in use of long-
acting reversible contraception (LARCs).466

Long-term multi-component strategy: commenting on the above study, Skinner and

Marino noted the convincing case that much of the reduction in teenage conception
could be attributed to the teenage pregnancy strategy, which they saw as ‘an impressive
example of how a sustained, multilevel, and multicomponent intervention could impact a
complex health and social issue with high cost-effectiveness’.467

Contraception: A 2020 study covering England, Scotland and Wales found evidence of
rising rates of LARC prescriptions for teenagers between 2004/05 and 2013/14 with a step
change after 2009/10 when GP contracts included a new target to offer women advice
about LARC methods.468

Education, ethnicity, and reduction in risk behaviours: In research based on data up to

2012, Girma and Paton argued for the importance of improved educational achievement
in driving falling teenage births. They also highlighted the impact of an increasing non-
white population, and reduced risk-taking behaviours, including drug and alcohol use.469

Social drivers: A 2020 study found that larger declines in local under-18 conception rates
between 1998 and 2017 were related to social drivers including growing Black or South
Asian teenage populations, less youth unemployment, and more educational attainment,
but concluded that the analysis could not explain all the reductions within local areas. 470

466
K. Wellings, M.J. Palmer, and R.S. Geary, ‘Changes in Conceptions in Women Younger than 18 Years and the
Circumstances of Young Mothers in England in 2000–12’, The Lancet, 388.10044 (2016), 586–595.
467
S. Rachel Skinner and Jennifer L. Marino, ‘England’s Teenage Pregnancy Strategy: A Hard-Won Success’, The
Lancet, 388.10044 (2016), 538–40.
468
Richard Ma and others, ‘Impact of a Pay-for-Performance Scheme for Long-Acting Reversible Contraceptive:
Advice on Contraceptive Uptake and Abortion in British Primary Care : An Interrupted Time Series Study’, PLoS
Medicine, 17(9) (2020), 1–18 .
469
Sourafel Girma and David Paton, ‘Is Education the Best Contraception: The Case of Teenage Pregnancy in
England?’, Social Science and Medicine, 131 (2015).
470
Katie L. Heap, Ann Berrington, and Roger Ingham, ‘Understanding the Decline in Under-18 Conception Rates
throughout England’s Local Authorities between 1998 and 2017’, Health and Place, 66 (2020).

158
 Changing attitudes: Few of these studies looked at attitudinal factors. However, a 2018
study by O’Brien, based on surveys and focus groups, found that recent generations of
young people were firmly focused on education and careers and the need to work hard at
school, often had negative views of both alcohol and promiscuity, enjoyed spending social
time with their family, and often socialised with friends more on the same screen than in
the same room. 471

None of the studies above considers all the potential risk factors and circumstances known
to contribute to teenage contraception. The remainder of this chapter will draw on outcome
data across several fields to present as rounded a picture as possible.
The biology: less sex, better contraception, or both?
The first question is whether there have been changes in the level of sexual activity and/or
contraception use amongst under-18s. Logically, unless there has been a change in the
ability of teenage girls to conceive, lower conception rates must reflect some combination
of later or less sexual activity by teenagers, and/or increased use of effective contraception.
There is evidence that both of these have occurred in the last two decades.
Rising age of first intercourse
For some time, researchers doubted whether there had been a reduction in the proportion
of teenagers who have had intercourse at young ages in England, this view being based on
data gathered in 2012 (from women aged 16 to 19 at the time) in the last National Survey of
Sexual Attitudes and Lifestyle (NATSAL).472 But evidence of a significant behaviour change
has started to emerge from other data sources.473
One important source is the World Health Organisation’s multi-country survey of Health
Behaviour in School-aged Children (HBSC).474 This has for some time asked 15-year-olds in
participating countries whether they have had intercourse. This data shows a striking fall in
England between 2002 and 2018. The proportion of 15-year-old girls reporting they had had
sex declined by more than half from 40 per cent to 18 per cent. These figures are illustrated
in Figure TP7 overleaf.475 The international average over the same period fell from 20 to 14
per cent: thus, England remained above average, but no longer such an outlier.
This trend is corroborated by evidence from cohort studies with different generations. The
LSYPE cohort study of young people born in 1989 and 1990 in England found that 10 per
cent of girls in the study had had sexual intercourse at age 14 or under. For girls in the

471
Katherine O’Brien, Social Media, SRE, and Sensible Drinking: Understanding the Dramatic Decline in Teenage
Pregnancy (British Pregnancy Advisory Service, 2018).
472
Scott, Wellings, and Lindberg.
473
It is disappointing that data on these matters is not collected more frequently in the UK: in the US it is
collected every two years, through the Centre for Disease Control and Prevention’s Youth Risk Behavior
Survey.https://www.cdc.gov/healthyyouth/data/yrbs/yrbs_data_summary_and_trends.htm
474
Fiona Brooks and others, HBSC England National Report: Findings from the 2018 HBSC Study for England
(University of Hertfordshire, 2020).
475
The last two cohorts are younger than those surveyed in the last wave of NATSAL.

159
Millennium Cohort Study, a UK study of children born ten years later, the corresponding
figure was much lower, at 2.1 per cent.476
Figure TP7: Trends in proportion of 15-year-old female pupils in England who have had
sexual intercourse by age 15

45
40
40
34
35
31
30
24
25
Per cent

20 18

15

10

0
2002 2006 2010 2014 2018

Source: Brooks et al, HBSC England National Report: Findings from the 2018 study

Increased use of more effective contraception

There is very strong evidence that the last two decades have also seen a trend for teenagers
to use more effective forms of contraception:

• According to HBSC survey data, the proportion of sexually active fifteen-year-old

girls in England using the pill at last intercourse rose from 23 to 33 per cent between
2006 and 2010 and has stayed above 30 per cent since then. (Figure TP8)

• Data from contraception services tells us that the proportion using the most
effective form of contraception, LARCs, has increased significantly. The statistics
from dedicated sexual health clinics in England show that the proportion of teenage
clients using LARCs grew from 10 per cent of 16–17-year-olds in 2006/07 to 34 per
cent in 2018/19: this trend is illustrated below. (Figure TP9)

• Primary care also saw substantial growth in prescriptions of LARCs to young people
between 2004/05 to 2013/14.477

476Department for Education, Youth Cohort Study & Longitudinal Study of Young People in England : The
Activities and Experiences of 18 Year Olds : England 2009 (DfE, 2010) Table 7.3.1; E Fitzsimons and others,
Determinants of Risky Behaviour in Adolescence: Evidence from the UK. (UCL Centre for Longitudinal Studies,
2018) Figure 39.
477
Ma and others.
160
Figure TP8: Trends in 15-year-old girls in England: use of pill at last intercourse

40
35
35 33 33

30

25 23
Per cent

20

15

10

0
2002 2006 2010 2014 2018

Source: HBSC international surveys, 2018 and previous years

Figure TP9: Trends in under-17s attending contraceptive clinics in England using Long-
Acting Reversible Contraceptives

40

35

30

25
Per cent

20

15

10

Under 16 16-17

Source: NHS, Sexual and Reproductive Health Services Statistics on attendances. First attendances for data
until 2013/14. Thereafter, main method.

161
Social factors: the wider drivers of behaviour change
These behavioural changes are significant but what drivers lie behind them? To understand
this, we should explore wider social changes, particularly in those factors which we know
are associated with teenage conception.
Improving sex and relationships education and information
Improvements in sex and relationships education are likely to have played some role. Good
sex education, received in school, is associated with later onset of sexual activity, and
increased contraceptive use. The earlier part of this chapter described the efforts to
improve sex and relationships education, which went hand in hand with media campaigns
encouraging young people not to feel pressured into having sex before they felt ready.
There was a significant increase in the proportion of young people receiving most of their
sex and relationships education at school in the decade leading up to 2010. Research with
16- to 24-year-olds between 2010 and 2012, found that 41.3 per cent of young women had
received their sex and relationships education mainly at school, up from below 30 per cent a
decade earlier. However, most young people still reported needing more information on a
broad range of topics. 478
Girls’ increasing engagement and success in education
Educational disengagement is closely associated with teenage conceptions. As discussed in
more detail in Chapters 1 and 5, the Labour government devoted time and attention to
improving secondary school attendance and attainment and there were sharp
improvements in outcomes. Overall absence amongst girls in secondary schools improved
from 8.4 per cent of all schooling in 2005/06 to 5.3 per cent in 2013/14. Between 2005 and
2015 the proportion of girls not obtaining level 2 qualifications by the age of 19 more than
halved. And between 1998 and 2018 the proportion of girls not in full-time education at age
17 more than halved, from just over 38 per cent to just over 17 per cent. These are changes
of a similar magnitude to the fall in teenage conceptions and took effect over a similar
period. These changes to girls’ educational performance and aspiration may well have
driven a more cautious approach to sex, relationships and contraception, as well as a rising
tendency for young women who did become pregnant to choose to have an abortion.
Declining alcohol use
Adolescent alcohol use is another factor associated with teenage conception. The declines
in teenage drinking described in Chapter 7 are therefore of interest. As that chapter sets
out, between 2003 and 2014, experience of alcohol amongst 13–15-year-olds more than
halved, and for 11-12 year-olds it fell by over two thirds. The proportion of 14- and 15-year-
olds who had drunk alcohol in the last week fell by two thirds. The proportion of girls who
had been drunk at least twice fell significantly between 2002 and 2014, from 27 per cent of
13 year-old girls to just six per cent, and the rate of under-18 hospital admissions for alcohol
amongst girls more than halved over the decade from 2006/07. Again, these are reductions

478
Clare Tanton and others, ‘Patterns and Trends in Sources of Information about Sex among Young People in
Britain: Evidence from Three National Surveys of Sexual Attitudes and Lifestyles’, BMJ Open, 5.3 (2015), 1–10.

162
of similar magnitude to the fall in teenage conceptions, in indicators closely associated with
early sexual intercourse and unprotected sex.
Ethnicity
It has been suggested that changing ethnic composition of the youth population may have
played a role in declining teenage conception.479 Sadly teenage conception data does not
cover ethnicity. As noted above, there is some evidence that girls from South Asian and
Black African backgrounds have lower teenage birth rates. This could be related to
differential social attitudes, educational factors, and attitudes to alcohol: however, as
discussed in Chapter 7, changing ethnic composition alone does not explain the scale of the
change in adolescent drinking at the population level.
Has growing use of mobile phones and social media played a role?
It has been suggested by some that increased use of mobile phone and online
communication has played the major role in reducing teenage conceptions.480 The growing
use of online communication is an important trend, but it does not correlate well with the
trends in teenage conceptions. Reductions in teenage conception started (at least in
London) before access to mobile phone and social media use was widespread. Reductions
were greatest in deprived areas, where young people were unlikely to have the resources to
be early adopters of new technology. Research into those who use social media most does
not suggest that they behave more cautiously as a result: for example, research into alcohol
use has found that young people who use social media most appear more likely to drink
alcohol than others.481

Discussion
Overall, therefore, this analysis suggests that the main drivers of falling teenage conceptions
were increased access to, and use of, effective contraception, combined with a trend
towards later first intercourse. These behavioural changes coincided with, and may have
been driven by, changes of similar magnitude in girls’ educational engagement and
attainment, reductions in alcohol use, and improved sex and relationships education. Policy
efforts in these areas may therefore have been mutually reinforcing.
To some extent, trends in teenage pregnancy resemble the other subjects covered in this
report, in that sustained policy attention and investment were followed by a period of
improving outcomes. But unlike many of those other subjects, although the dramatic
reduction of the earlier period slowed somewhat after 2013, the overall improving trend did
not plateau or reverse. It is not possible to say confidently why performance on this
indicator of disadvantage has proved more resilient. It could reflect an enduring change in

479
Heap, Berrington, and Ingham.
480
David Paton, ‘The Mysterious Fall of the Teenage Pregnancy Rate’, The Spectator, 2020.
481
Amy Pennay and others, ‘Researching the Decline in Adolescent Drinking: The Need for a Global and
Generational Approach’, Drug and Alcohol Review, 37.January (2018), S115–19; Hilde Pape, Ingeborg Rossow,
and Geir Scott Brunborg, ‘Adolescents Drink Less: How, Who and Why? A Review of the Recent Research
Literature’, Drug and Alcohol Review, 37.February (2018), S98–114.

163
contraception methods, a long-term change in the perceived acceptability or desirability of
young childbearing, or long-term trends in educational participation.
The falling rate of teenage pregnancy in England is widely seen as a success, and the fact
that it has continued for so long may have led to the belief that we are witnessing an
irreversible social trend that will simply keep going. But the data presented in this report
suggest a need for caution. The data shows us that before the pandemic, the pace of
reduction in conceptions showed some signs of slowing, and progress stalled for girls of 14
and under. Contraceptive access also appears to have become increasingly problematic. And
trends in some of the social risk factors for teenage conception (such as school absence)
have worsened either before or since the pandemic. For the next few years, the picture will
be clouded by the impact of the pandemic, the lockdowns and other restrictions associated
with it. But it would be unwise to assume that teenage pregnancy will go on falling in future
simply because it fell in the past.
So, despite past progress, there is no basis for complacency on this topic. Teenage parents
remain a very disadvantaged group with many vulnerabilities. With teenage birth rates still
well above those of most European countries, persisting inequalities in teenage pregnancy
rates between areas, and a worrying rise in some of the drivers of teenage conceptions,
there is a strong case for a continuing focus on teenage pregnancy, to ensure that young
people have the services and support they need to avoid unwanted conceptions.

164
Chapter 7 Adolescent Alcohol Use
Introduction
Alcohol misuse is a leading cause of ill health and preventable death, associated with many
adverse social consequences, and drinking alcohol in childhood and adolescence is
especially damaging. This is a social problem which in the past was particularly prevalent in
the UK. In 2002, the European school survey placed England in the top three countries for
high levels of adolescent drinking. But between 2002 and 2014, the falls in drinking by both
boys and girls were larger in England than in any comparator country, and England moved
to a better position in the league table 482 After 2014 progress appears to have stalled, and
hazardous drinking among young people remains an issue: in 2018, a third of 15 year olds
who had drunk in the last week said they had drunk more than 10 units that week. 483
This chapter looks at the story behind these headline statistics. It first sets out some key
background facts on adolescent drinking - the scale, who is most affected, and the cost of
alcohol misuse. It then summarises the policy approaches under Labour, Coalition, and
Conservative governments, and explores the trends in adolescent drinking in more detail.
Finally, it discusses possible causes for the striking patterns that have been seen.

Key data: alcohol use by young people in 2018/19

In 2019, just before the pandemic, nearly one in ten 11- to 12-year-olds in England had
already tried alcohol. As of 2018, a quarter of 15-year-olds had already been drunk on at
least two occasions. Around 14 per cent of 15-year-olds drank at least once a week. Eight
per cent of 13-year-old girls had been drunk twice. These figures – and the significant
decline they represent compared with earlier levels of teenage drinking - are set out in
more detail later in this chapter. (Figure AD1, Figure AD2 and Figure AD3).

Groups over-represented in adolescent alcohol use

Drinking is more common amongst young people with certain characteristics. To start with
family and peer group factors, a study by Yap and others in 2017 identified three risk factors
that predicted alcohol initiation and levels of later alcohol use/misuse - parental drinking,
favourable parental attitudes towards alcohol use, and parental provision of alcohol. The
same study found four protective factors - parental monitoring, parent–child relationship
quality, parental support, and parental involvement. 484 Analysis of the Millennium Cohort
found that having friends who drank was strongly associated with drinking at age 11. 485

482
Jo Inchley, Dorothy Currie, Alessio Vieno, and others, Adolescent Alcohol-Related Behaviours: Trends and
Inequalities in the WHO European Region, 2002-2014. (WHO Regional Office for Europe, 2018).
483
Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
England, 2021 (NHS Digital, 2022). Table 5.14. This series will hereafter be referred to as SDDYP with the
relevant year.
484
Marie B.H. Yap and others, ‘Modifiable Parenting Factors Associated with Adolescent Alcohol Misuse: A
Systematic Review and Meta-Analysis of Longitudinal Studies’, Addiction, 112.7 (2017), 1142–62.
485
Yvonne Kelly and others, ‘What Influences 11-Year-Olds to Drink? Findings from the Millennium Cohort
Study’, BMC Public Health, 16.1 (2016), 1–8.

165
School surveys consistently find that drinking levels are higher amongst respondents with
experience of school absence, school exclusion, and using drugs. 486 A 2010 analysis of the
first LSYPE cohort found that playing truant from school predicted trying alcohol amongst
14- and 15-year-olds who had not already had their first drink.487
Young people from low-income backgrounds are in general less likely than higher income
young people to have tried alcohol, and less likely to have drunk in the last week.488
Nonetheless, as of 2019 some areas in the most deprived decile (Hull, Blackpool, and
Liverpool) were in the top twenty for alcohol-related hospital admissions of under-18s.489

The cost of adolescent drinking

There is a large body of evidence on the harmful short-term and longer-term effects of
drinking in adolescence. Drinking damages brain development, and younger age at first use
is associated with increased risk of heavy alcohol use in later life.490 Regular or heavy
adolescent drinking also increases risks of drug use, unprotected sex, school absence, and
offending.491 Analysis of the first LSYPE Cohort found that at ages 14-15 young people who
drank on most days had over 4 times the odds compared to other young people of
increasing their truancy, and at ages 15-16 this had increased to over 10 times the odds.492
Follow-up of the 1970 birth cohort at age 30 found that, controlling for potential
confounding factors, binge drinkers at 16 went on to have an increased risk of school
exclusion, lack of qualifications, illicit drug use, homelessness, criminal convictions,
accidents, adult alcohol dependence, regular excess drinking, and lower adult social class.493
Hazardous drinkers affect not just themselves but also people who are close to them, or
who come in contact with them, and they impose major costs on public services such as the
NHS and the police. The role of alcohol in violent crime is particularly striking: the Crime
Survey for England and Wales regularly finds that a large percentage of victims of violent
crime believe the perpetrator was under the influence of alcohol at the time of the offence
(53 per cent of victims thought this as of 2013/14).494 The government has not revisited its
estimate of the total social and public service costs of alcohol abuse for over a decade: the
last figure it quoted was £21 billion a year, in 2012.495

486
SDDYP 2018. Tables 5.26 and 7.16. John Marsden and others, ‘Personal and Social Correlates of Alcohol
Consumption among Mid-Adolescents’, British Journal of Developmental Psychology, 23.3 (2005), 427–50.
487
Green and Ross.
488
SDDYP 2018 chapter 7 and tables 7.23 and 7.24 (area, no past trend).
489
Public Health England: https://fingertips.phe.org.uk/profile/child-health-profiles
490
Ralph W. Hingson, Timothy Heeren, and Michael R. Winter, ‘Age at Drinking Onset and Alcohol
Dependence’, Archives of Pediatrics and Adolescent Medicine, 160.7 (2006), 739–46; Wenbin Liang and Tanya
Chikritzhs, ‘Age at First Use of Alcohol Predicts the Risk of Heavy Alcohol Use in Early Adulthood: A
Longitudinal Study in the United States’, International Journal of Drug Policy, 26.2 (2015), 131–34. Donaldson.
491
Bellis and others; Coleman and Cater; Royal College of Physicians; Ipsos MORI, Teenage Drinking and the
Role of Parents and Guardians: Findings from Drinkaware Monitor 2016 (Drinkaware, 2016); Green and Ross.
492
Green and Ross.
493
Russell M. Viner and B. Taylor, ‘Adult Outcomes of Binge Drinking in Adolescence: Findings from a UK
National Birth Cohort’, Journal of Epidemiology and Community Health, 61.10 (2007), 902–7.
494
Office for National Statistics, Violent Crime and Sexual Offences: Alcohol-Related Violence (ONS, 2015).
495
Home Office, The UK Government Alcohol Strategy (London: HMG, 2012).

