Research and analysis

The economic and social cost of contact child sexual abuse

Published 13 December 2021

Authors: Freddie Radakin, Angie Scholes, Kien Soloman, Constance Thomas-Lacroix, Alex Davies.

The views expressed in this report are those of the authors, not necessarily those of the Home Office (nor do they represent Government policy).

Executive summary

This report provides an estimate of the financial and non-financial (monetised) costs relating to all children who began to experience contact sexual abuse, or who continued to experience contact sexual abuse, in England and Wales in the year ending 31st March 2019. This is estimated to be at least £10.1 billion (in 2018/19 prices). This estimate includes the costs of this cohort being victimised in previous and future years, in addition to lifetime consequences as a result of experiencing child sexual abuse (CSA). It should be noted that due to the way some costs are incurred over a victim’s lifetime this cannot be used as an annual or an in-year cost.

There are a few important things to note about this cost:

• it is mostly a ‘non-financial’ cost – that is, not all costs are directly paid by one organisation to another; some costs use notional (non-market) values which represent estimated harm in monetary terms^{[footnote 1]}
• the estimate represents the historic and future costs associated with victims of abuse who were identified during the year ending 31st March 2019
• this estimate cannot be used as an annual cost (for example, completing an equivalent exercise for the year ending 31st March 2020 and adding these costs together would lead to double counting)
• this estimate does not include the costs associated with online and non-contact sexual abuse

Following the same underlying approach used in ‘The economic and social costs of crime’ (Heeks et al., 2018) the estimate includes costs:

• in anticipation of child sexual abuse (expenditure on protective and preventative measures such as costs of education and training)
• as a consequence of child sexual abuse (physical and emotional harms to victims and survivors, lost economic output, and costs to health and victim services)
• in response to child sexual abuse (costs incurred by the police and criminal justice system, as well as the cost of safeguarding victims)

The authors are aware that this approach may seem reductive to those that have experienced child sexual abuse, and recognise the profound human costs of child sexual abuse to victims and survivors. The impacts of CSA are large and far reaching, and the entirety of this cannot be captured in this report. However, in assessing the cost of CSA in this way we are better able to articulate the cost of CSA on both victims and survivors, and the wider economy, demonstrating the impact in a clear and quantifiable manner.

Within this estimate most of the cost can be attributed to two areas, as shown in Table 1. The areas of lost output and physical and emotional harm account for nearly two thirds (64%) of the £10.1 billion total cost. This shows that most of the total cost is borne by victims and survivors^{[footnote 2]}. Government bears the next largest area of cost, with the cost of the police response, trials and prisons all being the responsibility of Government. There are also considerable costs to the voluntary sector in the form of victim services, CSA-related training and educational prevention.

Table 1 also shows whether the costs in each category are ‘financial’ costs or ‘non-financial’ (monetised) costs. ‘Financial’ costs are the direct cash and budget costs associated with CSA. ‘Non-financial’ costs refer to all other costs, including costs that use ‘notional’ (non-market) values which represent estimated harm in monetary terms. Some cost categories include both financial and non-financial costs.

Table 1 also indicates the time taken for these costs to accrue. This report considers the lifetime consequences as a result of experiencing CSA. As such, some costs (for example, emotional harm) accrue over multiple years. Other costs reflect the costs resulting from the volume of current prevalence in the year ending 31st March 2019 rather than the time of abuse or length of lifelong consequences of contact CSA. Here, in the absence of further evidence, these costs have been capped to one year.