166
Policies and spending programmes
Labour: 1997–2010
The Labour government’s early years in government sent mixed messages on alcohol. The
policy on alcohol taxation policy was described as seeking to deliver ‘a fairer balance in the
burden of taxation falling on different alcoholic drinks and different types of drink-
producers’. This meant, for example, that the duty on spirits was frozen in successive years,
but the taxation of spirit-based coolers (alcopops) was increased.496 A major overhaul of
licensing laws, which included the removal of 11.00 pm closing time for pubs, was legislated
in the Licensing Act 2003 and implemented in 2005. As the start date of this policy
approached, there was growing public concern about alcohol use, with extensive media
coverage highlighting the extent of binge drinking in town centres.
A 2003 report from the Prime Minister’s Strategy Unit, followed by a 2004 Government
Alcohol Harm Reduction Strategy, signalled a period of progressive tightening of policies on
adolescent drinking.497 An early focus was underage selling: surveys of young people
showed that underage purchasing was widespread, but prosecutions were rare. 498 The
government legislated for several new powers to tackle this, including making it possible for
the authorities to carry out test purchase operations involving young people, and making
enforcement easier by including underage sales within the new sanction of Fixed Penalty
Notices. From 2004 the Home Office funded police and local authorities to run ‘alcohol
misuse enforcement campaigns’ which encouraged concerted enforcement action on
underage sales and alcohol-related disorder.499 The frequency of enforcement action
increased. In 2000, there had been just 130 prosecutions for selling to under-18s.500 This
rose to 1,000 prosecutions in both 2005 and 2006, alongside thousands of fines of
individuals for selling to minors.501 The increased chance of sanctions brought a response
from industry, who created schemes such as Challenge 21 and Challenge 25 to require
alcohol purchasers to prove their age on request. Home Office data found that the
percentage of test purchasers who succeeded in buying alcohol underage fell from 50 per
cent to 15 per cent between 2004 and 2007.502
The government addressed teenage drinking through other initiatives. Treatment for young
people with an alcohol problem increased (see discussion of the expansion in substance
misuse treatment in Chapter 8). Young people with drink problems received support

496
HM Treasury, Budget 2002 - The Strength to Make Long-Term Decisions: Investing in an Enterprising, Fairer
Britain (HMT, 2002).
497
Prime Minister’s Strategy Unit, Alcohol Misuse: How Much Does It Cost? (Cabinet Office, 2003).
498
Prime Minister’s Strategy Unit, The Alcohol Harm Reduction Strategy for England (Cabinet Office, 2004).
499
Home Office, Saving Lives. Reducing Harm. Protecting the Public; Mark A Bellis, Zara Anderson, and Karen
Hughes, Effects of the Alcohol Misuse Enforcement Campaigns and the Licensing Act 2003 on Violence
(Liverpool: JMU Centre for Public Health, 2006); Home Office, Police Powers to Close Premises under the
Licensing Act 2003 (HO, 2011). As Home Office Director General of Crime and Policing between 2005 and
2008, I had oversight of this work.
500
Prime Minister’s Strategy Unit, The Alcohol Harm Reduction Strategy for England.
501
Home Office, FOI Release: Premises Prosecuted for Selling Alcohol to Underage Drinkers, 2008; Ministry of
Justice, Criminal Justice Statistics Quarterly: December 2013: Out of Court Disposals (MoJ, 2013).
502
National Audit Office, Reducing Alcohol Harm: Health Services in England for Alcohol Misuse (NAO, 2008).
167
through broader youth prevention work including through Connexions.503 And the
government introduced several sets of powers to tackle public drinking because of its links
with anti-social behaviour. The police gained powers to stop young people drinking alcohol
in public places and to confiscate alcohol from a young person on the street, and local
authorities acquired the power to designate public areas where drinking could be
stopped.504 The Policing and Crime Act 2009 brought into force a new offence of persistent
possession of alcohol in a public place by a person under 18. 505
Information was seen as a key part of alcohol strategy. The government-funded ‘Know Your
Limits’ campaign challenged irresponsible drinking. One 2008 advert showed young people
injuring themselves, urinating on their shoes, and getting vomit in their hair, with the slogan
‘You wouldn't start a night like this, so why end it that way’. 506 And in 2009, Sir Liam
Donaldson, the UK’s Chief Medical Officer, issued very clear medical guidance for parents on
children’s consumption of alcohol. The key points were that children should not drink
alcohol until at least 15, and that consumption thereafter should be supervised and no more
than one day a week. 507 The new guidance was widely covered in broadcast and print
media, with headlines from 17 December 2009 including: ‘Parents get booze alert over
liberal attitude to drink’ (Daily Mirror); ‘Health Chief warns the wine-weaning parents: you
could be lining children up for a life of addiction’ (Daily Mail); and ‘No alcohol for under-15s,
Liam Donaldson tells parents’ (Guardian).
Although many of these measures were welcomed by experts, many criticised the failure to
use tax as an instrument to restrain the rise of alcohol consumption.508 Labour finally
changed its position on alcohol tax in the 2008 Budget, when Chancellor Alistair Darling
increased alcohol duties by 6 per cent above simple indexation and announced that duty
rates would rise by 2 per cent more than inflation in future years as well, a policy known as
the ‘alcohol duty escalator’.509 This policy raised £1.5 billion over its first three years
compared with simple indexation, and contributed to a four-year reduction in alcohol
affordability – the first time in two decades that the trend of increasing affordability was
interrupted.510 However, when Sir Liam Donaldson used his 2009 annual report to argue
the case for a minimum unit price for alcohol, the Prime Minister himself was quick to damp
down speculation that the government would take it forward. Donaldson cited research
suggesting that a minimum price of 50 pence per unit of alcohol would decrease high-risk
drinkers’ consumption by over 10 per cent but that low-risk drinkers’ consumption would

503
Hoggarth, Smith, and Britain; Sheehy, Kumrai, and Woodhead.
504
Confiscation of Alcohol (Young Persons) Act 1997 and Criminal Justice and Police Act 2001
505
Home Office, Giving Directions to Individuals to Leave a Locality: Section 27 of the Violent Crime Reduction
Act 2006 (HO, 2010).
506
Mark Sweeney, ‘Home Office Booze Ads Show Dark Side of Getting Wrecked’, Guardian, 17 June 2008.
507
Liam Donaldson, Guidance on the Consumption of Alcohol by Children and Young People. (Department of
Health, 2009).
508
Martin Plant, ‘The Alcohol Harm Reduction Strategy for England’, BMJ (Clinical Research Ed.), 328.7445
(2004), 905–6.
509
HM Treasury, Budget 2008: Stability and Opportunity: Building a Strong, Sustainable Future (HMT, 2008).
510
NHS Digital, Statistics on Alcohol, England, Table 4

168
fall by less than 4 per cent. The Prime Minister's concern was the impact on moderate
drinkers, who would feel penalised. 511
Coalition and Conservative governments: 2010 to 2019
For a short period, the Coalition continued the policy of raising alcohol taxes and in 2012
signalled an intention to introduce a minimum unit price. 512 But the 2013 Budget brought
to an end the alcohol duty escalator by cutting beer duties, and in July 2013 the government
dropped its plan for a minimum unit price and substituted a ban on sales below cost price,
which would have a much smaller impact.513
Real reductions in duty rates continued through the rest of the Coalition period and since.
For the remainder of the period to 2019, all Coalition and Conservative budgets included
duty freezes or cuts with the exception of 2017, when Philip Hammond was Chancellor. The
alcohol duty indexation included in the March 2023 budget is therefore the first in a very
long time.514
The fiscal costs of these alcohol tax reductions have been striking. Table AD1 shows the
costs published in Budget documentation each year. Although these published costs are
shown only for five years, their impact does not stop after five years: the base is
permanently lower than it would have been if rates had been indexed every year. From the
table we can see that the four budgets from 2013 to 2016 made the public finances poorer
by a total £790 million in 2017/18. That cost endures into later years (unless duties are
indexed by more than inflation in a future budget). This fiscal cost suggests a notable
degree of priority for a policy that is essentially health-damaging. By contrast, public
funding for drug and alcohol treatment fell in real terms every year between 2014/15 and
2018/19, with spending on young people’s drug and alcohol services 29 per cent lower in
real terms by the end of the period.515
The Coalition promoted a partnership approach with the alcohol industry with the aim of
encouraging more responsible drinking.516 But a 2015 study found that the ‘industry
responsibility deals’ agreed were unlikely to have contributed significantly to reducing
alcohol consumption, with most industry participants committing to actions they would
have undertaken anyway. 517 A 2019 study of alcohol labelling found that, more than three

511
Donaldson and Liam, ‘Passive Drinking : The Collateral Damage’, in Chief Medical Officer’s Annual Report
(Department of Health, 2009); David Hencke and Andrew Sparrow, ‘Gordon Brown Rejects Call to Set
Minimum Prices for Alcohol’, The Guardian, 16 March 2009.
512
Home Office, The UK Government Alcohol Strategy.
513
Home Office, Next Steps Following the Consultation on Delivering the Government’s Alcohol Strategy (HO,
2013); Alan Brennan and others, ‘Potential Benefits of Minimum Unit Pricing for Alcohol versus a Ban on Below
Cost Selling in England 2014: Modelling Study’, BMJ (Online), 349.September (2014), 1–14.
514
HM Treasury, Spring Budget 2023 (HMT, 2023). Indexation is to take effect halfway through the financial
year, when it will coincide with changes to the duty structure which create standardised tax bands based on
alcohol by volume.
515
David Finch, Jo Bibby, and Tim Elwell-Sutton, Taking Our Health for Granted: Plugging the Public Health
Grant Funding Gap (Health Foundation, 2018).
516
Home Office, Next Steps Following the Consultation on Delivering the Government’s Alcohol Strategy.
517
Cécile Knai and others, ‘The Public Health Responsibility Deal: Has a Public-Private Partnership Brought
about Action on Alcohol Reduction?’, Addiction, 110.8 (2015), 1217–25.

169
years after the UK’s Chief Medical Officers updated their guidelines on low-risk drinking,
over 70 per cent of the most common products on shelves did not include the official, up-to-
date information on their labels. 518
Table AD1: Fiscal cost of alcohol duty cuts in Budgets 2013 to 2018 (£ million compared
with indexation)519

A 2018 study which estimated the proportion of alcohol sales revenue accounted for by
different groups of drinkers concluded that the alcohol industry would face significant
financial losses if consumers were to drink within guideline levels.520 Amongst alcohol and
health experts there is widespread concern about the lack of a cross-government strategy
on alcohol in England.521 But the government has continued to emphasise that it has no
plans for a stand-alone strategy on alcohol.522
Scotland and Wales
In recent years, alcohol policy in Scotland has taken a different turn from England, with a
framework of measures aimed at both young people and adults. 523 The measures include
implementation of a 50p minimum unit price per unit of alcohol in 2018 (following
legislation in 2012, with implementation delayed by legal challenges). Wales also

518
Alcohol Change UK, Drinking in the Dark: How Alcohol Labelling Fails Consumers (Alcohol Health Alliance
UK, 2020).
519
Figures from HM Treasury Budget Documents
520
Aveek Bhattacharya and others, ‘How Dependent Is the Alcohol Industry on Heavy Drinking in England?’,
Addiction, 113.12 (2018), 2225–32.
521
Alcohol Harms Commission, It’s Everywhere: Alcohol’s Public Face and Private Harm (Alcohol Health Alliance
UK, 2020); Grace Everest and others, Addressing the Leading Risk Factors for Ill Health (Health Foundation,
2022).
522
Hansard HOL Volume 801: Tuesday 21 January 2020 Baroness Williams. See also
https://hansard.parliament.uk/commons/2021-11-25/debates/BCD8D8C1-3FD0-49F9-8473-
E0DCEDB5E332/AlcoholHarm
523
Scottish Government, Alcohol Framework 2018: Preventing Harm (Scottish Government, 2018).

170
implemented a minimum unit price from 2020. A 2021 study in The Lancet reported that in
both Scotland and Wales, the introduction of the minimum unit price was associated with a
reduction in grams of alcohol purchased, compared with areas across the border in England.
The reduction was concentrated in households that bought the most alcohol. 524 More
recent research on Scottish experience continues to suggest that minimum unit pricing has
led to reduced consumption, including among heavier drinkers (but with some evidence of
financial strain among people with alcohol dependence). 525 The latest Public Health
Scotland evaluation says that their best estimate is that the minimum unit price significantly
reduced deaths wholly attributable to alcohol consumption (by 13.4 per cent).526

Outcomes
There is good data on adolescent drinking in England over this period, particularly in relation
to 11–15-year-olds. The data shows a significant fall in alcohol use from around 2003 to
2014 with improvement seeming to have stalled after 2014.
The data comes from several sources including the Health Survey England, the Smoking
Drinking and Drugs Use Survey (abbreviated to the SDDYP Survey), the international Health
Behaviour in School-aged Children Survey (the HBSC Survey) 527 and other NHS data. The
SDDYP Survey recently published data from 2021.528 This is shown in the relevant charts as a
dotted line.
Young people who have tried alcohol
Young people’s experience of alcohol fell sharply between 2003 and 2014. In this period, the
proportion of young people who had drunk alcohol halved for 13- to 15-year-olds, and fell
by nearly three quarters for 11- to 12-year-olds. Experience of alcohol amongst 8- to 10-
year-olds shrank from 17 per cent to almost nothing. The figures have plateaued since
2014. (Figure AD1).

Peter Anderson and others, ‘Impact of Minimum Unit Pricing on Alcohol Purchases in Scotland and Wales:
524

Controlled Interrupted Time Series Analyses’, The Lancet Public Health, 2667.21 (2021), 1–9.
525
John Holmes, ‘Is Minimum Unit Pricing for Alcohol Having the Intended Effects on Alcohol Consumption in
Scotland?’, Addiction, 2023, 1–8.
526
Grant M.A. Wyper and others, Evaluating the Impact of Alcohol Minimum Unit Pricing (MUP) on Alcohol-
Attributable Deaths and Hospital Admissions in Scotland (Public Health Scotland, 2023).
527
Jo Inchley, Dorothy Currie, Sanja Budisavljevic, and others, Spotlight on Adolescent Health and Well-Being.
Findings from the 2017/2018 Health Behaviour in School-aged Children (HBSC) Survey in Europe and Canada.
International Report. Volume 2. (WHO Regional Office for Europe, 2020).
https://digital.nhs.uk/data-and-information/publications/statistical/smoking-drinking-and-drug-use-among-
528

young-people-in-england/2021

171
Figure AD1: Trends in children who have ever had a ‘proper alcoholic drink’, England

80 74

70

60

50
Per cent

38
40 34 35

30

17
20
10 9
10
2
1
0

8-10 years 11-12 years 13-15 years

Source: Health Survey England, 2019

Regular drinking
The SDDYP Survey has long-run trend data on how frequently young people drink alcohol.
Figure AD2 shows the trends for 13- and 15-year-olds.
In 2000, as the left-hand panel shows, 43 per cent of 15-year-old boys and 38 percent of
15-year-old girls drank at least once a week. For both boys and girls this fell to 10 per cent
by 2014, then plateaued at the higher level of 14 per cent. In 2000, 5 per cent of 15-year-
old boys said they drank almost every day. This fell to almost zero in 2014, but has edged
up again since then.
The trend for 13-year-olds, in the right-hand panel below, shows a similar substantial fall in
drinking at least once a week, from 15 per cent of boys in 2000, to 2 per cent in 2014, with a
small rise again since 2014.

172
Figure AD2: Trends in proportion of pupils who usually drink almost every day, or at least
once a week, England
15 year olds' drinking frequency 50 13 year olds' drinking frequency

50 45

45 43 40

40 35
35 30
38
per cent of all pupils

30
25
25
20
20
14 15
15
10
10
10

5 5

0 0
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
2021

13 year old boys: almost every day

15 year old boys: almost every day 13 year old girls: almost every day
15 year old girls: almost every day
15 year old boys: at least once a week 13 year old boys: at least once a week
15 year old girls: at least once a week 13 year old girls: at least once a week

Source: SDDYP Survey, successive years

Experience of being drunk

The HBSC survey provides trend data on school pupils who have been drunk two or more
times. This data, shown in Figure AD3 below shows large falls for England between 2002
and 2014. The trend again largely plateaued after 2014.
Figure AD3: Trends in young people in England who have been drunk at least twice

60 55

50
40
29.9 31 15 y/o
Per cent

girls
30 15 y/o
25 boys
20 13 y/o
girls
10 5 13 y/o
boys
11y/o
0 girls
1998 2002 2006 2010 2014 2018
Source: HBSC international overview reports

173
Hospital admissions
Alcohol-related hospitalisations of under-18s were rising in the early 2000s. They began to
fall after 2005, and more than halved over the following decade. (Figure AD4) A gender gap
persists, with higher rates for girls. The rate of decline has been slow since 2014/15.

Figure AD4: Trends in hospital admission rates for under-18s for alcohol-specific
conditions, England
90

80

70

60
Rate per 100,000

50

40

30

20

10

Male single year

Female single year
Male 3 yr average

Sources: (a) pre 2006: Bellis et al, Contributions of alcohol use to teenage pregnancy, 2009. (b) post
2006, PHE Fingertips tool

What about over-16s?

The trend of lower drinking by under-16s appears to have carried on into the young adult
years. Drinking in the last week by 16- to 24-year-olds began to fall after 2005, dropping by
about one third by 2019. The trend for older age groups was a slower decline, or indeed no
decline at all for the over 65 age band. (Figure AD5)

174
Figure AD5: Trends in proportion of different age groups who have drunk alcohol in the
last week, England

80

70

60

50

40
Per cent

30

20

10

0
2002

2010

2018
1998
2002
2006
2010
2014
2018
1998
2002
2006
2010
2014
2018
1998

2006

2014

1998
2002
2006
2010
2014
2018
1998
2002
2006
2010
2014
2018
11-15 pupils (from 16 to 24 age range 25 to 44 age range 45 to 64 age range 65+ age range
SDDYP Survey)

Source: Health Survey England, 2019, and SDDYP Survey 2018

Inequalities
During the period of falling adolescent drinking, some inequalities were reduced.
Heavy drinkers vs lighter drinkers
Research by Melissa Oldham and others has found that, within the reduction in school
pupils’ weekly alcohol units between 2001 and 2016, those who drank at higher levels
experienced the largest falls in consumption.529
Regional gaps

Data on under-18 hospital admissions shows improvement across the country between
2007 and 2014, with reducing regional inequalities driven by dramatic improvements in the
North West and North East. But progress has slowed since 2014 and stalled entirely in some
regions (Figure AD6).

529
Melissa Oldham, Sarah Callinan, and others, ‘The Decline in Youth Drinking in England—Is Everyone
Drinking Less? A Quantile Regression Analysis’, Addiction, 115.2 (2020), 230–38.

175
Figure AD6: Trends in alcohol-specific hospital admissions of under-18s by region (three
year averages)

140
North East
120
North West

100 South West

rate per 100,000

Yorks Humberside
80
South East

60 West Midlands

East of England
40
E Midlands
20
London

Source: Office for Health Improvement and Disparities, Public Health Profiles

Ethnicity

An ethnic breakdown of school survey data, published in occasional years, is set out in
Figure AD7. It shows that white pupils drank the most at the beginning of the period and cut
consumption by the largest amount by 2014.

Figure AD7: Trends in 11-15 year old pupils who have drunk alcohol in last week, by broad
ethnic group, England

30

25

20
Per cent

15

10

0
2003- 2014 2016 2018 2021
5 White Mixed race Asian Black

Source: SDDYP successive years

176
Statistics on adolescent drinking in Scotland

Surveys of alcohol use by pupils in Scotland are undertaken triennially (normally) but the
last one dates from 2018. This therefore does not record experiences after the
implementation of the minimum unit price. The survey covers a sample of 13- and 15-year-
old boys and girls and found that among all groups, the proportion of pupils who had ever
had an alcoholic drink had increased since 2015, having previously been on a downward
trend. Just over half of 13 year olds (53 per cent) and around two-thirds of 15 year olds (70
per cent), who had ever had an alcoholic drink, had been drunk at least once, again an
increase on 2015.530

International
Falling adolescent drinking is a phenomenon seen in many high-income countries over the
last few decades. However, the timing has varied between countries, and the UK stands out,
both for the high starting point and the speed of reduction. 531 Between 2002 and 2014,
English boys and girls had the largest fall in weekly drinking out of 36 European countries
measured by the HBSC survey.532 This is illustrated for 15 year olds and selected countries in
Figure AD 8. This graph also shows that in England, Scotland, Wales and Ireland, boys and
girls have similar prevalence of drinking – unlike most other comparators.
Figure AD 8: Trends in weekly drinking by 15 year olds, by gender, selected countries

60
England Wales
Italy
50 Netherlands
Scotland
Germany
per cent of pupils

40

France
30
Ireland

20

10

0
2002
2006
2010
2014
2002
2006
2010
2014

2002
2006
2010
2014

2002
2006
2010
2014

2002
2006
2010
2014

2002
2006
2010
2014

2002
2006
2010
2014

2002
2006
2010
2014

Boys Girls
Source: HBSC surveys, successive years

530
Scottish Government, SALSUS Scottish Schools Adolescent Lifestyle and Substance Use Survey Full Report
(Scottish Government, 2018).
531
Rakhi Vashishtha and others, ‘Trends in Adolescent Drinking across 39 High-Income Countries: Exploring the
Timing and Magnitude of Decline’, European Journal of Public Health, 31.2 (2021), 424–31.
532
Inchley, Currie, Vieno, and others.

177
The HBSC 2018 report did not include the indicator used in Figure AD8. However, the 24
alcohol indicators published in both 2014 and 2018 HBSC Surveys are analysed in Table
AD2. This shows that more indicators deteriorated in Wales, England, and Scotland than
the average of all countries.
Table AD2: Change in adolescent alcohol indicators in HBSC survey 2014 and 2018

Number of alcohol indicators published in both 2014 and 2018 which …

.. deteriorated ..were stable .. improved
England 16 4 4
Wales 17 533 7 -
Scotland 11 4 9
HBSC average - 9 15

Plausible explanations of the trends in England

Explanations of trends in alcohol use in England need to account for a substantial and fairly
steady improvement after around 2003, followed by a stalling from 2014. During the period
of reduction, the fall was largest amongst those who drank most.
A number of studies have examined possible factors behind the period of reduction.
Attitudinal changes amongst parents and children, changes in affordability, and changes in
young people’s cultures and places of socialising are amongst the principal hypotheses
thought plausible in the literature so far.534 However, so far no study has attempted to
explain the post-2014 stalling.
In light of the data discussed above, I suggest that a rounded explanation would include
nearly a dozen factors over the last two decades. Many of them moved initially in a
direction likely to reduce adolescent drinking, then in a less positive one. Some of these
factors are closely inter-related.
1. Changed ethnic composition
It is likely that the changing ethnic mix of the youth population made some contribution to
falling levels of alcohol use: ethnic minority groups with lower average alcohol consumption
now represent a larger share of the youth population. But the scale of this change is
nowhere near enough to explain the halving of adolescent drinking.535 Moreover, as noted
above, detailed analysis of the adolescent drinking statistics found that the greatest
reduction in drinking during the decade was amongst white young people.
2. Later age of first introduction to alcohol
A later age of introduction to alcohol is clearly an important factor. As Figure AD1 showed,
between 2003 and 2014 the proportion of 13 to 15-year-olds who had tried alcohol halved,

533
For four of the Welsh indicators the deterioration was statistically significant.
534
Aveek Bhattacharya, Youthful Abandon: Why Are Young People Drinking Less? (London: Institute of Alcohol
Studies, 2016); Jonathan Birdwell and Ian Wybron, Character and Moderation: Encouraging the Next
Generation of Responsible Drinkers (Demos, 2015); O’Brien; Victoria Whitaker and others, ‘Young People’s
Explanations for the Decline in Youth Drinking in England’, BMC Public Health, 23.1 (2023), 1–13.
535
Bhattacharya.

178
and for 11 to 12-year-olds it fell by nearly three quarters. The reduction contributed to
lower drinking directly by cutting off several years of alcohol use. It also contributed
indirectly by reducing an important risk factor for alcohol dependence, and changing the
peer context, so teenagers became less likely to have multiple friends who drink. The
stalling of improvement since 2014 is of concern.
3. Changed attitudes and behaviours by parents
There is extensive evidence that during the 2000s parental attitudes became more cautious
towards alcohol and teenage drinking. 536 Drinking levels by 25-44 year olds also declined
during the decade, although not as strikingly as for younger generations (Figure AD5).
Survey evidence shows that between 2008 and 2014 a rising proportion of 11-15 year olds
believed their parents would not approve of them drinking. (Figure AD9) However after
2014, this indicator fell back again. These trends are potentially significant in both the
decline of adolescent drinking and the subsequent stalling of progress. More evidence about
trends in parental monitoring and their possible role in reduced adolescent drinking is given
in a 2014 analysis by Bhattacharya.537
Figure AD9: Trends in 11-15 year old pupils who believe their parent(s) doesn't or
wouldn't like them drinking, England

100% 2 1 1 1 2 2 3
90%

80%
47 43
48 49 49
53 49
70%

60%

50%

40%

30%
52 56
51 50 49 49
45
20%

10%

0%
2008 2010 2012 2014 2016 2018 2021 (post pandemic)

Does or would let pupil drink as much as he or she likes

Doesn't or wouldn’t mind pupil drinking if not too much
Doesn't or wouldn’t like pupil drinking

Source: SDDYP 2021

536
Bhattacharya.
537
Bhattacharya.

179
 4. Young people’s changed attitudes towards alcohol
A similar pattern of declining approval until 2014 is seen in data on the attitudes of pupils
themselves. Between 2003 and 2014, the proportion of school pupils who thought it was
‘OK to get drunk once a week’ fell by more than half, albeit with fluctuations from year to
year. Young people’s attitudes to drinking and drunkenness then appear to have become
more tolerant again after 2014. (Figure AD10). The drivers of these attitudinal changes are
likely to be complex and may include many of the other factors listed in this section, and/or
public health education and messaging. Perceptions of health and other risks seem to be a
significant element of the attitudinal picture: survey and interview research conducted with
12 to 19 year olds in 2018 and 2019 found that awareness of alcohol-related harm was the
most popular reason given by young people for why their generation drank less than
previous cohorts. The harms listed included both long-term health issues such as liver
disease, cancer, mental ill health, dependence and death, but also concern about vomiting,
hangovers, accidents, the potential for sexual or physical violence, and social consequences
to do with public drunkenness and damage to personal relationships.538
Figure AD10: Trends in 11–15-year-old pupils' attitudes to drinking by pupils of their age,
England

50

45

40

35

30

25
Per cent

20
20

15

10 7

OK to try getting drunk to see what it's like OK to drink alcohol once a week
OK to get drunk once a week

Source: Smoking Drinking and Drug use by Young People. 2021

538
Whitaker and others.

180
 5. Increased staying-on in education after 16, changes in absence and exclusion.
The large changes in teenagers’ educational experiences discussed in more detail in other
chapters of this report may also have influenced drinking patterns. The significant rise in
pupils staying on in full-time education after 16 during the period we are considering will
have restricted the time teenagers could devote to drinking, and changed young people’s
priorities for their time use. Participants in the interview and survey research described
above spoke of the need to do well in school in order to get on, with one respondent
commenting: ‘we don’t have the time or the privilege to waste getting drunk’. 539 The trends
in permanent exclusions and persistent absence discussed in Chapters 4 and 5 are also
relevant as both exclusion and absence are associated with higher levels of drinking. Until
around 2013 exclusion and absence were falling and therefore reducing the size of this
group at high risk of drinking. Rising exclusions and rises in severe absence since 2013 will
have had the opposite effect.
6. Changes in drug-taking and smoking
Other risk behaviours closely associated with drinking also fell during the period when
adolescent drinking was declining. Recent drug taking by 11- to 15-year-olds halved
between 2003 and 2011 and the percentage of 11- to 15-year-olds who were regular
smokers also halved between 2002 and 2010. Changes in drug use and smoking may well
have contributed to changes in drinking levels, through social peer group effects, where a
young person who meets friends to smoke or take drugs, may also find alcohol available in
the same setting. Drug-taking by teenagers has risen again since 2014 (see Chapter 8) but
smoking has not (Figure OY1).
7. Reduced ability to buy alcohol underage
As noted earlier, self-purchase of alcohol by young people is associated with higher drinking
levels. This was one reason why restricting underage selling by licensed premises was a
focus of policy from the mid-2000s onwards.
What the SDDYP Survey can tell us about self-purchase of alcohol by 11-15 year old drinkers
is shown in Figure AD11. There has been a reduction in self-purchase. In 2000, 54 per cent
of school pupils who drank bought alcohol, at least occasionally: by 2010 this had fallen to
44 per cent and by 2018 it was 39 per cent. Within these figures, the proportion of young
drinkers who bought from licensed premises fell. These changes may reflect tougher
enforcement of licensing laws, and possibly also increases in the cost of alcohol, discussed
later.
8. Changed location of drinking
Figure AD12 below shows SDDYP Survey answers about where young drinkers drink. The
graph shows a large reduction after 2006 in the proportion of young people drinking outside
(in the street, or in a park – the black line on the graph). This ‘homification’ of drinking
could have contributed to lower levels of drinking and drunkenness: street drinking is
associated with higher levels of alcohol use, and drinking in domestic settings implies
greater adult influence over what is drunk and how much.

539
Whitaker and others.

181
Figure AD11: Trends in self-purchase of alcohol by young people who drink, England

Current drinkers who ever buy Buy from licensed premises Buy informally
alcohol
60 60
54 60

50 50 50
44
Per cent of current drinkers

39
40 40 40

30 30 30

20 20 20

10 10 10

0 0 0
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
2020

1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
2020
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
2020
Buy from off-licence
Buy from
Proportion of current drinkers friend/relative
who buy alcohol at all Buy from shop or
supermarket Buy from someone else
Buy from pub/bar

Source: SDDYP 2021

Figure AD12: Trends in locations of drinking for 11-15 year olds who drink alcohol, England

90

80
Per cent of current drinkers

70

60

50

40

30
31
20 13
10

0
2004 2006 2008 2010 2012 2014 2016 2018
At home or someone else's home
Parties with friends
On the street, in a park, or somewhere else outside4
Pub or bar
Club or disco
Somewhere else
Source: SDDYP Survey 2021, Table 6.8

182
 9. Reduced face to face socialising by teenagers
These switches in drinking location sit within a context of much broader change in young
people’s socialising patterns. During the 2000s, teenagers in many countries increasingly
reported spending less time directly with peers and more time with parents. Researchers
looking at data between 2002 to 2010 found that England had a particularly large decline in
young people’s face to face socialising. 540 They rejected the idea that this was linked with
greater use of mobile phones and social media as the trends started before the rise of
large-scale social media access. Instead, they hypothesised that increased parental
supervision of their children, or a toughened climate of alcohol enforcement could be the
reason.541 More recent research with young people finds a complex pattern behind
changed socialising, including infrequent going out during the school week, especially for
those in compulsory education, and a greater choice of entertainment – particularly,
though not exclusively, associated with the use of social media.542

10. Increased access to substance misuse treatment for young people

Availability of treatment and support for young people with drinking problems is also likely
to be part of the picture. The overall number of young people in specialist treatment for
alcohol is small, but almost doubled during the Labour government, and reached a peak of
nearly 9,000 in 2008/09. By 2018/19, this had fallen back to less than 2,300.543

11. Affordability
One final contributory factor to be considered is affordability. This is recognised as a major
influence on alcohol consumption both at the population level and for young people. A
study of data on adolescent drinking in 33 countries concluded that reducing the
affordability of alcohol stood out as the most successful single measure. 544
A long-standing NHS indicator tracks alcohol affordability in the UK using an index that
accounts for price, tax, and household incomes. This shows that alcohol became more
affordable in every year between 1994 to 2007. However, the March 2008 increase in
alcohol duties began a five-year period of annual real increases in alcohol taxes. Combined
with the effect of the recession on incomes, this caused affordability to decline for four
years (until 2011). The alcohol duty escalator was cancelled by the Coalition in its 2013
budget and alcohol then became steadily more affordable. 545

540
Margaretha De Looze and others, ‘The Decline in Adolescent Substance Use across Europe and North
America in the Early Twenty-First Century: A Result of the Digital Revolution?’, International Journal of Public
Health, 64.2 (2019), 229–40.
541
De Looze and others.
542
Whitaker and others.
543
Public Health England, Young People’s Statistics from the National Drug Treatment Monitoring System (PHE,
2019).
544
Robyn Burton and others, ‘A Rapid Evidence Review of the Effectiveness and Cost-Effectiveness of Alcohol
Control Policies: An English Perspective’, The Lancet, 389.10078 (2017), 1558–80; E. Leal-López and others,
‘Association of Alcohol Control Policies with Adolescent Alcohol Consumption and with Social Inequality in
Adolescent Alcohol Consumption: A Multilevel Study in 33 Countries and Regions’, International Journal of
Drug Policy, 84 (2020), 102854.
545
https://digital.nhs.uk/data-and-information/publications/statistical/statistics-on-alcohol/2020/part-7

183
Discussion
The analysis above allows us to compare the large shifts in alcohol policy, enforcement,
context and social attitudes in England with the equally large shifts in outcomes. During the
period from 2003 to around 2013 the prevalence of adolescent drinking and drunkenness
improved dramatically with many indicators falling by a half or even three quarters.
These wider trends cannot, of course, prove causality. However, they provide suggestive
evidence of why adolescent drinking fell so sharply in England. Certainly, the reduction in
adolescent drinking becomes easier to understand when one appreciates how much less
likely it became, comparing 2013 with 2003, that a teenager would have tried alcohol as a
child, would know other drinkers, or would be able to buy alcohol themselves and drink it in
pubs, clubs, or parks. Many of these factors were the target of health education and
policing policies. Policy changes in educational participation and taxation may also have
played a role, alongside a wide range of social factors, including changes in parental drinking
habits.
While the eleven factors listed were moving in the right direction for part or all of the period
when alcohol use was declining, many of them have stalled or gone in the wrong direction
since 2013 or 2014. The reduction in early experience of alcohol plateaued after 2014.
Young people’s attitudes to alcohol, and their perception of their parents’ attitudes, became
less negative. Adolescent drug taking increased again, and levels of permanent exclusion
and severe absence grew. Alcohol became more affordable, and the number of young
people in treatment for alcohol fell. Several of these adverse developments reflects alcohol
policy decisions on tax and treatment funding; and some reflect wider policy changes in
relation to education and the availability of support for young people. Again, though
causality cannot be proved, it is striking that this weakening of policy and relaxation of
attitudes has coincided with the stalling of progress in reducing adolescent drinking and
drunkenness between 2014 and 2018.
So, what now, after the pandemic? Have the Covid lockdowns and restrictions on socialising
radically restricted adolescent drinking? There is very little new data since 2018. The one
published survey on adolescent drinking in England, conducted in 2021, shows limited
change and its results may still reflect the impact of the pandemic.546 For adults, the
pandemic is thought to have had mixed effects on alcohol consumption, with some drinkers
reining back, but some heavier drinkers consuming more. This interpretation was
corroborated by the latest figures for alcohol-related deaths: the number of deaths (all
ages) from alcohol-specific causes was 27 per cent higher in 2021 than in 2019.547 548
Writing in 2022, it is hard to assess the likelihood of a significant rise in adolescent drinking
in the future. A recent review by John Holmes and others described the trend of adolescent

546
Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
England, 2021.
547
Dr Robyn Burton and others, Monitoring Alcohol Consumption and Harm during the COVID-19 Pandemic
(Public Health England, 2021).
548
Office for National Statistics, Alcohol-Specific Deaths in the UK: Registered in 2021, 2022.

184
alcohol use as ‘stabilising’, but recognised that it could deteriorate.549 In truth, it will take
some time for a post-Covid trend to become clear.
The analysis in this chapter suggests several policy steps that could be taken to re-establish
a downward trend in adolescent alcohol use - restoring the real value of alcohol taxation or
introducing a minimum unit price; stepping up enforcement of underage drinking rules;
taking meaningful action on the risk factors of absence, exclusion, and drug use, and putting
renewed effort into influencing parents’ and young people’s knowledge about the risk of
alcohol. If government were to put in place a cross-department strategy to look again at the
levers that can influence adolescent drinking – taxation, enforcement, health guidance,
prevention and education - the benefits would very likely be seen in reduced social and
economic costs over many decades ahead.