Table 1: Total cost of contact CSA in England and Wales for 2018/19

Cost area	Cost category	Contribution to overall cost (£ millions)	Contribution to unit cost (and method) [1]	Proportion of total cost	Financial or non-financial? [2]	Time to realise [3]
Costs in anticipation	Educational prevention programmes	£3.3	£30 (Top-down)	0.0%	Financial	1 year
Costs in anticipation	Offender prevention programmes	£2.4	£20 (Top-down)	0.0%	Financial and non-financial	1 year
Costs in anticipation	CSA training costs	£2.2	£20 (Top-down)	0.0%	Financial and non-financial	1 year
Costs in anticipation	Subtotal	£7.8	£70
Costs as a consequence	Physical and emotional harm	£5,203.1	£45,985 (Bottom-up)	51.5%	Non-financial	Up to 7.17 years
Costs as a consequence	Healthcare cost	£174.6	£1,545 (Bottom-up)	1.7%	Financial	2.17 years
Costs as a consequence	Lost output	£1,306.6	£11,550 (Bottom-up)	12.9%	Non-financial	41.4 years
Costs as a consequence	Victim services	£25.3	£225 (Top-down)	0.3%	Financial and non-financial	1 year
Costs as a consequence	Subtotal	£6,709.6	£59,300
Costs in response	Police costs	£975.9	£8,625 (Top-down)	9.7%	Non-financial	1 year
Costs in response	Court costs	£41.2	£365 (Top-down)	0.4%	Financial and non-financial	1 year
Costs in response	Prison costs	£348.8	£3,085 (Top-down)	3.5%	Financial	5.47 years
Costs in response	Safeguarding costs	£2,013.7	£17,800 (Top-down)	19.9%	Financial	2.21 years
Costs in response	Subtotal	£3,379.7	£29,870
	Total	£10,097.1	£89,240

[1] ‘Bottom-up’ unit costs are a direct cost per victim measure. ‘Top-down’ costs refer to unit costs estimated by calculating a total cost and dividing by the prevalence estimate.
[2] ‘Financial’ costs are the direct cash and budget costs associated with CSA. ‘Non-financial’ costs refer to all other costs, including costs that use ‘notional’ (non-market) values which represent estimated harm in monetary terms.
[3] This is the time period over which these costs are realised. Note that the time to realise physical and emotional harm costs, healthcare costs and lost output costs include a lag until after victimisation.

Here, and throughout this report, unit costs are presented rounded to the nearest £5 whilst overall costs are presented rounded to the nearest £0.1m. Any discrepancy between subtotals, totals and individual cost contributions is a result of this rounding convention.

The report also provides an estimate of the unit cost of contact CSA, as a companion to the estimated total cost. This unit cost is estimated to be £89,240 and represents the average cost of a single victim of contact CSA in the cohort of current victims in the year ending 31st March 2019.

The unit cost is calculated by aggregating the per victim ‘contribution’ to each cost category. Some per victim contributions are based on a ‘top-down’ method. This means the cost for all contact CSA activity was estimated for this category and then divided by prevalence to arrive at a per victim estimate. Other contributions are ‘bottom-up’. This means the initial estimate was on a per victim basis. Section 2.5 gives details on how to apply the unit cost for the purpose of appraising policy interventions and estimating the costs avoided from reducing victimisation of contact CSA. In addition, each section of this report gives details on the methodology used to calculate the costs in each category.

The unit cost is multiplied by the prevalence (113,144 victims) to produce the total cost^{[footnote 3]}. As Section 3 sets out, accurately estimating the level of contact CSA happening within a country is a difficult task. Previous cost of crime estimates have used police recorded crime, or data from the Crime Survey for England and Wales (e.g. Heeks et al., 2018), but these are not suitable for contact CSA. Various sources were considered, with the decision being taken to use an updated version of the Radford et al. (2013) prevalence estimate. This is based on a survey of young people and their care givers, asked about their experiences of sexual abuse in the year prior to the survey.

Throughout the report conservative assumptions have been made to ensure a more robust estimate, specifically the estimates for the length of sexual abuse and physical and emotional harms suffered by victims. This means lower values have been chosen resulting in a cost estimate that likely underestimates the cost of contact CSA but is more reliable. While the £10.1 billion estimate of the cost of contact CSA may appear large, this is likely to be an underestimate of the true cost.

This estimate of the economic and social costs of contact CSA is the most comprehensive yet. The results of this report reinforce the need to tackle contact CSA within society and the economic harm that results from such sexual abuse.