549
John Holmes and others, ‘Youth Drinking in Decline: What Are the Implications for Public Health, Public
Policy and Public Debate?’, International Journal of Drug Policy, 102 (2022), 103606.

185
Chapter 8 Adolescent Drug Use
Introduction
In the late 1990s, the UK had more 15 to 16 year old drug users than any country in
Europe.550 But between the mid-2000s and 2013 or 2014, the proportion of school pupils in
England experimenting with drugs or taking them regularly fell by at least half, and the gap
with other countries narrowed. But this improvement appears to have stalled in the mid-
2010s. This chapter looks at the circumstances behind these striking trends. It first sets out
some key background facts on drug use by young people - its scale, who is most affected,
and the costs associated with drug use. It then charts the development of policy under
Labour, Coalition and Conservatives, sets out the trends in drug use by young people, and
discusses possible reasons for the trends.551

The geographical coverage of the chapter is in some respects England, but policing policy
is exercised at the England and Wales level, and legislation on drug classification is UK-
wide. Outcome data is England-only except where noted otherwise. The chapter mainly
covers policy and data up to 2019, but some key policy developments and data since 2019
have been included, to make the analysis as up to date as possible.

The scale of drug use by young people

As of 2018, just over one third of 15 year old pupils in England had experience of drug-
taking.
Figure DR1 below gives more detail of 15 year old pupils’ use of individual types of drug as
of 2018.
Figure DR1: Proportion of 15 year old pupils using different types of drugs: 2018, England
25 22.5
Per cent of 15 year old pupils

19.3
20

15
10.7
10 8.4
7.0

5 3.6

0
Cannabis Any psychoactive Glue, gas, Any stimulant Any psychadelic Any opiate Any class A
substance aerosols or
solvents
Ever taken Used in last year Used in last month
Source: Smoking Drinking and Drug Use by Young People, 2018, tables 8.7c, f and i

550
European Monitoring Centre For Drugs And Drug Addiction, Annual Report on the State of the Drugs
Problem in the European Union (EMCDDA, 1998).
551
I was involved in drugs policy as Home Office Director General of Crime and Policing between 2005 and
2008.

186
Smaller proportions of younger secondary school children had tried drugs, with the smallest
proportion - amongst 11-year-olds - being 7 per cent.552

Costs
Drugs pose a range of costs for those who take them, those around them, and society as a
whole:

• A wide range of physical and mental health risks are associated with drug use,
varying with the drug, the nature of use and other factors. The health risks include
drug dependence, injection-related conditions and infections, heart and blood
pressure problems, unintentional injuries, mental health problems, and associations
with self-harm and suicide.553

• There are associations between regular cannabis use and higher risk of adverse
educational outcomes and school drop-out. 554

• Young people who are exposed to drugs and alcohol in their early teens have been
found to be at increased risk for later poor outcomes, including substance
dependence, early pregnancy, and criminal conviction, after adjusting for other risk
factors.555

• There is a large body of evidence on the drug-crime association: a review of multiple

research studies in 2008 showed that the odds of offending were three to four times
greater for drug-users than non-drug-users, with the odds of offending highest
among crack users and lowest among recreational drug users.556

• Home Office research in 2013 estimated that between 44 and 48 per cent of
acquisitive crimes (excluding fraud) were drug-related.557 The link is not limited to
acquisitive crime: around half of the increase in homicides in England and Wales

552
See Figure DR 3 later in this chapter for fuller figures including trends.
553
Louisa Degenhardt and others, ‘The Global Burden of Disease Attributable to Alcohol and Drug Use in 195
Countries and Territories, 1990–2016: A Systematic Analysis for the Global Burden of Disease Study 2016’, The
Lancet Psychiatry, 5.12 (2018), 987–1012; Dame Carol Black, Review of Drugs. Evidence Relating to Drug Use,
Supply and Effects, Including Current Trends and Future Risks (Home Office, 2020); Becky Mars and others,
‘Predictors of Future Suicide Attempt among Adolescents with Suicidal Thoughts or Non-Suicidal Self-Harm: A
Population-Based Birth Cohort Study’, The Lancet Psychiatry, 6.4 (2019), 327–37; Paul Moran and others, ‘The
Natural History of Self-Harm from Adolescence to Young Adulthood: A Population-Based Cohort Study’, The
Lancet, 379.9812 (2012), 236–43.
554
Alexander I. Stiby and others, ‘Adolescent Cannabis and Tobacco Use and Educational Outcomes at Age 16:
Birth Cohort Study’, Addiction, 110.4 (2015), 658–68; Jan C. van Ours and Jenny Williams, ‘Why Parents Worry:
Initiation into Cannabis Use by Youth and Their Educational Attainment’, Journal of Health Economics, 28.1
(2009), 132–42. See also Chapters 4 and 5 of this report.
555
Candice L. Odgers and others, ‘Is It Important to Prevent Early Exposure to Drugs and Alcohol among
Adolescents?’, Psychological Science, 19.10 (2008), 1037–44; Wayne D. Hall and others, ‘Why Young People’s
Substance Use Matters for Global Health’, The Lancet Psychiatry, 3.3 (2016), 265–79.
556
Trevor Bennett, Katy Holloway, and David Farrington, ‘The Statistical Association between Drug Misuse and
Crime: A Meta-Analysis’, Aggression and Violent Behavior, 13.2 (2008), 107–18.
557
Hannah Mills, Sara Skodbo, and Peter Blyth, Understanding Organised Crime : Estimating the Scale and the
Social and Economic Costs (Home Office, 2013).

187
 between 2013/14 and 2017/18 was due to cases involving drug dealers or drug users
or a drug-related motive.558

As of 2017-18, the government estimated that the cost of harms related to illicit drug use in
England was £19.3 billion. This figure includes significant costs to the health service, social
care, enforcement, families of drug users, and the impact of drug-related deaths. However,
by far the largest element of the costs, at £9.3 billion, was the cost of drug-related crime.559
Many young people ‘mature out’ of drug use. But it has been estimated that, without
treatment, 37 to 44 per cent of young people in substance abuse treatment would develop
drug/ alcohol misuse problems as adults. For Class A drug users, the proportion is higher,
with up to 95 per cent of teenage Class A users being estimated to continue using drugs in
adulthood.560

Groups over-represented in drug use

Much of our data on adolescent drug use comes from the Smoking Drinking and Drug Use by
Young People surveys conducted regularly in mainstream schools in England (hereafter
abbreviated to the SDDYP survey). In this population, analysis in 2018 found that the
characteristics most strongly associated with recent drug taking were (in descending order
of association):

• smoking
• drinking
• having a family that does not discourage drug use
• playing truant
• living in London
• being older
• having low happiness yesterday
• being excluded from school
• being male. 561
In the same survey in 2018, the most common reasons given for first trying drugs were
wanting to see what it was like (50 per cent), wanting to get high or feel good (22 per cent),
because friends were doing it (16 per cent) and wanting to forget problems (also 16 per
cent).562 In relation to ethnic and socio-economic differences, the SDDYP survey found that:

• in 2018, Asian pupils were less likely than other ethnic groups to have taken drugs in
the last year.563

558
Black.
559
Black.
560
Frontier Economics, Specialist Drug and Alcohol Services for Young People – A Cost Benefit Analysis (DfE,
2010).
561
Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
England 2018 (NHS Digital, 2019).
562
Pupils could give more than one answer.
563
13 per cent of Asian pupils had taken drugs in the last year, compared with 17 per cent of white pupils, 18
per cent of black pupils, and 23 per cent of mixed ethnicity pupils. SDDYP 2018, Table 8.38

188
 • there were slightly higher rates both for ever having taken drugs and recent drug use
in the second least deprived quintile of areas.564

An older sibling who uses cannabis is strongly associated with using cannabis or other illegal
drugs according to a 2006 study of 15–16-year-olds in six countries, including the UK. The
next strongest associations in that study were for ‘going out most evenings’ and ‘friends use
cannabis’.565
Some groups who are under-represented in school surveys are known to have higher rates
of drug use: this includes excluded pupils, school absentees, looked after young people,
gang members, and young offenders.566

Policies and spending programmes

Labour: 1997-2010
Overall approach
Drugs policy was an early priority for the 1997 Labour Government, and a cross-
governmental strategy was published in April 1998. It included policies that applied to all
age groups – treatment, tackling drug availability and drug-related crime, and protecting
communities – as well as specific objectives to help young people resist drugs. It
acknowledged the start made by the previous Conservative government’s ‘Tackling Drugs
Together’ programme, launched in 1995, which had set up Drug Action Teams. 567
Increases in expenditure
The 1998 Spending Review announced an additional £211 million for proactive anti-drugs
work over three years and set a cross-departmental PSA target both to reduce the
proportion of people under 25 misusing illegal drugs, and to increase problem drug users’
participation in treatment. Drug budgets were the focus of substantial further investment
after 2000. Between 2000/01 and 2005/06 spending on drug treatment in England more
than doubled (from £234 million to £573 million), and expenditure on the young people and
communities elements of the drugs programme more than quadrupled, rising from £108

564
SDDYP 2018, Table 10.23. Note that other surveys have found a different pattern, for example the What
about Youth? Survey in 2014 found that 15 year olds who were eligible for free school meals had slightly
higher experience of cannabis.
565
Anna Kokkevi and others, ‘Psychosocial Correlates of Substance Use in Adolescence: A Cross-National Study
in Six European Countries’, Drug and Alcohol Dependence, 86.1 (2007), 67–74; See also Stephen J. Bahr, John P.
Hoffmann, and Xiaoyan Yang, ‘Parental and Peer Influences on the Risk of Adolescent Drug Use’, Journal of
Primary Prevention, 26.6 (2005), 529–51.
566
Mason Jenny, Nicola Pearce-Smith, and Caryl Beynon, Smoking, Drinking and Drug Use among Hard to
Reach Children and Young People: An Evidence Synthesis Report (Public Health England, 2018); Melissa
Oldham, Michael Livingston, and others, ‘Trends in the Psychosocial Characteristics of 11–15-Year-Olds Who
Still Drink, Smoke, Take Drugs and Engage in Poly-Substance Use in England’, Drug and Alcohol Review, 40.4
(2021), 597–606; Children’s Commissioner, Keeping Kids Safe: Improving Safeguarding Responses to Gang
Violence and Criminal Exploitation (CCO, 2019).
567
HM Government, Tackling Drugs to Build a Better Britain: The Government’s Ten-Year Strategy for Tackling
Drugs Misuse (The Stationery Office, 1998).

189
million to over £500 million.568 Spending on drugs programmes then remained stable in cash
terms between 2006/07 and 2009/10.569
Expansion of treatment
The extra resources from these spending reviews allowed expansion of drug treatment,
linked to new schemes to identify Class A drug-users in the criminal justice system and
channel them towards treatment. 570 These schemes were progressively developed to cover
the different stages of the criminal justice process (arrest, charge, bail, or sentence) and
were grouped together as the ‘Drug Interventions Programme’, led from the Home Office.
All areas in England and Wales were expected to follow the approach.571 A new National
Treatment Agency was set up in 2001 to expand drug treatment and set standards. The
number of drug users in treatment contact increased substantially over this period, more
than doubling from an estimated 85,000 in 1998 to over 200,000 in 2009/10, with the
majority of those in treatment in the 20 to 40 age bracket. 572
Treatment for young people
Treatment settings dominated by adult drug users are often inappropriate for under-18s,
and dedicated services for young people were initially in short supply.573 By 2002, young
people’s services were provided in 80 per cent of Drug Action Team areas and all Youth
Offending Teams had named drug workers to assess and arrange for support.574 From 2003
onwards, a version of the Drug Interventions Programme was introduced for under-18s.575
In 2004, funding streams from several departments were pooled to form a £58 million
Young People Substance Misuse Partnership Grant.576 Drug Action Teams and Children’s
Services were encouraged to work together to identify vulnerable groups at high risk of
using drugs, with regional support teams based in Government Offices.577 Data recording

568
DrugScope, Annual Report on the UK Drug Situation 2001, ed. by Stephane Aujean and others (EMCDDA,
2001); UK Focal Point on Drugs, United Kingdom Drug Situation 2006: National Report to the European
Monitoring Centre for Drugs and Drug Addiction, ed. by G Eaton and others (EMCDDA, 2006). Figures for
England.
569
UK Focal Point on Drugs, United Kingdom Drug Situation 2010: National Report to the European Monitoring
Centre for Drugs and Drug Addiction, ed. by Charlotte Davies and others (EMCDDA, 2010). Table 1.1.
570
HM Treasury, Public Services for the Future : Modernisation, Reform, Accountability.
571
Home Office, Departmental Report 2007/08 (Home Office, 2008).
572
National Treatment Agency, Statistics from the National Drug Treatment Monitoring System 2005-06 (NTA,
2006); National Treatment Agency, Statistics from the National Drug Treatment Monitoring System 2009-10
(NTA, 2010); HM Government, Drugs: Protecting Families and Communities: The 2008 Drug Strategy (HMG,
2008).
573
Ilana B. Crome, ‘Overview: Beyond Guidelines and Guidance - Psychosocial Perspectives on Treatment
Interventions for Young People with Substance Problems in the United Kingdom’, Drugs: Education, Prevention
and Policy, 13.3 (2006), 203–24.
574
Home Office, Updated Drug Strategy 2002 (Home Office, 2002).
575
Matrix Research and Consultancy and Institute for Criminal Policy Research, Evaluation of Drug
Interventions Programme Pilot for Children and Young People: Arrest Referral, Drug Testing and Drug
Treatment and Testing Requirements (Home Office, 2007).
576
Department of Health and North West Public Health Observatory, United Kingdom Drug Situation 2005:
National Report to the European Monitoring Centre for Drugs and Drug Addiction, ed. by G Eaton and others
(EMCDDA, 2005).
577
Department for Education and Skills, Home Office, and Department for Health, Every Child Matters : Change
for Children Young People and Drugs (HMG, 2005).

190
the number of under-18s in specialist substance misuse in England began to be published
systematically from 2005/06. It shows that overall numbers rose from 17,000 in that year to
just over 24,000 in 2008/09, with some decline thereafter. This trend is shown with the
post-2010 trend in Figure DR2 below.578
Figure DR2: Trends in number of young people in substance use treatment in England by
age
30,000

24,055
25,000

20,000 17,001
14,485
15,000

10,000

5,000

Under 12 12 years old 13 years old 14 years old

15 years old 16 years old 17 years old Total under 18
Source: Young people's Statistics from the National Drug Treatment Monitoring System 2018/19

The impact of treatment

There is a well-developed evidence base for the cost-effectiveness of treatment for both
adults and young people.579 For example, the government-commissioned Drug Treatment
Outcomes Research Study followed a large cohort of adults seeking treatment in 2006/07,
and found that regular use of drugs reduced significantly between baseline and follow-up,
especially amongst the most problematic drug users. While 40 per cent of the sample said
that they had committed an acquisitive crime such as burglary, robbery, theft or bag-
snatching in the four weeks before their first interview, this halved to 21 per cent at first
follow-up and fell to 16 per cent by the second follow-up. Overall, this study found that
treatment had a cost-benefit ratio of around 2.5 to 1.580 Analysis of drug treatment for
young people in 2010 found that, four years after treatment, the ‘re-presentation rates’ for
young people (i.e. the number who came back for treatment) ranged from 16 per cent for
alcohol, 17 per cent for cannabis, up to 40 per cent for heroin and crack users. This

578
These figures are for both drug and alcohol treatment as no separate series is published for the whole
period.
579
McSweeney, Turnbull, and Hough, The Treatment and Supervision of Drug-Dependent Offenders: A Review
of the Literature; Peter Burkinshaw and others, An Evidence Review of the Outcomes That Can Be Expected of
Drug Misuse Treatment in England (Public Health England, 2017).
580
Michael Donmall and others, Research Report 23: Summary of Key Findings from the Drug Treatment
Outcomes Research Study (DTORS) (Home Office, 2009).

191
compares favourably with the continued substance misuse rates expected without
treatment. 581
Policing and communities
The emphasis on channelling arrestees and offenders into drug treatment was just one of
many changes in the engagement of police and enforcement agencies in drug issues over
these years. Some of the changes were at national level, such as the creation of the Serious
Organised Crime Agency. But much of the focus was on policing at neighbourhood level,
where increased officer numbers and the recruitment of 16,000 police community support
officers came with expectations of reductions in both crime and its impact on communities.
The government’s early Public Service Agreement targets for crime (described in Chapter 2)
focused on acquisitive crimes such as vehicle crime, domestic burglary and robbery, which
were often linked with drug use. New powers and new funding streams accompanied these
targets. For example, the police acquired new powers to shut down crack houses rather
than simply prosecute the owner.582 And a Communities Against Drugs funding stream
created in the 2001 Budget channelled £220 million over three years to Crime and Disorder
Reduction Partnerships to help communities mobilise against drugs.583
Prevention with young people
A range of policies were introduced by Labour to try to prevent disengaged young people
dropping out of school and getting involved in drugs and other risk behaviours. Initiatives to
reduce school absence and exclusion (described in Chapters 4 and 5) and the cross cutting
youth policies described in Chapter 3 created a workforce of learning mentors, police in
schools, youth workers and Connexions advisors able to support young people and try to nip
problems in the bud.584 585 The goal of providing more constructive things for young people
to do was seen as particularly important for preventing drug use. From 2000 onwards, the
Home Office and the Youth Justice Board funded ‘Positive Futures’ in partnership with Sport
England and the Football Foundation, using sport to reduce anti-social behaviour, crime and
drug use among at risk 10-16 year olds. In 2007, HM Treasury and the Department for
Children Schools and Families set out a 10 year plan backed up by £185m between 2008 and
2011 to improve youth facilities and support youth volunteering.586
Education and information
The wider strategy to raise young people’s understanding of drugs and their consequences
included more funding for in-school drug education (DfES invested £21 million beginning in
2000 to train teachers in effective drug education) and the establishment of the web and
phone-based ‘FRANK’ resource in 2003. FRANK was designed to be a youth-friendly and

581
Frontier Economics.
582
Anti-social Behaviour Act 2003
583
Tim McSweeney, Paul J Turnbull, and Mike Hough, UK Drug Policy Commission : Tackling Drug Markets and
Distribution Networks in the UK (UK Drug Policy Commission, 2008).
584
Hallam, Castle, and Rogers; Department for Education and Skills, ACPO, and Youth Justice Board, Safer
School Partnerships Guidance (DfES, 2006); Hoggarth, Smith, and Britain.
585
Advisory Group on Drug and Alcohol Education, Drug Education: An Entitlement For All (DCSF, 2008).
586
Department for Children Schools and Families, Aiming High for Young People: A Ten Year Strategy for
Positive Activities (DCSF, 2007).