1. Introduction

This report provides an estimate of the financial and non-financial costs relating to all children who began to experience contact sexual abuse, or who continued to experience contact sexual abuse, in England and Wales in the year ending 31st March 2019.

The cost estimate presented in this paper is a victim-focused cost that estimates current victims in the year ending 31st March 2019 (whether new or continuing cases), and considers the costs of previous and future abuse, in addition to lifetime consequences. This gives a focused understanding of the full social and economic consequences of current contact CSA prevalence.

The unit cost estimate for contact CSA presented in this paper is the cost to the average victim of contact CSA over their lifetime, based on the number of victims in 2018/19^{[footnote 4]}.

Within this report the term ‘victim’ has been used to describe a person who has been sexually abused in childhood (up to the age of 18). Some people who have experienced sexual abuse prefer the term ‘survivor’, and the term ‘complainant’ is used in the criminal justice system when accusations have not been proven. For clarity, we use the term ‘victim’ throughout.

1.1 A note on costing contact CSA

The authors would like to offer a ‘content warning’ to those that may have been affected by child sexual abuse. The authors of this report are aware that costing contact CSA is a complex and sensitive undertaking and that the impacts of sexual abuse can go above and beyond what can be quantified or assigned a monetary value. In the calculation of this costing, there is stark discussion about the length of time that victims are sexually abused as well as discussion of the number of rapes and sexual assault that take place within sexual abuse cases. Additionally, the discussion around physical and emotional harms is highly focused around the harm suffered in contact CSA cases.

1.2 Defining contact child sexual abuse

In line with the existing cross-government definition, throughout this report child sexual abuse is defined as:

Forcing or enticing a child or young person to take part in sexual activities, not necessarily involving a high level of violence, whether or not the child is aware of what is happening. The activities may involve physical contact, including assault by penetration (for example, rape or oral sex) or non-penetrative acts such as masturbation, kissing, rubbing and touching outside of clothing. They may also include non-contact activities, such as involving children in looking at, or in the production of, sexual images, watching sexual activities, encouraging children to behave in sexually inappropriate ways, or grooming a child in preparation for abuse. Sexual abuse can take place online, and technology can be used to facilitate offline abuse. Sexual abuse is not solely perpetrated by adult males. Women can also commit acts of sexual abuse, as can other children.

(Department for Education, 2018a)

CSA is often split into contact and non-contact offending. Contact offending is defined as any sexual abuse that “involve[s] physical contact, including assault by penetration (for example, rape or oral sex) or non-penetrative acts such as masturbation, kissing, rubbing and touching outside of clothing” (Department for Education, 2018a). Non-contact offending is defined as where “a child is abused without being touched by the abuser. This can be in person or online and includes flashing, showing pornography and forcing a child to make, view or share child abuse images or videos” (NSPCC, 2020).

There is of course overlap between contact and non-contact abuse, and abuse that takes place online. For example, offenders may elicit indecent images from children they interact with online without ever meeting them in the ‘offline’ world. Alternatively, contact offenders may create child sexual abuse material in the ‘offline’ world, that is then subsequently shared and used by offenders in online spaces, who never have contact with that child. Offenders may begin grooming a child online and then meet them to commit contact offences, or the grooming may begin ‘offline’, and communication continued online. There are a myriad of ways in which the online and offline worlds may overlap. It is important to note that for children and young people, the distinction between ‘online’ and ‘offline’ may be blurred and not as rigid as suggested in this report.

For this report, the focus is on costing contact CSA. While to some this may seem an arbitrary distinction due to the increasingly online nature of many children’s lives, and the overlap described above, this approach has been taken as a first step in understanding the costs of CSA. It is a complex area, with a number of evidence gaps. By developing this framework, it is hoped that future work can then build on this to develop costs of non-contact CSA.

Evidence shows that CSA affects a considerable proportion of the population, with the Crime Survey for England and Wales (CSEW) estimating that 6% of the adult (18-74) population experienced contact sexual abuse at some point in their childhood (defined by the ONS as before the age of 16) (Office for National Statistics, 2020a). Within this 6%, 2% experienced rape and 5.8% experienced other contact sexual abuse (for example, groping or kissing or being forced to touch someone else’s body for sexual purposes)^{[footnote 5]}. Women are more likely than men to have been victims of CSA with 9.4% saying they experienced contact CSA in childhood compared to 2.7% of men.