192
non-judgemental source of information about drugs. In 2005/06 its website received 5.7
million hits from over 2 million visitors and the telephone helpline received approximately
1,350 calls per day.587 By 2009 over a third of pupils in the SDDYP survey listed it as a helpful
source of helpful information about drugs.588
Cannabis
FRANK played a particular role in campaigns about the dangers of specific drugs, such as
cocaine and cannabis.589 The ‘Brain Warehouse’ cannabis campaign run in 2006/07 was
reported to have reached 67 per cent of young people through its television adverts.590 This
campaign was part of the government’s response to rising evidence of the mental health
risks associated with cannabis. This concern eventually led to cannabis being reclassified
(upwards) from Class C to Class B in 2009, having only been downgraded from B to C in
2004. More detail on cannabis classification is set out in the box below.

Cannabis

In 2001, Home Secretary David Blunkett commissioned the Advisory Committee on the
Misuse of Drugs (ACMD) to review the classification of cannabis, and in July 2002
accepted their recommendation to reclassify it from Class B to Class C. 591 This was
supported by the police who wanted scope to handle cannabis offences more flexibly, and
devote more time to Class A drug use.592 The change took effect in 2004.

The government soon became concerned about the mental health effects of cannabis use
and the increased use of higher potency ‘skunk’. The new Home Secretary, Charles
Clarke, asked for a reconsideration. The AMCD did not recommend reclassification but
noted ‘an association between cannabis use and the development of psychotic
symptoms’.593 In 2007, Prime Minister Gordon Brown announced a further review. The
ACMD again advised no change, but the Government said that it must err on the side of
caution and moved cannabis back to Class B from January 2009. 594 595
Labour's last drug strategy was published in 2008. It took credit for a reduction in young
people’s drug use, the doubling of the number of people accessing drug treatment, shorter
waiting times for drug treatment, and a fall in recorded acquisitive crime of around 20 per

587
UK Focal Point on Drugs, 2007 National Report to the European Monitoring Centre for Drugs and Drug
Addiction, ed. by G Eaton and Others (EMCDDA, 2007).
588
Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
England in 2009 (NHS Digital, 2010).
589
https://www.bbc.co.uk/news/magazine-21242664
590
UK Focal Point on Drugs, 2008 National Report to the European Monitoring Centre for Drugs and Drug
Addiction (EMCDDA, 2008).
591
https://hansard.parliament.uk/commons/2002-07-10/debates/8ce622ec-55db-42af-b325-
7a32c719224f/DrugsStrategy
592
The Police Foundation, Drugs and the Law: Report of the Independent Inquiry into the Misuse of Drugs Act
1971 (Police Foundation, 2000).
593
Advisory Council on the Misuse of Drugs, Further Consideration of the Classification of Cannabis under the
Misuse of Drugs Act 1971 (Home Office, 2005).
594
Advisory Council on the Misuse of Drugs, Cannabis : Classification and Public Health (Home Office, 2008).
595
Home Office, Explanatory Memorandum to the Misuse of Drugs Act 1971 (Amendment) Order 2008
<http://www.legislation.gov.uk/uksi/2015/215/pdfs/uksiem_20150215_en.pdf>.

193
cent. It set as future priorities a continued focus on policing and treatment, more help for
families affected by drug use, and benefits incentives for people with drug problems to
move towards treatment, training and employment. 596
Coalition and Conservatives: 2010 to 2019
Overall approach
When the Coalition took office in 2010, drugs policy moved to a more devolved model with
reduced central funding. The 2010 Public Health White Paper announced that the
government intended to ‘end central control and give local government the freedom,
responsibility and funding to innovate and develop their own ways of improving public
health’.597 Following this through, from 2013, Local Authorities became responsible for local
drug services and treatment, and the National Treatment Agency ceased to exist, its
functions being transferred to the newly formed Public Health England. The 2010 Drug
Strategy, issued by the Home Office, also pushed accountability for reducing drug-related
crime to a local level, to elected Police and Crime Commissioners, who were to take office in
2012.598 The devolution of police and crime to local levels included the right to drop
programmes of drug testing and treatment on arrest and conviction (such as the Drugs
Intervention Programme).599
Reducing prevention spending
These devolution statements were followed by funding cuts. From 2011/12, the Coalition
merged funding for drugs prevention with other funding streams, and cut the total, to
create the Early Intervention Grant. The Home Office later estimated that central
government spending on drugs early intervention and demand reduction ‘may have fallen
from £269 million in 2010/11 to around £215 million in 2014/15’ but emphasised that these
were only estimates.600 At the end of 2011/12, half of Drug Action Teams surveyed reported
a significant decline in substance misuse prevention activity funding. 601 The drugs
information service FRANK continued to exist (and still does) but there was no marketing
budget for it in three out of five years under the Coalition. ‘Positive Futures’ came to an end
in 2013.602 More background on cross-cutting prevention spending for the
Coalition/Conservative period is reported in Chapter 3 (Child Poverty and Youth Support).
Reducing expenditure on treatment for young people
Funding for drug and alcohol treatment by local authorities began to be squeezed under the
Coalition as local budgets faced the impact of austerity. In 2011/12, two fifths of Drugs
Action Teams surveyed reported a significant decrease in funding for young people’s

596
HM Government, Drugs: Protecting Families and Communities: The 2008 Drug Strategy.
597
Department of Health, Healthy Lives, Healthy People: Our Strategy for Public Health in England (HMG,
2010).
598
Home Office, Drug Strategy 2010: Reducing Demand, Restricting Supply, Building Recovery: Supporting
People to Live A Drug Free Life (HMG, 2010).
599
Brendan J. Collins, Kevin Cuddy, and Antony P. Martin, ‘Assessing the Effectiveness and Cost-Effectiveness
of Drug Intervention Programs: UK Case Study’, Journal of Addictive Diseases, 36.1 (2017), 5–13.
600
Home Office, An Evaluation of the Government’s Drug Strategy 2010 (HMG, 2017).
601
Helen Beck, Charting New Waters: Delivering Drug Policy at a Time of Radical Reform and Financial
Austerity (UK Drug Policy Commission, 2012).
602
Home Office, An Evaluation of the Government’s Drug Strategy 2010.

194
substance misuse treatment services.603 The decreases continued under the Conservatives.
The Public Health grant from central government fell in real terms every year between
2014/15 and 2018/19, and spending on young people’s drug and alcohol services fell 29 per
cent in real terms over that period.604 By 2018/19 the number of young people receiving
specialist drug and alcohol treatment was a third down on 2010/11 as shown in Figure DR 2
earlier in the chapter.
Policing
The 2010 spending review police spending cuts described in Chapter 2 naturally had an
impact on the policing of drugs. By 2014/15 police officer numbers had fallen by 12 per cent
and community support officer numbers were down by more than a quarter. Both fell
further after 2015 under the Conservatives. In 2011, three fifths of police forces said that
they expected to spend less on tackling illicit drugs. and by 2015 drug seizures by police
forces has fallen by a quarter compared with 2010. 605 606
Other expenditure
The government introduced some new programmes relevant to drugs. An element of
spending on the Troubled Families Programme was seen as part of the drug strategy.607
However, the independent evaluation of the programme found no significant impacts in
relation to drug use.608 The government introduced ‘Payment by Results’ pilots for drug and
alcohol treatment in eight areas from 2012 to 2014. Evaluation found mixed results, with
pilot areas doing better on some outcomes and worse on others.609
One further change made after 2014 was that the longstanding annual SDDYP survey of
young people was moved to a biennial basis. This has led to a lack of timely information
during a period of concerning developments in drug (and alcohol) use, compounded by the
loss of regular data gathering during the pandemic.
Emergence of the County Lines problem
During 2014, evidence began to emerge of a new model of drug supply known as ‘County
Lines’.610 As described by the National Crime Agency (NCA), this generally involved ‘a group
from an urban area expanding their operations by crossing one or more police force
boundaries to more rural areas, setting up a secure base and using runners to conduct day
to day dealing’. The main drugs concerned were Class A, particularly heroin and crack

603
Beck, Charting New Waters: Delivering Drug Policy at a Time of Radical Reform and Financial Austerity.
604
Finch, Bibby, and Elwell-Sutton.
605
Helen Beck, Drug Enforcement in an Age of Austerity: Key Findings from a Survey of Police Forces in England
(UK Drug Policy Commission, 2011).
606
https://www.gov.uk/government/statistics/seizures-of-drugs-in-england-and-wales-financial-year-ending-
2021/seizures-of-drugs-in-england-and-wales-financial-year-ending-2021 Summary Table 1
607
Home Office, An Evaluation of the Government’s Drug Strategy 2010.
608
Day and others.
609
Emma Disley and others, Evaluation of the Drugs and Alcohol Recovery Payment by Results Pilot Programme
Interim Final Report (University of Manchester, 2017). See also
https://www.manchester.ac.uk/discover/news/payment-by-results-for-drug-misuse-treatment-gets-mixed-
reception/
610
Alexandra Topping, ‘London Gangs Using Children as Drug Mules as They Seek to Expand Markets’, The
Guardian, 5 January 2014.

195
cocaine. Vulnerable young people were being exploited as runners, and guns and knives
were being used in inter-gang violence. 611 By 2017 the NCA found that only five police force
areas in England and Wales were unaffected by County Lines and their 2018 report noted
the groups’ increasing focus on recruiting runners around pupil referral units, special
schools, foster homes and homeless shelters. 612 613 The development of County Lines was
widely seen as contributing to worsening knife crime and homicides of young people, and
the proportion of people sentenced for drug supply who were aged under 21 increased
between 2013 and 2018.614 615
County Lines Government response
The government's response to county lines emerged surprisingly slowly. In 2017 it legislated
to create powers to close down phone numbers used for drug dealing.616 In April 2018, the
government produced its Serious Violence Strategy, which acknowledged that drug-related
violence was one of the key factors behind worsening serious violence trends: the strategy
set up an Early Intervention Youth Fund (of £11 million over two years), called on Police and
Crime Commissioners to prioritise serious violence, provided £3.6m funding to support a
National County Lines Co‑ordination Centre, and promised new legislation on sale and
possession of knives. In March 2019, a new Youth Endowment Fund was created with a
£200m government endowment, ‘to prevent children and young people becoming involved
in violence […] by finding out what works and building a movement to put this knowledge
into practice’. 617 The government’s strategy was criticised by the Home Affairs Select
Committee in July 2019 as ‘a completely inadequate response to this wave of violence’ with
criticism for the government’s ‘hands-off approach’ and lack of funding for prevention.618
Drug treatment
The position on drug treatment was also attracting increasing criticism with the ACMD
warning in 2017 that treatment was ‘facing disproportionate decrease in resources, likely to
reduce treatment penetration and the quality of treatment in England.619 Data about drug
use now suggested that the downward trend in young people’s drug use had stopped. In
2017, the SDDYP survey found a large increase between 2014 and 2016 in the proportion of
pupils who had taken drugs. This increase was visible across all year groups and several drug
types including class A. Yet the publication declared that ‘an estimate from the next survey
in 2018 is required before we can be confident that these survey results reflect a genuine
trend’.620 The 2018 SDDYP survey school statistics, when published in 2019, showed that the

611
National Crime Agency, Intelligence Assessment: County Lines, Gangs and Safeguarding (NCA, 2015).
612
National Crime Agency, County Lines Drug Supply, Vulnerability and Harm.
613
National Crime Agency, County Lines Violence, Exploitation and Drug Supply (NCA, 2017).
614
Black.
615
The trends on knife crime are set out in Chapter 2 and data on young homicide victims is in Appendix 1.
616
Digital Economy Act 2017
617
https://youthendowmentfund.org.uk/about-us/
618
House of Commons Home Affairs Committee, Serious Youth Violence (HoC, 2019).
619
Advisory Council on the Misuse of Drugs, Commissioning Impact on Drug Treatment (Home Office, 2017).
620
Health and Social Care Information Centre, Smoking , Drinking and Drug Use among Young People: England,
2016 (NHS Digital, 2017).

196
higher level in 2016 had been sustained.621 The picture was similar for the over-16s:
statistics released in July 2018 showed that what had previously been described as
fluctuations in Class A drug use by 16-24 year olds, now showed a statistically significant
increase between 2011/12 and 2017/18, and ‘a genuine rise in Class A drug use among this
age group’.622
Developments in 2019
The government took further measures to step up its response to drugs and serious violence
during 2019 under both the May and Johnson premierships. These included extra funding to
recruit more police officers, and a £100 million Serious Violence Fund, supporting 18 areas
to set up Violence Reduction Units. 623
Dame Carol Black review of drugs
In February 2019 Home Secretary Sajid Javid appointed Dame Carol Black to lead a review of
illegal drugs. The findings from her review led the government to announce a major
overhaul of drugs policy in December 2021, including new investment in treatment and
workforce development, a move to revive local drug partnerships, and the re-creation of a
unit to coordinate drugs policy across departments. 624 The new approach included plans to
reinstate provision lost over the previous decade. Grants of £85 million were allocated to
fifty local authorities for 2022/23 as part of a ‘phased approach’ to ‘rebuild the drug
treatment and recovery system’. 625 In February 2023, the government announced
allocations of £154 million for 2023/24, and indicative funding of £266 million for
2024/25.626

Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
621

England 2018. (NHS Digital, 2019).

622
Home Office, Drug Misuse: Findings from the 2017/18 Crime Survey for England and Wales (Home Office,
2018).
Home Office, Serious Youth Violence: Government Response to the Home Affairs Committee’s Sixteenth
623

Report of Session 2017–2019 (HoC, 2020).

624
HM Government, From Harm to Hope: A 10-Year Drugs Plan to Cut Crime and Save Lives (HMG, 2021).
625
https://www.gov.uk/government/publications/extra-funding-for-drug-and-alcohol-treatment-2022-
to-2023/additional-drug-and-alcohol-treatment-funding-allocations-2022-to-2023
626
https://www.gov.uk/government/news/421-million-to-boost-drug-and-alcohol-treatment-across-england
197
 Dame Carol Black review of drugs: key extracts from analysis
• The most alarming development in the recent evolution of the UK drugs market has been
the widespread involvement of children and young people in drug supply. [..] The
Children’s Commissioner estimates that around 27,000 young people in England and
Wales identify as gang members.

• Drug use among children (aged 11 to 15) has increased by over 40% since 2014, following
a long-term downward trend. This appears to be occurring across a wide range of
substances and across most demographics.

• However, there has been a sustained and significant decrease in the number of young
people receiving specialist interventions for their drug use.

• The trends in young people becoming involved in drug supply, drug consumption and in
serious violence have occurred against a backdrop of increasing numbers of children in
care and children in need, falling local government budgets, cuts to young people’s
services and increasing child poverty. Another key factor appears to be the rapid increase
in permanent exclusions from school over the past five years.

• Spending on treatment has reduced significantly because Local Government budgets have
been squeezed and central Government funding and oversight has fallen away. [..] The
amount of un-met need is growing, some treatment services are disappearing, and the
treatment workforce is declining in number and quality.

• Over the past decade, tackling drugs has fallen down the priority list for nearly all police
forces. This has partly been in response to funding cuts (of around 20% from 2010) and the
emergence of other priorities, such as tackling domestic abuse and mental health
incidents. [..] The number of drug seizures by police forces has fallen over the past decade
for all of the main drug types.

• The decline in referral pathways between the criminal justice system and community
treatment have coincided with the Home Office’s decision to cancel central funding with
the Drugs Intervention Programme in 2013, as well as disinvestment in drug treatment
services.

• The demise of Drug Action Teams and the Drug Interventions Programme over recent
years has contributed to the fragmentation of partnership working in relation to drugs at a
local level and this has particularly impacted on police engagement.

Source: Dame Carol Black, Review of Drugs: Summary (2020) 627

Outcomes
We now turn to what the data tells us about drug use by young people. The big picture,
for both 11–15-year-olds and 16–19-year-olds, is one of declining use from around 2003
until around 2014, followed by a rebound until the pandemic. This pattern applies both
for cannabis and Class A drugs.

https://www.gov.uk/government/publications/review-of-drugs-phase-one-report/review-of-drugs-
627

summary

198
The data on 11–15 year olds is from the SDDYP Survey, and is England only. Data on 16–
19-year-olds comes from the Crime Survey for England and Wales. Both of these series
were interrupted during the pandemic but have recently published new data covering
2021 to 2022.628 In the charts below, 2021/22 data is included as a dotted line.
Drug use by 11–15-year-olds
Drug use ever
As of 2003, around half of 15-year-olds had ever taken drugs. This proportion halved
between 2003 and 2014, then rose again after 2014. Figure DR3 shows this along with the
trends for other ages which also show a striking fall until the early 2010s, with some
increases thereafter. The post-pandemic figures show no further rise and some reduction.
Figure DR3: Trends in proportion of 11–15 year old pupils in England who have ever used
drugs

60

49
50

40
34
Per cent

30 28 27
24

20 18

11
10

11 yr old 12 yr old 13 yr old 14 yr old 15 yr old

Note: To measure drug use on consistent basis, the pre 2016 definition is used throughout this graph, ie
not including nitrous oxide and new psychoactive substances, where data is only available since 2016.
Source: SDDYP in successive years.

Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
628

England, 2021; Office for National Statistics, Drug Misuse in England and Wales: Year Ending June 2022 (ONS,
2022).

199
Last year use: cannabis, Class A
By far the most common drug for young people to have used is cannabis. Cannabis use
amongst under-16 school pupils halved in the decade after 2003 then edged up again
between 2014 and 2018.
Use of Class A drugs in the last year amongst under-16 school pupils also fell then rose
during the period we are studying. Class A use was on a broadly downward trend from the
mid-2000s until 2013 (from 9 per cent of 15-year-olds in 2005, to 4 per cent in 2013) then
rebounded after 2014.
Cannabis and Class A trends for this group are shown in Figure DR4 below. The most recent
data on this age group shows a reduction: these figures were collected between September
2021 and February 2022, when some Covid restrictions were still in place in England.
Figure DR4: Trends in proportion of 13 to 15 year olds in England who have used cannabis
or Class A drugs in last year

Cannabis use in last year by 13, 14 and Class A drug use in last year by 13, 14
15 year olds and 15 year olds
35 35
31
30 30

25 25
21
19.3
20 20
Per cent

15 15.8 15

9
10 7.5 10 7.0
4.0
5 5

0 0
1998

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

2020

2009
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008

2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021

13 year olds 14 year olds 13 year olds 14 year olds

15 year olds 15 year olds

Source: SDDYP successive years

Drug use by 16–19-year-olds

Data on 16–19-year-olds (England and Wales) shows similar trends to the data on
under-16s in England. For cannabis use in the last year there was a steep decline from
1997 to 2012/13 (albeit with interruptions) and a rise again between 2013/14 and
2019/20. The pattern for Class A drugs amongst 16–19-year-olds is bumpier, but overall
represents a decline from a peak in 1999 to a low in 2012/13, then an apparent upward
trend until 2018/19. The most recent data on this age group shows a large reduction for
these age groups: this data covers the period from October 2021 until June 2022. For part
of this period, some Covid restrictions were still in place in England.