While the CSEW estimates the proportion of adults who experienced CSA as children, it is more difficult to estimate current prevalence of CSA, i.e., the number of children currently experiencing sexual abuse. It is often not disclosed to professionals or identified by professionals when it occurs. Over 60% of those who experienced sexual abuse before the age of 16 did not tell anyone at the time (Office for National Statistics, 2020a). Less than 10% of those who made disclosures made them to someone in an official position such as the police or a doctor (Office for National Statistics, 2020a). The reasons for this are numerous; the most common reason for not telling anyone at the time was “embarrassment” (Office for National Statistics, 2020a).

Looking at offences that are reported to and recorded by the police in the year ending 31st March 2019, there were 63,621 contact CSA offences (including sexual grooming) recorded by the police in England and Wales. This is a 4% increase on the previous year (Home Office, 2020). However, only a small number of reported offences result in a charge – in the same period there were 4,412 charges for contact CSA offences. A further 34,143 offences did not reach a charge due to evidential difficulties. Numbers of prosecutions and convictions for contact CSA offences are also low: in 2019, 3,768 offenders were prosecuted for contact CSA offences, and just 2,728 offenders were convicted (Ministry of Justice, 2020).

1.3 Previous estimates of the cost of child sexual abuse

There are three previous pieces of work looking at the cost of CSA in the UK. The first, Saied-Tessier (2014), aimed to estimate the total cost to the economy from all CSA victims in a given year, including adult victims still affected by sexual abuse suffered in childhood as well as current victims. Saied-Tessier estimated this cost to be £3.2 billion (2014 prices). The second, by Conti et al. (2017) took a wider approach, looking at the average lifetime cost of being a victim of child maltreatment more generally, including CSA, estimating it to be £89,390 (2017 prices). Conti et al. took this approach, rather than giving a lifetime cost, because of the difficulty of estimating the prevalence of child maltreatment.

The Home Office has previously produced an estimate for the cost of a subset of CSA in the form of estimating the cost of child sexual exploitation (CSE)^{[footnote 6]}. This uses an adjusted estimate based on a model estimating the cost of sexual slavery. The cost per victim (i.e. unit cost) for CSE was estimated as £348,410 with the total annual cost estimated at £2.3 billion in financial year 2015/16 (Fell et al., 2019).

1.4 The approach of this report

In order to ensure consistency with other Home Office costs of crime work, and to produce comparable estimates, this report follows the methodology and framework of previous Home Office research reports into the costs of crime, notably Heeks et al. (2018). This framework presents the costs in three parts: in anticipation of CSA, as a consequence of CSA, and in response to CSA. While the overall approach is the same, there are some notable differences in this work. Previous Home Office estimates have used police recorded crime (PRC) or Crime Survey for England and Wales (CSEW) data to estimate the number of victims, but as discussed in Section 3 this report has taken a different approach. A different approach to estimating lost output is also necessary, to reflect the fact that ‘lost output’ in cases of CSA refers to lost output later in life once victims of contact CSA have started working. This is in contrast to previous Home Office estimates which focus on crimes affecting adults, who are already of working age.

In developing this cost estimate the authors have drawn on expertise from an external steering group. This was comprised of subject matter experts from law enforcement, academia, the third sector and Home Office economists. The steering group advised on the overall methodology, how different aspects were costed, and how key assumptions were made. The external steering group also reviewed the report at an initial stage. One consistent recommendation of the steering group was to be conservative with assumptions. An example of this is the use of minimum values when estimating the frequency of rapes within CSA cases. The consequence of the use of conservative assumptions is that the cost estimates produced by this report are likely underestimates of the true cost of contact CSA.

As well as the use of conservative assumptions, the external steering group agreed that some areas of cost should be excluded from this estimate to ensure the cost is reliable, if an underestimate. These areas are the long-term health cost of contact CSA, and any potential risk of future offending by victims. For both there was limited quantitative evidence in the format needed to include these in the costs modelling, and thus they have been excluded from this analysis. Data from Sexual Assault Referral Centres (SARCs) has also not been included in the estimate, again because it is not available in a format suitable for inclusion in this model.

All costs in this report are in 2018/19 financial year prices. Where the data used gives costs for a calendar, rather than financial, year, they have been uprated to 2018 prices. This is close to the base year as it represents a 9-month overlap with the 2018/19 financial year, so further adjustments have not been made (see Table AP4 in Annex).

This report considers the lifetime consequences as a result of experiencing Child Sexual Abuse. As such, some costs (for example, emotional harm) accrue over multiple years. Other costs reflect the costs resulting from the volume of current prevalence in the year ending 31st March 2019 rather than over the time of abuse or length of lifelong consequences of contact CSA. Here, in the absence of further evidence, these costs have been capped to one year.

Some costs in this report are ‘financial’ costs, whilst others are ‘non-financial’ (monetised) costs. ‘Financial’ costs are the direct cash and budget costs associated with CSA. ‘Non-financial’ costs refer to all other costs, including costs that use ‘notional’ (non-market) values which represent estimated harm in monetary terms.

While the estimates included in this report represent the ‘average’ cost of being a victim of contact CSA, for many victims the cost may exceed this or even never be quantifiable. The unit cost produced is an average and for some victims the cost may be far greater than we have estimated.

Throughout this report, unit costs are presented rounded to the nearest £5 whilst overall costs are presented rounded to the nearest £0.1m. Any discrepancy between totals and individual cost contributions is a result of this rounding convention.

2. What is included in this report?

This report focusses on the costs of contact CSA and therefore, as discussed in section 1.1, non-contact offences are out of the scope of this report. The current analysis breaks down the costs of CSA into three categories: costs in anticipation, as a consequence and in response.

2.1 Costs in anticipation

Educational prevention costs

Educational prevention refers to programmes that are intended to raise awareness of contact CSA, teach children about the risks of contact sexual abuse and the importance of consent within relationships. These costs cover voluntary sector spending, and National Crime Agency (NCA) spending on the Child Exploitation and Online Protection command (CEOP)’s educational work.

Offender prevention costs

Offender prevention is any spending on services that work with offenders or individuals who may be at risk of offending in order to reduce the risk that they offend against children. The costs cover voluntary sector spending.

CSA training costs

CSA training refers to measures used to enable adults to adequately respond to contact CSA, as well as equipping them with the skills to identify when contact CSA is happening. This training is positioned as a cost in anticipation rather than in response because it is provided/undertaken on the expectation that adults may be required to identify and/or respond to contact CSA, rather than being provided in relation to specific cases of abuse those adults have identified/responded to. These costs cover voluntary sector spending and police spending.

Some costs that are likely to be incurred in anticipation of contact CSA have not been included in this estimate, due to difficulties in quantifying the spending or the proportion of the cost attributable to contact CSA specifically. These include:

• educational prevention programmes through relationship and sex education
• training for those working in education, health and social care
• the Child Sex Offender Disclosure Scheme (CSODS)^{[footnote 7]}
• early intervention schemes, such as those targeted at at-risk young people, because these tend to be designed to target a range of vulnerabilities
• the costs of Disclosure and Barring Services (DBS)

Therefore, this section is likely an underestimate of the true costs in anticipation of contact CSA.

2. 1 Costs as a consequence

Physical and emotional harm

The costs of physical and emotional harm concern injuries to victims as a consequence of CSA. These costs use the quality adjusted life year (QALY) methodology, which considers the reduction in quality of life from different injuries, based on the duration, severity and frequency of each injury. These physical and emotional harms represent immediate costs to the victim – for example bruising, scratches and cuts. It has not been possible to include estimates of long-term healthcare costs such as consequent chronic injuries or long-term mental health impacts.