200
Figure DR5: Trends in 16 to 19 year olds in England and Wales who have used cannabis or
Class A drugs in last year

Cannabis use in last year Class A drug use in last year

16 to 19 year olds 16 to 19 year olds
30 30
28.5

25 25

20 20
Per cent

15 15
14.4

10 10 8.0

5 5
3.5

0 0

2001/02
1997
1999

2003/04
2005/06
2007/08
2009/10
2011/12
2013/14
2015/16
2017/18
2019/20
2021/22
1997
1999
2001/02
2003/04
2005/06
2007/08
2009/10
2011/12
2013/14
2015/16
2017/18
2019/20
2021/22

Source: Crime Survey for England and Wales, Drug Misuse Appendix Tables, Table 3.03

Frequency of drug use by under -16s.

It is hard to tell from the charts above what happened to the proportion of young people
who use drugs regularly. Figure DR6 explores this question and shows very large changes in
the proportion of 15-year-old pupils who have taken drugs on more than ten occasions. For
15-year-old boys, this proportion fell from 14 per cent in 2005 to 4 per cent in 2014; for
girls the fall was from 11 per cent to 2 per cent. Proportions of both rose again between
2014 and 2018.
Figure DR6: Trends in 15 year old pupils in England who have taken drugs on more than
ten occasions, by gender
16
14
14

12

10 6.7
Per cent

11
8
4
6

4 5.6
2
2
0

Boys Girls
Source: SDDYP successive years

201
Inequalities
Distributional breakdowns of the trends in drug use are limited. The gender split in Figure
DR6 above shows girls’ use slightly below that of boys in most years. The same pattern is
seen in Class A drug use (not shown).629 Regional data has been published periodically and is
shown in Figure DR7 below. It shows broadly consistent trends across regions, but with a
large rise in London between 2014 and 2016.
Figure DR7: Trends in proportion of 11-15 year old pupils in England who took drugs in last
year by region

25

20

15
Per cent

10

0
2006-2008 2009 2010 - 2011-2012 2013 2014 2015 2016 2017 2018
2011
North East North West Yorkshire & the Humber
East Midlands West Midlands East of England
London South East South West

Source: SDDYP successive years

International context: how unusual are these trends?

There has been a general trend of falling adolescent substance use in developed countries in
the last two decades. However the charts that follow demonstrate that the fall in England
and the UK stands out, both for its very high initial starting point, and its scale and timing.
The ESPAD European survey provides comparisons across a wide range of drugs, although
sadly, for the UK, comparable figures only go up to 2007. These figures are illustrated in
Figure DR8 and show that UK drug use was much higher than the survey average in 1995,
but then narrowed the gap with the European average between 1995 and 2007. Figure DR9
shows more recent data for cannabis alone, this time for England, from the four-yearly HBSC
international survey of 15-year-olds. For both girls and boys, the gap between England and
the international average was initially large but then narrowed considerably between 2002
and 2006. For boys, the gap widened again in 2018.

Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
629

England, 2021. Tables 8.6d,e,g, and h

202
Figure DR8: Trends in drug use by 15 and 16 year old students in the UK and the ESPAD
international survey

30

25

20
Per cent of students

15

10

0
1995 1999 2003 2007 1995 1999 2003 2007 1995 1999 2003 2007 1995 1999 2003 2007 1995 1999 2003 2007
Lifetime use of inhalants Use of marijuana/hashish Lifetime use of illicit drugs Cannabis use at 13 or Lifetime use of
to get high in last 30 days other than marijuana or younger ecstasy
hashish

UK Survey average

Source: 2011 ESPAD Report: Substance Use among Students in 36 European Countries (EMCDDA, 2102) Tables 62, 63, 64,
65, and 68

Figure DR9: Trends in HBSC data on cannabis use by 15 year olds, England and Survey
average

Proportion of 15 year old boys who have Proportion of 15 year old girls who have
ever used cannabis ever used cannabis
England, and HBSC Survey average England, and HBSC Survey average
45 45
40 40
35 35
30 30
Per cent

25 25
20 20
15 15
10 10
5 5
0 0
2002 2006 2010 2014 2018 2002 2006 2010 2014 2018

England HBSC Survey average England HBSC Survey average

Source: Inchley et al, Spotlight on adolescent health and well-being. Findings from the 2017/2018 Health Behaviour in
School-aged Children (HBSC) survey in Europe and Canada. International report. Volume 2, and previous HBSC
international reports

203
Plausible explanations of the trends in England
There has been no dedicated study of the changes in young people’s drug use in England
over this time period. As has been suggested in other chapters of this report, explanations
of such striking changes are likely to be multicausal.
On the basis of the data and wider research literature, seven hypotheses are suggested as
plausible contributors.
Later age of first introduction to drugs
As Figure DR3 showed, experience of drugs declined steeply amongst secondary school
pupils in the 2000s, with signs of a small recovery since around 2014. This has parallels to
the falls in early onset of sexual intercourse and alcohol use described in Chapters 6 and 7.
This reduction in early onset of drug-taking is important for several reasons: it directly cuts
off several years of potential drug taking, reduces a risk factor for later dependence, and
changes the peer context so that teenagers became less likely to have friends who take
drugs. Any rise in early onset of drug use would be concerning for the same reasons.
Changed attitudes to drugs
The SDDYP survey shows that young people’s approval of drug use fell in the 2000s, with
signs of an increase again in during the 2010s. (Figure DR10)
Figure DR10: Trends in attitudes to drug use among 11-15 year old school pupils in
England
12

10

8
per cent

0
2002

2011
1999
2000
2001

2003
2004
2005
2006
2007
2008
2009
2010

2012
2013
2014
2015
2016
2017
2018
2019
2020
2021

OK to take cannabis once a week OK to sniff glue once a week

OK to take cocaine once a week

Source: Smoking Drinking and Drug use by Young People 2021

The decline then resurgence in approval of regular cannabis use is particularly striking and
would be expected to have an impact on behaviour. It is not it clear what caused the

204
attitudinal changes. It is possible that the hardening of young people’s attitudes towards
cannabis could have been driven by increased awareness of cannabis use disorder.630 Or
changing attitudes may have reflected a wider risk awareness which the government was
actively trying to encourage through drugs education, FRANK and related communications
campaigns, and cannabis reclassification.
It is worth noting that perceived parental attitudes did not change much over the period.
For the last twenty years, a consistent 99 per cent of pupils have said their parents would be
against them taking drugs.631
Reduced face to face socialising
Another possible contributor to lower drug use is the change in young people’s socialising
patterns during the 2000s to spend less time directly with peers (Figure OY9 in Appendix 1).
Two international studies using HBSC survey data have found links between reduced face to
face socialising and declining adolescent cannabis use, speculating that the link could be
through reduced opportunity to obtain drugs and social influences on their use. 632 The
possible reasons for this change, as discussed in Chapter 7, may include tougher alcohol
enforcement, increased school conscientiousness, and increasing time spent online by
young people. However research has found that that those who make most use of
electronic media communication are not less likely to drink or take drugs – in fact the
reverse was the case.633
Falls in smoking and drinking
Smoking and drinking are strongly associated with drug taking. Both smoking and drinking
declined significantly amongst young people from the mid-2000s onwards.634 The similarity
between these trends may be coincidental, or driven by common factors. There could also
be a causal link. For example, a young person who meets friends to smoke or drink may find
drugs available in the same setting. And recent US research has also found evidence that
adolescent nicotine exposure influences long-term molecular, biochemical, and functional
changes in the brain that encourage subsequent drug abuse.635 In the light of this, the
recent increase in the use of e-cigarettes is of concern. (See Figure OY1 in Appendix 1).

630
For a discussion of cannabis use disorder see Liz McCulloch, ‘Why Did Cannabis Treatment Presentations
Rise in England from 2004-2005 to 2013-2014?’, Drugs and Alcohol Today, 17.4 (2017), 218–31.
631
Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
England, 2021. Table 10.15.
632
Emmanuel Kuntsche and others, ‘Decrease in Adolescent Cannabis Use from 2002 to 2006 and Links to
Evenings out with Friends in 31 European and North American Countries and Regions’, Archives of Pediatrics
and Adolescent Medicine, 163.2 (2009), 119–25; De Looze and others.
633
De Looze and others.
634
For trends on drinking see Chapter 7, and for trends in smoking see Appendix 1.
635
Eric R. Kandel and Denise B. Kandel, ‘A Molecular Basis for Nicotine as a Gateway Drug’, New England
Journal of Medicine, 371.10 (2014), 932–43; Michelle Ren and Shahrdad Lotfipour, ‘Nicotine Gateway Effects
on Adolescent Substance Use’, Western Journal of Emergency Medicine, 20.5 (2019), 696–709.

205
Changes in school absence and exclusion
Chapters 4 and 5 of this report show that that the number of young people out of school
because of permanent exclusion and absence fell steeply until 2013, before starting to rise
again. Could this help to explain changes in young people’s drug use?
The SDDYP survey captures some data on the matter, as it asks children to say if they have
ever been excluded or played truant, and significant numbers of children answer that they
have. This tells us two things (Figure DR11). First, as would be expected, between 2006 and
2013 progressively fewer pupils in mainstream schools said that they had experience of
exclusion or truancy, but after 2014 the numbers started to rise again. Second, the extent to
which this group used drugs also changed – having started at very high levels, it declined
from 2003 to 2014, then edged up again.
These graphs mean that between around 2006 and 2014, progressively fewer pupils in the
SDDYP survey reported that they had experienced exclusion or absence, and fewer of those
who did were using drugs regularly. Then after 2014, the trends reversed: more pupils
started experiencing exclusion or absence, and more of them were regular drug users.
Figure DR11: Trends in drug use by 11-15 year old pupils who have truanted or been
excluded, England

Proportion of 11 to 15 year old pupils who Proportion of 11 to 15 year old pupils who
have truanted or been excluded usually take drugs at least once a month, by
as a proportion of all respondents whether ever truanted /been excluded

30% 25

21
24.2%
25%
20

18.9%
20%
16.1% 15
Per cent

15%
9
10
8
10%
6

5
5%

0% 0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021

2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021

Truanted or excluded
Never truanted or excluded
All pupils
Source: SDDYP 2021, Table 8.21

206
This could be interpreted several different ways. Levels of drug use could be contributing to
change in the rates of exclusion and school absence. Alternatively, or additionally, the link
could go in the other direction with changes in exclusion and absence rates influencing
patterns of drug use.
Treatment
The discussion of government policies earlier in this chapter highlighted the expansion in
specialist substance misuse treatment for young people in the 2000s, and its subsequent
decline. Given what we know of treatment effectiveness, the expansion of treatment
capacity is likely to have contributed to the overall fall in drug use, as well as helping to
reduce drug-related offending.
The number of under-18s engaged in treatment was not enormous - 24,000 in treatment in
the peak year of 2008. So treatment of under-18s cannot on its own be responsible for a
reduction in drug use across the adolescent population as a whole. However, to this number
one should perhaps add the number of over-18s in treatment in the adult system. In
2008/09 there were just over 30,000 18-24 year olds and over 44,000 25-29 year olds in
substance misuse treatment in England.636 With almost 100,000 under-30s in treatment in
England in one year, a significant number of teenagers and young adults were being helped
to reduce their substance use, which could plausibly have had a wide effect, not least
through declining prevalence of drug use among older siblings and older friends.
Availability
Finally, we should look at the question of drug availability. The SDDYP Survey tells us that
between 2003 and 2011 the proportion of 13-, 14- and 15-year-olds who thought it would
be easy to obtain drugs fell by about 10 percentage points, and the proportion who had
been offered drugs fell by even more. These proportions then increased slightly in the
2010s.
These trends are shown in Figure DR12. They are important because part of young people’s
drug use is opportunity-driven. Consistently about ten per cent of pupils who have taken
drugs say they first took them ‘just because I was offered’, and most young people who use
drugs obtain them from a friend of their own age or an older friend.637 These changes in
availability may reflect both the trends in drug use amongst peers, and penetration of drugs
in the community generally. The downturn in availability in 2021 is likely to reflect
increased time spent at home during the pandemic and reduced socialising.

636
National Treatment Agency, Statistics from the National Drug Treatment Monitoring System 2010-2011
(NTA, 2011).
637
SDDYP 2021. Table 9.1 and Table 10.1.

207
Figure DR12: Trends in proportion of 11-15 year old pupils in England who have been
offered drugs or think they would be easy to obtain, by age

Have ever been offered any drugs, Perceived it 'very easy' or

by age 'fairly easy' to get illegal drugs, by age
70 70

60 60

50 50

Per cent
Per cent

40 40

30 30

20 20

10 10

0 0

2006

2009

2012
2001
2002
2003
2004
2005

2007
2008

2010
2011

2013
2014
2015
2016
2017
2018
2019
2020
2021
2002

2011

2020
2000
2001

2003
2004
2005
2006
2007
2008
2009
2010

2012
2013
2014
2015
2016
2017
2018
2019

2021

11 years 12 years 13 years 11 years 12 years 13 years

14 years 15 years 14 years 15 years

Source: SDDYP successive years

Discussion
The trends shown in adolescent drug use described in this chapter show a marked similarity
with many of the other indicators in this report. Between the mid-2000s and 2013 or 2014,
the proportions of young teenagers experimenting with drugs or taking them regularly fell
by at least half. Levels of adolescent drug use in England moved closer to the international
average. After 2014, progress stalled and some indicators worsened, but the first set of
post-pandemic figures are lower than those for 2018.
As is the case for other forms of adolescent disadvantage discussed in this report, the
striking trends we observe in adolescent drug use are likely to be multi-causal. During the
period that adolescent drug use was falling, a set of significant related factors were
changing in a direction likely to reduce drug use. Young people became significantly less
likely to try drugs in their early teens. Young people’s attitudes to drugs became more
negative. There was a decline in the prevalence of adolescent drinking and smoking,
behaviours closely associated with drug use. Going out with friends in the evening became
less common. The number of young people persistently absent or excluded fell. Availability
of drugs seems to have reduced so that young people were less likely to be offered drugs or

208
to know where to get them. Treatment for young people with drug problems became more
available. These changes are likely to have been mutually reinforcing.
During the 2010s, many of the factors that were restraining drug use started to move in the
opposite direction. Early onset of drug use became a little more frequent, attitudes to drugs
became less cautious again, exclusion and absence rose again, treatment availability fell, a
new model of drug dealing developed that directly recruited vulnerable young people, and
young people became more likely to be offered drugs again. Again, these changes may well
have been mutually reinforcing, but this time in an adverse direction.
Some, but not all of these changes, can be traced back to policy and the decisions of
successive governments first to prioritise prevention, treatment and policing, and then to
deprioritise them. De-prioritisation may seem a harsh word for the Coalition’s policies: at
the time they were probably thought of as devolution, and with so many of the indicators
going in the right direction in the early 2010s, drugs may have been thought of as
yesterday’s problem. But given the harm that drugs can do, and the money to be made by
those who sell them, a more cautious approach would have been wiser. At the very least,
the impact of the decision to devolve both drug treatment and crime management to local
areas should have been monitored. Lack of policy ownership and focus at national level, and
the decision to drop annual monitoring of young people’s drug use, meant that the situation
deteriorated significantly in the second half of the 2010s without a meaningful government
response.
Data gathered in 2021 show some moderation in young people’s drug use, but it is too early
to say whether that is a temporary effect of reduced social contact during the pandemic.
Other impacts of the pandemic, such as the rise in school absence, and the level of mental
health need amongst young people, represent possible risk factors for future drug use. In
terms of the policy prospect, much hangs on the resourcing and implementation of the new
strategy announced in 2021.
Whatever happens to policy and spending, much better monitoring of trends in young
people’s drug use is surely essential. Annual surveys of young people’s drug use would take
some of the guesswork out of this important area, and allow the government to respond
more quickly to adverse trends. Just as importantly, the government should commission
regular research on drug use amongst the young people systematically under-represented
in our information sources – those who are absent from school, unplaced, excluded, or in
alternative provision. Acknowledging and measuring the scale of drug use amongst all young
people is a vital first step to designing effective policy and avoiding both great misery and
significant cost to the taxpayer further down the line.

209
Chapter 9 Conclusion
This report has explored the record of policy and outcomes for adolescent disadvantage
across a period of just over two decades. Its focus has been on a set of particularly
damaging and expensive forms of disadvantage which were explicitly prioritised by the
Labour government. The report has followed the story through two decades, under Labour
and its Coalition and Conservative successors, to track the evolution of policy and the trends
in outcomes.

The policy story

The report recounts sizeable shifts in multiple social policies affecting young people, both
after 1997 when Labour took over from a Conservative government, and in the 2010s when
the Coalition and Conservative administrations in turn came to office. During the first period
there was a major increase of activity as Labour expanded services, addressed cross-cutting
issues and service inequalities, and put more money in the hands of poor families. After
2010, the focus was on deficit reduction and austerity across government spending,
alongside reducing central government intervention and making structural changes to the
school system.

The outcome story

The report also recounts sizeable changes in outcomes. Over Labour’s time in office as a
whole, the key indicators of youth disadvantage discussed in this report improved
substantially and many continued to improve during the first few years of the Coalition.
Then in the mid-Coalition period, between 2012 and 2014, many of these indicators began a
period of stalling and, in some cases, deterioration.
The outcome trends for eight key indicators of adolescent disadvantage are illustrated in the
panel of charts at Figure S1 below. The graphs show that at the end of Labour’s term all
eight were substantially lower than at the beginning, with some indicators having halved.
Many saw steep falls virtually in parallel over the decade from 2003 to 2012. (This period is
shaded green in the graph).
The picture is not uniform. Two indicators (10-17 year olds entering the criminal justice
system, and secondary pupils missing over 50 per cent of schooling) show a steep fall from a
slightly later starting point, around 2007. In the case of 10-17 year olds entering the
criminal justice system, 2007 marked a change in trend (discussed in Chapter 2). For
secondary pupils missing over 50 per cent of schooling, (‘severe absence’), we do not know
the earlier trend, as data were not collected before 2006/07.
For child poverty, the improvement was concentrated in the early part of Labour’s term. The
measure shown is the fixed real income threshold, also known as absolute poverty: by 2010-
11, the proportion of 11–18-year-olds in poverty was 17 percentage points lower than in
1996-97. A small rise and fall over the second decade left the numbers one percentage point
lower overall in 2018-19 compared with 2010-11.