Healthcare costs

Healthcare is the response to and provision of care for physical and emotional injuries, as a consequence of contact CSA. This covers costs incurred by the NHS and other healthcare providers.

Lost output

Lost output refers to the reduced earnings or potential future earnings due to the increased likelihood of unemployment that CSA victims are exposed to throughout the duration of their working life.

Victim services

Victim services refer to the costs of providing support to victims of contact CSA and their friends and family. This covers the initial cost of support as well as the opportunity cost of volunteer time in delivering victim services.

There are some ‘costs as a consequence’ that have not been included in this costing due to difficulties quantifying the costs involved. These include specialist healthcare costs and the long-term healthcare costs (including diminished life expectancy) related to contact CSA. Specialist healthcare costs are concentrated in Sexual Assault Referral Centres (SARCs) which are specialist medical and forensic services centres providing services for victims of rape and sexual assault. Therefore, this section is likely an underestimate of the costs as a consequence of contact CSA.

2.3 Costs in response

Police costs

As a large part of the police’s resources are spent dealing with crime, the cost captured here is therefore the opportunity cost of police time and resources taken up by responding to contact CSA crimes rather than engaging in other activities, such as responding to non-crime activities.

Court costs

This is the cost to the Government of holding and preparing for the trials of those proceeded against for contact CSA offences. The costs include the cost of juries, the cost to the Crown Prosecution Service (CPS) and the cost of legal aid in response to contact CSA.

Prison and probation costs

Prison and probation costs refer to the cost of custodial and non-custodial sentences in response to contact CSA. This considers the number of custodial sentences in the year ending 31st March 2019 in response to child sexual abuse, the cost per year in prison and the average length of time offenders are sentenced for. Additionally, the cost of those given non-custodial sentences have also been costed. The two costs are combined to give the prison costs.

Safeguarding costs

Safeguarding costs refer to measures that are put in place to help protect victims of contact CSA. The costs cover two of the Department for Education’s (DfE) safeguarding roles: Children in Need (CiN) and Looked After Children (LAC). This section also includes data on Children Receiving Care and Support (CRCS) in Wales.

The costs in response estimated above does not include all spending as a response to contact CSA. There are additional costs to the criminal justice system (CJS) beyond those related purely to trials and prison and probation^{[footnote 8]}. Additionally, aside from the police costs, proxy figures have been used where contact CSA averages were unavailable. These may not reflect the true cost of responding to contact CSA. It has also not been possible to estimate probation costs, nor the cost of offenders on licence or under notification requirements (commonly known as being on the ‘sex offenders register’). Therefore, the costs in response to contact CSA as set out in this report is likely an underestimate of the true cost.

2.4 Unit costs and total costs

As well as the total cost, the report also provides an estimate of the ‘unit cost’ of contact CSA. The unit cost represents the average cost of a single victim of contact CSA in the cohort of current victims in the year ending 31st March 2019.

The unit cost is calculated by aggregating the per victim contribution to each cost category. Some of these are calculated using a ‘top-down’ method, where the cost for all contact CSA-related activity was estimated for the category, and then divided by the prevalence figure to arrive at a per-victim estimate. This is the same approach as used in Heeks et al. (2018). Because of this, some of per-victim costs can seem counter-intuitive, such as police or prison costs. While the cost of one investigation, or one incarceration, is high, it can result in a smaller contribution to the unit cost because the total cost is divided by the (larger) prevalence estimate, rather than the (smaller) number of offences recorded by police.

Other cost categories use a ‘bottom-up’ method. In this instance the initial estimates are on a per-victim basis, which is then scaled up by the prevalence estimate to give a total cost for that category.

2.5 How to use the cost estimates in this report

Total costs

It is important to note that the cost estimates presented in this report are non-financial costs – that is, some of the costs use ‘notional’ (non-market) values which represent estimated harm in monetary terms, rather than all being costs directly paid by one organisation to another.

This estimate takes a snapshot of the number of victims who experienced sexual abuse in the year ending 31st March 2019 and estimates the costs incurred over the lifetime of the victims abused in this year. As such, the cost estimated here is not an in-year cost. Some victims who experienced sexual abuse in this period will have experienced abuse in the years before this time period, the costs of which are included in this report.