210
 0.0%
0.2%
0.4%
0.6%
0.8%
1.0%
1.2%
1.4%
1.6%
1.8%
10%
20%
30%
40%
50%
60%

0%
1997/98
1998/99 1996-97
1999/00 1997-98
2000/01 1998-99
2001/02 1999-00
2002/03 2000-01
2003/04 2001-02
2004/05 2002-03
2005/06 2003-04
2006/07 2004-05
2007/08 2005-06
2008/09 2006-07
2009/10 2007-08
2010/11 2008-09
2011/12 2009-10
2012/13 2010-11
2013/14 2011-12
constant prices, AHC)

2014/15 2012-13
2015/16 2013-14
2016/17 2014-15

50 per cent of schooling

Abolute poverty: 11 to 18s
(below 60 % of 2010/11 median in

2017/18 2015-16
2018/19 2016-17

Secondary pupils missing over

2017-18
2018-19

10
20
30
40
60

50

0
%

10
30
40
50
70

20
60

0
rate/
1,000
1998 1997/98
1999 1998/99
2000 1999/00
2001 2000/01
2002 2001/02

14 year olds
2003 2002/03

15 year olds
2004 2003/04
2005 2004/05

16 year olds
2005/06
2006
2006/07
2007
2007/08
2008
2008/09
2009 mother 2009/10
2010
2010/11
2 by age 16

2011
2011/12
2012
2012/13

17 year olds
2013
2013/14
2014
2014/15
2015
2015/16
Figure S1: Long term trends in adolescent outcomes - summary

2016
2016/17
Proportion not attaining Level

2017
2017/18
Teenage conceptions by age of

2018
2018/19
2019

10
20
30
40
50
60

0
%

211
100,000
120,000

0
20,000
40,000
60,000
80,000

1997
1998
1997/98
1999 1998/99
2000 1999/00
2001 2000/01
2002 2001/02
13 year olds

2003 2002/03
2004 2003/04
2004/05
14 year olds

2005
2006 2005/06
2007 2006/07
2007/08
2008
2008/09
2009
2009/10
2010 2010/11
2011 2011/12
15 year olds

2012 2012/13
2013 2013/14
2014 2014/15
10-17 year olds entering

drank alcohol in last week

2015 2015/16
13, 14 and 15 year olds who

2016/17
criminal justice system (E&W)

2016
2017 2017/18
2018 2018/19
10
12
14

0
2
4
8

6
%

0.05
0.15
0.25
0.35

0
0.2
0.3
0.4

0.1

1998
1999 1997/98
1998/99
2000
1999/00
2001
2000/01
2002
2001/02
2003
2002/03
2004 2003/04
2005 2004/05
2006 2005/06
2007 2006/07
2008 2007/08
2009 2008/09
2010 2009/10
2011 2010/11
in last month

2012 2011/12
secondary schools

2013 2012/13
Green shading is decade 2003 to 2012. Refer to chapters for sourcing. Alcohol figures are from Smoking Drinking and Drug use by Young People 2018, Table 5.5

2014 2013/14
2014/15
Permanent exclusion rate,

2015
2016 2015/16
2016/17
11-15 year olds who took drugs

2017
2017/18
2018
2018/19
The trends did not break abruptly with a switch of government. But after the mid-Coalition
period, things started to change for many indicators. After 2011/12, Level 2 attainment at
16 began to deteriorate and the cohorts who turned 16 in the next three years each had
slightly more young people not achieving Level 2. Indeed, despite a slight improvement later
in the decade, Level 2 attainment had still not recovered its 2011/12 level by 2018/19. The
trends in permanent exclusion, teenage drug use, and severe absence all bottomed out
between 2012 and 2014 then started to rise again. After 2014, alcohol use probably
plateaued, taking all indicators into account. Only two indicators continued to fall
consistently between 2010 and 2019 – teenage conceptions, and 10-17 year olds entering
the criminal justice system. It is not clear how far the latter reflects changes in offending, as
against changing criminal justice practices and resourcing.

A changing picture on inequalities

During the period of improving outcomes, there are some striking example of disadvantaged
groups and areas improving faster than the average.

• Pupils eligible for free school meals and pupils with special educational needs
narrowed the gap with other pupils in terms of Level 2 attainment, school absence,
and permanent exclusion.
• The deprivation gap in teenage pregnancy also narrowed.
• Black Caribbean pupils had been heavily overrepresented in permanent exclusions
and amongst those not achieving level 2 by age 16, but saw some of the largest
improvements until 2011/12.
• White pupils were the ethnic group with the highest drinking prevalence in 2003,
and then saw the biggest reduction by 2014.
• Regionally, London moved from having the highest rates of both teenage
conceptions and school absence to being among the lowest. The beginnings of this
shift were evident early in the first decade, alongside an improvement in secondary
school attainment.
• Some North-South inequalities began to narrow, as the North East had the greatest
fall in alcohol-related hospital admissions for under-18s and, along with Yorkshire
and Humberside, saw the largest increase in level 2 attainment between 2001/02
and 2011/12.
After 2012, some of these inequalities started to widen again.

• Increases in permanent exclusion were sharpest for pupils eligible for free school
meals, pupils with special needs but no statement or education and health care plan,
and pupils of Black Caribbean, Mixed White/Black Caribbean, and Gypsy Roma
ethnicities. The increases in exclusion were also particularly pronounced in the North
East, and North West.
• Pupils eligible for free school meals, pupils with special needs, and pupils of Black
Caribbean and Mixed White and Black Caribbean ethnicity all saw large increases in
overall absence and the regions with the largest increases in persistent absence after
2013/14 were the North East and Yorkshire and Humberside.

212
 • The deterioration in Level 2 attainment at 16 was also disproportionate for pupils on
free school meals, and those with special educational needs but no statement or
education and health care plan, and the North East was the region with the largest
deterioration between 2011/12 and 2017/18.
• During the decade, the attainment gap in secondary schools has stopped narrowing
on all the key measures.
• Black Caribbean pupils saw a 12.9 percentage point decline in their Level 2
attainment between 2012 and 2019.

Change between cohorts

All this means that there was a striking change in prevalence of several forms of adolescent
disadvantage over a very short space of time. This is particularly marked for the generation
who turned 16 around 2012 and 2013. Compared with their predecessors born five or ten
years earlier, this generation experienced much lower levels of teenage drug taking, alcohol
use, absence and exclusion from school, as well as lower rates of entry into the criminal
justice system. They also had fewer unwanted conceptions, better Level 2 attainment at Age
16, and higher staying-on rates in school. On some of these measures, this group of young
people also fared better than the generation which came after them. This striking data
about the teenage years of this generation create the possibility of a cohort effect which
may well become visible in a range of other outcomes in later life.

International comparisons
International comparisons underscore the importance of the changes described. During the
2000s, England became less of an outlier in teenage birth rates; had the largest fall in
weekly drinking for both boys and girls, out of 36 European countries; and UK levels of
adolescent drug use came much closer to international averages. Between 2006 and 2014
the UK caught up with the OECD average educational participation rate for 15-19 year olds,
closing an eleven percentage point gap. A study of 23 countries between the mid-1990s and
2010 found that only eight of the countries covered reduced child poverty, with the UK
seeing the largest reduction. However, between 2013 and 2018 the United Kingdom had the
third largest increase in relative child poverty out of over 30 comparator countries. Between
2014 and 2018, cannabis use by boys in England started to diverge from the international
average again, and since 2014 adolescent drinking in England has not matched the further
falls seen in other countries.

How can we explain the trends?

These striking trends in overall outcomes and in their distribution are intriguing and call for
explanation. But there is no research method that can determine the causes with certainty.
The data this report examines is not the output of a controlled experiment, but instead
covers the entire adolescence of millions of young people over two decades, whose
childhood and adolescence were affected by multiple policy changes and many other
factors. This is an important caveat to any attempt to explain these trends. But it does not
mean that we cannot marshal the evidence we do have, and use that to consider plausible
hypotheses that might explain what has happened.

213
Having done this, my assessment is that the pattern of outcome trends this report examines
is likely to have been caused by a combination of exogenous social changes, direct policy
impacts, and wider reinforcement effects between different areas of young people’s lives.
Social, demographic, or technological change?
To take these in turn, there is clearly some role for social, demographic, and indeed
technological change in some of the trends observed. But, as discussed in the individual
chapters, exogenous factors of this kind cannot on their own explain the magnitude of the
outcome changes that have occurred, or their distribution, or their stalling. To take one
example, while it is likely that the changing ethnic mix of the youth population has made
some contribution to falling drinking levels, this cannot account for the scale of the fall, the
change that has been seen in locations of drinking and how alcohol is obtained, the fact that
the fall was greatest among white young people, or the stalling of the change after 2014. To
take a different example, it is clear that the extraordinary growth of mobile phone
ownership and use has played a part in changing patterns of socialising among young people
that may have impacted upon the use of alcohol, drugs and sexual behaviour. But many
other factors appear likely to have played a part, and research cited in the chapters finds
that those who communicate most online are more likely than others to use alcohol and
drugs.
Policy?
The report suggests instead that the changes in the outcomes discussed are multi-causal,
with policy changes likely to have been part of the explanation. The report has drawn on a
wide range of analysis to reconstruct the policy picture, including government-
commissioned evaluations where they exist, the National Audit Office’s archive of value for
money reports, the work of parliamentary select committees, inspectorates, and bodies
such as the Office of the Children’s Commissioner, and a wide range of independent
research from this country and abroad. These help to reconstruct the picture of what was
done at different points in the period, and the inputs and outputs that were delivered. The
evidence varies in its rigour, and many policies were not evaluated to the most robust
standard (or even evaluated at all).638 Nonetheless, there is substantial evidence in every
chapter suggesting that government policy initiatives, in different ways and at different
times, are likely to have contributed directly to the changes in outcomes that have occurred,
both when they were improving, and when they deteriorated.
Reinforcement effects?
Then we have the question of indirect and reinforcement effects. This is where the research
literature becomes rather scarce. By this I mean that the debate about the impact of past
policy on real world outcomes tends to be conducted within the confines of each policy and
academic discipline. It is, of course, widely known that risk factors and adverse outcomes
tend to cluster together, and the literature on ‘what works’ rightly stresses the importance
of multi-component programmes to address adolescent risk behaviours.639 Despite this,

638
Jill Rutter, Evidence and Evaluation in Policy Making: A Problem of Supply or Demand? (Institute for
Government, 2012); National Audit Office, Evaluating Government Spending (NAO, 2021).
639
Jackson and others; Hale, Fitzgerald-Yau, and Viner; Fitzsimons and others.

214
there is not a great deal of research looking into whether these ‘crossover’ and
reinforcement impacts have actually occurred in the recent policy environment, for example
through changes in educational participation affecting changes in crime, or falling teenage
alcohol use impacting on the teenage conception rate.
In the individual chapters, I draw attention to several possible reinforcement effects of this
type. Chapter 7, for example, identifies nearly a dozen factors across multiple policy areas
which all, for a period, were moving in a direction likely to reduce adolescent drinking, and
could have been very powerful in combination. Many other chapters end with a similar list.
At the aggregate level, the trends shown in Figure S1 are consistent with the interpretation
that there were multiple mutually reinforcing changes in young people’s environment,
which together caused them to go through adolescence in a very different way from their
predecessors.
Cohort effects?
This brings us back to the possibility of a cohort effect. The young people whose outcomes
improved so sharply in the 2000s went through childhood and adolescence in an era of
falling family poverty and rising school resourcing, and some of them will have benefitted
from Sure Start in their early years. Growing up, they were exposed to lower levels of crime,
and their peer groups engaged less in alcohol and drug use. In secondary school they
benefited from a very different 14-19 curriculum, with much more support to help them
stay in learning after age 16. These changes in their environment could easily have had
substantial effects on their performance in education as well as on their involvement in risky
behaviours.
The generation of young people who turned 16 in the early coalition years were also
subject to multiple policy changes compared with their predecessors. Many will have been
affected by changes to benefits; young people were no longer able to study vocational
qualifications to the same extent; spending on mental health, youth services, and drug
services was falling; Connexions and the Education Maintenance Allowance had been
abolished; and virtually all frontline services that young people might need to draw on were
going through a time of funding pressure and organisational turbulence. Looked at in that
context, the slowing and stalling of progress from mid-Coalition onwards becomes easier to
understand, as a cumulative result of multiple policies whose likely interaction appears not
to have been recognised in advance.
Policy is not just about spending
Before we leave the question of policy changes, one last point needs to be made. Policy is a
very broad term, and as will be obvious from the detailed chapters, it covers many different
dimensions of the way governments operate. These include the priorities that governments
set; the areas they decide not to prioritise; how these priorities are reflected in spending
budgets; the design of incentives for institutions, markets and individuals; collaboration and
accountability structures; national/local relationships; and the degree of interest in tracking
performance. The analysis in the detailed chapter suggests that all of these are important,
although they do not all attract the same attention in political debate.

215
This means, to be blunt, that in weighing the impact of government policies we need to look
at more than just spending levels. This is an important point to recognise in relation to the
three administrations we are discussing. Labour did indeed raise spending on many social
programmes – both to expand existing services and to create whole new spending
programmes in areas such as the Education Maintenance Allowance, the Connexions
service, or the teenage pregnancy strategy. But some of its policies cost very little – for
example, increasing the enforcement of alcohol licensing, giving parents clear guidance on
adolescent drinking, making contraception services more youth-friendly, and giving young
people clear information on drugs. One policy – increasing alcohol duties – actually raised
money as well as contributing to public health.
Just as Labour’s policy changes cannot all be categorised simply as spending increases, the
Coalition and Conservative governments’ policy changes cannot all be bracketed under the
umbrella of deficit reduction. Certainly, austerity loomed large across many youth spending
areas. But many Coalition and Conservatives policies cost money rather than saving it (for
example the academy and free schools programmes, and the repeated reductions in alcohol
duty). And many of the most significant policy changes under the Coalition and
Conservatives were structural and design changes which sprang from concerns other than
austerity - for instance, the desire for a smaller state, the preference against regulation, the
desire to reduce local authorities’ role in running schools, and the view that secondary
education should be more traditionally academic.640

Lessons and recommendations for the future

This consideration of policy and outcomes on adolescent disadvantage offers many

potential lessons for the future. But those lessons will have to be applied in a new context
that is made much more difficult by the experience and aftermath of the pandemic, and the
state of the economy.
There are many issues competing for political attention in 2023. However, policy on children
and young people should be very high on the list. I would argue that this should always be
the case. But it is particularly important at present because the factors that are new in our
current context - the educational, social and mental health issues which young people have
experienced during the pandemic, and the likely impact of the cost of living crisis on
vulnerable families – have been overlaid on a system that was already struggling in many
aspects.
The detailed chapters of this report offer plenty of insight into specific elements of policy
that need attention. Some of them – drugs, for instance - are already the subject of new
government initiatives, described in the detailed chapters, which one hopes will be
adequately resourced and implemented with persistence. In other areas, such as special
needs, and school absence, the government appears to be becoming more active, although

640
Ruth Lupton, Tania Burchardt, and others, Social Policy in a Cold Climate: Policies and Their Consequences
since the Crisis (Bristol: Policy Press, 2016).

216
the jury is out on the adequacy of what is planned. But these examples I pick out are just a
few of the items on a potentially long ‘to do list’ of youth policy issues to tackle.
In making some closing recommendations I will focus on a short list of overarching lessons
for policy makers which I believe are relevant to all the youth policy areas discussed in this
report.
i. Joining-up and leadership at national level: the first lesson is about national leadership
and coordination. Individual youth policy areas will be easier and cheaper to tackle if
the government does it in a coordinated way. This will maximise the synergies between
policies, help to avoid perverse incentives, reduce the scope for gaps and duplication,
and help develop preventive investments. These points have been powerfully made in
several recent reports in relation to children and young people.641 I strongly agree with
their arguments that underplaying the government leadership role is a false economy.
ii. Local coordination: Neglecting the cross-cutting role at local level is also a mistake. It is
easy for Whitehall to rearrange (or remove) the requirements on local partners to
collaborate, but these apparently bureaucratic changes can send powerful signals. It
matters enormously whether local services are encouraged to work together, or not.
Central government should take a more consistent and constructive interest in this,
and above all needs to satisfy itself that there are enough frontline staff, with enough
time, to support young people and their families who are at risk of, or already
experiencing, serious difficulties. There are many different models for this –
Connexions was one model in the past, and the Commission on Young Lives idea of a
‘Sure Start for teenagers’ is another possible model.642 Currently, however, capacity of
this kind is fragile and overstretched in many parts of the country. It is in central
government's interests to develop a sustainable model for the future. Indeed, it is very
hard to see how effective or lasting solutions can be found to the large number of
children missing school, the scale of criminal exploitation, or the number of 16- to 17-
year-olds whose activity is unknown, without a well-resourced and visible frontline
resource taking on such a role on a continuing basis.
iii. More emphasis on outcomes with better use of data and evidence: As this report has
shown, a great deal of data is already available, most of it produced by the government
itself. But it does not appear that government is currently bringing this picture
together at the aggregate level to track the overall outcome picture.643 Good policy
would be served by an investment in collating this data for the public and policy
makers, and by investment in filling the key data gaps.644 Taking this forward would
support the NAO’s recommendation that central government needs ‘a shared strategic,
data-led view of the complex set of problems it is trying to address, and a strong
evidence base to determine the most efficient and effective ways of addressing them’.
645

641
Commission on Young Lives; House of Commons Committee of Public Accounts.
642
Commission on Young Lives.
643
Public Accounts Committee.
644
Amongst the data gaps mentioned in the report are: restoring annual drug and alcohol surveys of young
people; reinstituting self-report crime surveys; gathering more data on hard-to-reach groups; and tracking
trends in the numbers of young people experiencing multiple disadvantage.

217
 iv. Urgency: Above all, the analysis in this report points to the need for greater urgency
in addressing the problems of youth policy. In researching the policy context across
so many different fields, I have been struck by the number of policy areas where the
response to problems has been very slow to appear, or where actions are promised
but seem to take years to deal with. One can understand the extraordinary pressures
that Ministers and civil servants have been under in recent years. But children only
get one childhood. While the clock ticks on, it is the frontline and children
themselves who are left trying to cope in the system as it is.

There is, without doubt, a demanding agenda to tackle. But this report can end on a more
hopeful note. Although there are many areas of youth policy that need attention, and the
challenges are daunting, the ultimate message of this report should offer encouragement
to those faced with these issues. Significant change in young people’s outcomes has been
achieved in our own recent past. Those issues appeared entrenched when efforts to tackle
them first began. But it turned out that there were many things that could be tried, and
many of them appear to have worked. Targets that some thought fanciful were, in many
cases, achieved. Forms of disadvantage that were thought to be intractable turned out not
to be. Many young people overcame challenges that had impeded previous generations.
These lessons of the past can offer encouragement for the future, at a time when effective
youth policies are much needed.

645 National Audit Office, Support for Vulnerable Adolescents.

218
Appendix 1 Other Youth Indicators
This section contains data on some wider issues affecting young people which the reader
may wish to refer to as background to the main chapters.
The charts and brief explanatory notes cover the following issues:

• Health behaviours
• Obesity and being overweight
• Mental health
• Wellbeing
• Ethnicity
• Leisure time and face-to-face socialising
• Technology use
• Experience of being bullied
• Involvement in fighting
• Children as victims of crime
• Self harm
• Suicide

Several of the sections draw on published data from the international Health Behaviour
in School-aged Children survey (HBSC) which produces both national and cross-national
reports.646 Use is also made of comparisons between the two cohorts of the Longitudinal
Survey of Young People in England (LSYPE): these cohorts began in 2004 and 2013
respectively, with 13–14-year-olds.647

646
The 2018 cross-national reports are at https://hbsc.org/publications/reports/spotlight-on-adolescent-
health-and-well-being/
647
More detail can be found at https://closer.ac.uk/study/lsype-2/

219
Health behaviours
Smoking and use of e-cigarettes
Smoking
Information on smoking by school pupils is collected in the Smoking Drinking and Drug Use
Survey, the same source as much of the alcohol and drug data in this report. Figure OY1
below shows the proportion of 11–15-year-old pupils who smoke regularly. This fell slightly
between 2000 and 2006, then more sharply, the rate for girls more than halving between
2006 and 2012. For almost all of this period more girls than boys were regular smokers.
There has been a continued decline since then to just one per cent in 2021.
E-cigarettes
Figure OY1 also includes data on e-cigarette use, which has been collected since 2014. The
number of regular users rose from low levels to reach 3.5 per cent of 11–15-year-old boys
and five per cent of 11–15-year-old girls by 2021. The figures for 15-year-olds are much
higher: in the 2021 survey, 9 per cent of 15-year-old boys and 12 per cent of 15-year-old
girls said they were regular e-cigarette users.648
Figure OY1: Trends in proportion of 11–15-year-old pupils who are regular smokers or e-
cigarette users, England
14

12

10

8
Per cent

Regular e-cigarette users: boys Regular e-cigarette users: girls

Regular smokers: boys Regular smokers: girls

Source: Smoking Drinking and Drugs use by Young People 2021, Tables 1.3 and 4.2

Overlapping use of alcohol, drugs and smoking

The Smoking Drinking and Drug Use survey collects information on the overlaps between
the use of drugs, alcohol and smoking. This information goes back as far as 2005. The
figures for 15-year-olds are set out in Figure OY2 and show that the proportion recently
using both drugs and alcohol fell from 15 per cent to 4 per cent between 2005 and 2014, but
has seen some rebound since then.

Health and Social Care Information Centre, Smoking, Drinking and Drug Use among Young People in
648

England, 2021. Table 4.3.

220
Figure OY2: Trends in proportion of 15-year-olds who have recently engaged in two or
more of alcohol use, drug use, and smoking, England

Physical activity
The World Health Organisation recommends that young people engage in at least one hour
of moderate physical activity per day. Very few 11- to 15-year-olds in England meet this
threshold, and, as Figure OY3 shows, the proportion has fallen since 2002: the fall has
been greater for boys, but girls remain less likely than boys to meet the recommended
level.
Figure OY3: One hour’s moderate physical activity per day: 11- to 15-year-olds, England
recommendation
35
of one hour per day of moderate physical activity, England

29
30 28

25 23
22
Per cent of pupils

19
20
16
15 15
14
15 12

10

0
2002 2006 2010 2014 2018
One hour daily, boys One hour daily, girls

Source: Brooks et al, HBSC England National Report: Findings from the 2018 HBSC study for England

221
Obesity and being overweight
The Health Survey England publishes data on 11–15-year-olds who are overweight or obese.
These figures have fluctuated, at a high level, as illustrated in Figure OY4. The highest figure
during this period was in 2004, falling back to 32 per cent by 2010. The 2019 figure was 39
per cent.
Figure OY4: Trends in proportion of 11–15-year-olds who are overweight or obese,
England

50
45 42
39
40
35 32
30
30
Per cent

25
20
15
10
5
0

Source: Health Survey England, 2019

Mental health conditions

Long-term trend information on the prevalence of mental health conditions in England
relies on a survey which has been conducted only infrequently. The Mental Health of
Children and Young People Survey provides data on 11- to 15-year-olds for three individual
years within the scope of this report: 1999, 2004, and 2017and these are shown in Figure
OY5
The ONS commentary on the 2017 Survey noted that for 5-to 15-year-olds over this period:

• the prevalence of mental disorders overall, of any type, has increased

• young people have become more likely to experience emotional disorders

• there are indications that rates of behavioural disorder in boys aged 11 to 15 have
fallen.649

Katharine Sadler and others, Mental Health of Children and Young People in England, 2017: Trends and
649

Characteristics (NHS Digital, 2018). p30.

222
Figure OY5: Trends in any mental disorder and specific disorders in 11- to 15-year-olds by
sex, 1999, 2004, 2017, England

Boys Girls
16 16
14.2
14 14 13.0

12 12

10 10

8 8

6 6

4 4

2 2

0 0
1999 2004 2017 1999 2004 2017

Any disorder Anxiety disorders

Source: Mental Health of Children and Young People in Depressive disorders Behavioural disorders
England, 2017, Table 3
Hyperactivity disorders Less common disorders

Wellbeing
Happiness with different aspects of children’s lives
Data on children’s happiness with various aspects of their lives has been collected for several
decades by the British Household Panel Survey, succeeded by Understanding Society.
Figure OY6 below sets out trends on three aspects of children’s happiness - with life as a
whole, with family, and with school. Average happiness scores for life as a whole and school
were significantly lower in 2018/19 than in 2009/10.650
A disaggregated analysis of data for England published by the Department for Education
showed that within the overall fall in average ratings of wellbeing by 10–15-year-olds
between 2009/10 and 2016/17, the deterioration for 13- to 15-year-olds was slightly larger
than for 10- to 12-year-olds. 651

650
The Children’s Society, The Good Childhood Report 2021 (The Children’s Society, 2021).
651
Department for Education, State of the Nation 2019: Children and Young People’s Wellbeing (DfE, 2019).

223
Figure OY6: Trends in young people’s happiness with different aspects of life, UK

9.5

9
Mean happiness score (0 to 10)

8.5

7.5

Life as a whole Family School

Source: Children's Society, Good Childhood reports for 2020 and 2021. Figures up to 2009 from BHPS, and cover 11- to
15-year-olds. Figures from 2009/10 onwards from Understanding Society, and cover 10- to 15-year-olds.

Perceptions of school
Data from the HBSC school survey can give us more insight into trends in young people’s
experience of school in England, albeit collected only at four-year intervals.
The panel of charts at Figure OY7 shows trends on six indicators of school perception which
have been collected in England for part or all of the period 2002 to 2018. They show an
overall picture of improvement in positive perceptions up to 2014 (in some cases measured
back to 2002, in some cases only measured back to 2010), b ut a decline in all the positive
perception indicators between the 2014 and 2018 surveys.

224
Figure OY7: Trends in perceptions of school, 11- to 15-year-olds, England

Feel they belong in school Feel safe in school Feel teacher cares about them as a person
100 100 100
76
80 80 80
60
60 74 60 60
Per cent

40 52 40 40

20 20 20

0 0 0
2002 2006 2010 2014 2018 2002 2006 2010 2014 2018 2002 2006 2010 2014 2018
Boys Girls Boys Girls Boys Girls

Feel other students are kind and helpful Like school a lot Feel a lot of pressure
100 100 100

80 80 80

60 60 60
Per cent

38
40 40 30 33 40
20 23
36 32
20 20 20
25
18
0 0 0
2002 2006 2010 2014 2018 2002 2006 2010 2014 2018 2002 2006 2010 2014 2018
Boys Girls Boys Girls Boys Girls

Source: Brooks et al, HBSC England National Report: Findings from the 2018 HBSC study for England

225
Ethnicity
During the last two decades, the ethnic composition of the secondary school population
has changed. Figure OY 8 below shows the picture at 5-year intervals since 2003, which is
as far back as a consistent series can be tracked.
Figure OY8: Secondary school pupils by broad ethnic minority group, selected years,
England

100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%
2003 2008 2013 2018

White Mixed Asian Black Chinese Any other ethnic group Unclassified

Source: DfE, School, pupils and their characteristics

Leisure time / face to face socialising

There have been important changes in young people’s time use over the period covered by
this report. Comparison of the two LSYPE cohorts shows a significant reduction in the
proportion of 13- to 14-year-olds listing going out with friends and spending time at each
other’s houses as their main spare time activities: these reductions were offset by an
increase in time spent by themselves. (Figure OY9)

226
Figure OY9: How young people mostly spend their spare time, comparison of the two
LSYPE cohorts, England

60

50
50
42
40
Per cent

30

21 22 22
20
20
13
9
10

0
Go out somewhere with Go round to a friend's Spend time with family Spend time by themselves
friends house (or friends go
round to theirs)

13/14 in 2004 13/14 in 2013

Source: Clare Baker et al, Longitudinal study of young people in England: cohort 2, wave 1 (DfE, 2014)

Data from the HBSC school survey analysed in another research study showed a decline in
peer socialising in many countries during the 2000s. In 2002 the proportion of adolescents
in the UK having daily face to face contact with peers in the evening was second highest out
of 26 countries surveyed, at 33.7 per cent, but then saw the largest fall between 2002 and
2010 (falling by 10.2 percentage points. 652

Technology use
Increasing amounts of data have been collected in surveys about young people’s use of
technology during the two decades covered by this report. This picture has to be pieced
together from different data sources: understandably, many surveys have changed the
questions they ask in response to the rapid change in this sector.
Electronic communication with friends
Increasing access to email and texting led to significant changes in young people’s ability to
keep in touch outside school. HBSC data for the period 2002 to 2010 show a steady rise in
the percentage of 11-, 13-, and 15-year-olds who communicated with friends by telephone,
e-mail or text every day. (Figure OY10).

652
De Looze and others.

227
Figure OY10: Trends in young people communicating with friends by telephone, e-mail or
text every day, England

80
Percentage in daily electronic

70
60
communication

50
40
30
20
10
0
11 year old boys 11 year old girls 13 year old boys 13 year old girls 15 year old boys 15 year old girls
England
2002 2006 2010
Source: HBSC survey, successive years

Access to the internet at home and ownership of phones

Both young people’s internet access at home and young people’s phone ownership grew
rapidly during the 2000s.
By 2009, 90 per cent of children aged 12 to 17 had internet access at home. There was a
sizeable gap in access by income – 98 per cent access for households with annual income of
more than £40,000, compared with 75 per cent access for households on incomes under
£20,000. 653
As of 2007, a UK wide survey found that 64 per cent of children had acquired their first
phone by the age of 10. This percentage remained stable for the rest of that decade.654 By
2010 mobile phone ownership by young people aged 12 to 15 stood at 87 per cent.655
Hours spent online
Young people’s screen time rose in many countries during the 2000s. HBSC data shows that
the proportion of 11 to 15 year olds in England exceeding two hours screen time a day rose
from 70.8 per cent in 2002 to 86.5 per cent in 2006 and 91.7 per cent in 2010.656 Annual
surveys of UK young people’s internet use have been conducted since 2007: the data (Figure
OY11) shows a steady rise from average 13.7 hours per week in 2007 to more than 20 hours
in 2018.

653
Synovate UK Ltd., Staying Safe Survey 2009: Young People and Parents ’ Attitudes around Internet Safety
(DCSF, 2009). (UK figures.)
654
Ofcom, UK Children’s Media Literacy (Ofcom, 2011).
655
Ofcom, Children and Parents: Media Use and Attitudes Report (Ofcom, 2014).
656
Ariane Ghekiere and others, ‘Trends in Sleeping Difficulties among European Adolescents: Are These
Associated with Physical Inactivity and Excessive Screen Time?’, International Journal of Public Health, 64.4
(2019), 487–98.

228
Figure OY11: Trends in typical weekly hours spent online by UK 12- to 15-year-olds, self-
report

25

20
Hours per week

15

10

0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

Source: Ofcom, Children and parents media use and attitudes report, successive years

Experience of being bullied

The longest time-series on bullying in England is that collected in the international Health
Behaviour in School-aged Children (HBSC) survey. Figure OY12 shows a reduction in the
proportion who had been bullied in the last two months between 2002 and 2010, with a
resurgence again by 2018.
Figure OY12: Trends in 11–15-year-olds who have experienced bullying in the last two
months, England
50
45
40 37
34 36
35
30 28 36
30 34
Per cent

31
25
27 28
20
15
10
5
0

Boys Girls

Source: HBSC England National Report: Findings from the 2018 HBSC study for England, Brooks et al, 2020

229
Comparison of the two LSYPE cohorts also shows a decline in the experience of bullying
between year 10s in 2005 and year 10s in 2014. (Cyberbullying was excluded from this
comparison, for comparability.) Forty-one per cent of young people in year 10 in 2005 said
they had been bullied in the last 12 months, compared with 36 per cent in 2014. In terms of
the form of bullying, threats of violence had fallen from 20 per cent to 14 per cent, and
actual violence from 15 per cent to 10 per cent.657

Involvement in fighting

The HBSC survey also collects self-report data on school pupils’ experience of fighting.
Figure OY13 shows that the prevalence of frequent fighting has fallen over the last two
decades for both boys and girls. It fell steeply for 11-year-old boys from 2002 onwards (the
first year of the data) and for 15-year-old boys from 2006. For 13-year-old boys there was a
sharp fall between 2006 and 2010 but a stalling in progress thereafter.
Figure OY13: Trends in pupils who have been involved in a physical fight at least three
times in the last year, England
30

25
per cent of pupils

20

15

10

0
2002 2006 2010 2014 2018 2002 2006 2010 2014 2018 2002 2006 2010 2014 2018
11 year olds 13 year olds 15 year olds

Boys Girls
Source: HBSC international reports, successive years

Children as victims of crime

Overall crime
The Crime Survey for England and Wales has surveyed 10–15-year-olds about their
experience of crime since 2009/10. Figure OY14 shows the overall trends. The overall trend
is heavily influenced by violence. In both cases, we see a decline, then some levelling out.

Sarah Lasher and Clare Baker, Bullying : Evidence from the Longitudinal Study of Young People in England 2 ,
657

Wave (Department for Education, 2015).

230
Figure OY14: Trends in crimes experienced by 10–15-year-olds, England and Wales

1400
1200
Thousands of crimes

1000
800
600
400
200
0
1998/99

2005/06

2012/13
1997/98

1999/00

2000/01

2001/02

2002/03

2003/04

2004/05

2006/07

2007/08

2008/09

2009/10

2010/11

2011/12

2013/14

2014/15

2015/16

2016/17

2017/18

2018/19
Overall crime Violence
Source: Crime in England and Wales, year ending December 2019 - Appendix tables

Young homicide victims

Figure OY15 shows trends in homicide victimisation for young people from 2005/06
onwards. The figures for 5–15-year-olds are subject to considerable fluctuation. A clearer
pattern is evident in relation to 16–24-year-old male victims of homicide. The victimisation
rate fell sharply between 2007/08 and 2011/12, stayed low for some years, but rose sharply
in 2016/17.
Figure OY15: Trends in young homicide victims, rate per million, England and Wales

5-15 year olds 16 to 24 year olds

45 45
40
39
40 40

35 35

30 30

25 25

20 20
21

15 15

10 10

5 5

0 0
1997/98
1998/99
1999/00
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
2017/18
2018/19
1997/98
1998/99
1999/00
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
2017/18
2018/19

5 to 15 Male 5 to 15 Female 16 to 24 Male 16 to 24 Female

Source: ONS, Homicide Appendix tables

231
 Self Harm
Self-report studies
Self-report studies suggest that the prevalence of self harm amongst adolescent girls is high
and has risen sharply in recent years.
The 2018 Health Behaviour in School-aged Children report for England found that 35 per
cent of 15-year-old female pupils surveyed in 2018 had ever self-harmed (up from 32 per
cent in 2014). This is shown in Figure OY16. Self harm by boys was at a lower level, but had
also increased, from 11 per cent ever having self-harmed, to 16 per cent in 2018. For some
young people self-harm was frequent: of the 15-year-old girls who had ever self-harmed, 6
per cent did so every day, and 10 per cent several times a week. 658
Figure OY16: Percentage of 15-year-olds in England reporting they had self-harmed

40
35
35 32

30

25
Percent

20 16
15 11
10

0
2014 2018

15 year old boys 15 year old girls

Source: HBSC England National Report 2018

Self harm hospital admission rates

From 2012 onwards, Public Health England published rates of hospital admissions for 10–
14-year-olds and 15–19-year-olds. These show sharp rises in self-harm admission rates for
both age groups, between 2011/12 and 2013/14, then a continued rise for the older group
and levelling off for the younger group. (Figure OY17).

658
Brooks and others.

232
Figure OY17: Trends in hospital admissions for self-harm, 10-14 and 15–19-year-olds,
England
700
600
Admissions per 100,000

500
400
300
200
100
0
1997-98

1998-99

1999-00

2000-01

2001-02

2002-03

2003-04

2004-05

2005-06

2006-07

2007-08

2008-09

2009-10

2010-11

2011-12

2012-13

2013-14

2014-15

2015-16

2016-17

2017-18

2018-19
10-14 year olds 15-19 year olds

Source: PHE Fingertips, accessed July 2022

Suicide
In 2019, the number of 15–19-year-old suicides registered in England was 176. This
represents a rate of 5.7 per 100,000. The rate of 15-19-year-old suicides in England now
is close to the rate of 1998, having fallen and risen again in the intervening two decades,
as Figure OY18 shows.659 The number of 10-14 year old suicides is much smaller, usually
in single figures, and fluctuates. There were 10 suicides by 10–14-year-olds recorded in
England for each of 2017, 2018 and 2019.660
Figure OY18: Trends in suicide rates of 15-19s, England
7
5.2 5.7
6
Rate per 100,000

5
4
3.2

3
2
1
0

ONS, Suicides in England and Wales

660
Note that In England and Wales, in July 2018, the standard of proof used by coroners to determine whether
a death was caused by suicide was lowered to the “civil standard” – balance of probabilities. ONS do not think
that recently observed increases in suicide among males and females in England are due to this change. For
more details see Office for National Statistics, Change in the Standard of Proof Used by Coroners and Its Impact
on Suicide Death Registrations Data in England and Wales (ONS, 2020).

233
Bibliography
A full bibliography for this report is available separately at:
https://sticerd.lse.ac.uk/dps/case/spdo/spdorp15ref.pdf

234