Since this cost estimate is not an in-year cost, this cost should not be treated as an annual cost of contact CSA. If one were to complete an equivalent exercise to estimate the lifetime cost for those who suffered abuse in the year ending 31st March 2020, that estimate would count the lifetime cost to some of the same victims as in this report. This is because this estimate assumes the average length of an abuse period to be 2.17 years. As such, taking a snapshot of the number of people who suffered abuse next year would include some of the same people included in this year’s snapshot. Estimating the lifetime costs to these victims would therefore lead to double counting.

The Home Office does not plan to estimate this cost annually, or to use this cost estimate in this way. It is important that this cost is treated as the lifetime cost for those who suffered abuse in the year ending 31st March 2019 and not an annual cost of contact CSA per year.

Unit cost

The estimated unit cost of contact CSA is designed to help policy makers and practitioners weigh up the crime reduction benefits of policies and assess the cost effectiveness of particular interventions. Users should therefore understand how to use the estimate provided here to ensure the net benefit from crime reduction policies is calculated correctly. The total cost of £10.1 billion provides an estimate of the costs relating to all children who began to experience contact sexual abuse, or who continued to experience contact sexual abuse, in England and Wales in the year ending 31st March 2019, which is broader than just that identified by the authorities. As set out in Section 2.4 this unit cost is derived from a series of category costs, some of which use a ‘top-down’ method, and some of which use a ‘bottom-up’ method. The total overall cost is estimated by multiplying the unit cost by prevalence (see Section 3). It is not possible to update the overall total cost figure in the future by multiplying the total unit cost by a new prevalence figure as the ‘top-down’ elements of the unit cost are specific to this report’s prevalence estimate.

This is an important distinction to understand when estimating the benefits of any crime reduction policy. For each victim prevented as a result of a policy intervention, the cost saving is not the total unit cost, as some of the spend on, for example, prevention measures, would still be incurred. It is therefore advised that the unit cost is adjusted to only include a subset of cost areas that contribute to the total cost that would be avoided ‘at the margin’ of intervention, and exclude those costs tied to wider prevalence (see Executive summary; Table 1).

The total unit cost presented here is for the ‘average’ victim, whether or not the sexual abuse was identified by or reported to the police. As evidence shows that this is only a small proportion of contact CSA the contribution to the unit cost should be interpreted as only including a share (i.e. average) of the law enforcement and criminal justice system costs, and not the full per-victim cost. This share is known as a multiplier and can be considered an empirical take on the probability a crime will be reported to the police.

3. Estimating the prevalence of contact child sexual abuse

3.1 Finding an appropriate prevalence estimate

In order to estimate the cost of contact CSA an understanding of the number of victims of contact CSA is needed. This is referred to as the ‘prevalence estimate’. Accurately estimating the level of CSA happening within a country is a difficult task. As discussed in Section 1.2, understanding of the prevalence of CSA is limited due to under reporting and under identification. Previous estimates of costs of crime (e.g. Heeks et al., 2018) have used police recorded crime (PRC), or data from the Crime Survey for England and Wales, but these are not appropriate for CSA (see below for more detail). This report therefore uses a prevalence estimate from Radford et al. (2013). Various other options were considered for the prevalence estimate, including the Crime Survey for England and Wales estimate, contact CSA offence sub codes in sexual offences on those aged under 18 in PRC, and an estimate produced by the Office for the Children’s’ Commissioner. This section sets out the advantages and disadvantages of each of these options.

Necessary features of a prevalence estimate

To be suitable for use in this report a prevalence estimate required four features:

• it should be a yearly estimate of the number of incidents of contact CSA committed
• it should be for England and Wales
• it should cover all forms of contact CSA (including sexual abuse perpetrated by those aged under 18 as well as adults, including extra and intra-familiar sexual abuse) and include all victims of contact CSA, comprehensively covering all victims from birth to turning 18
• the estimate should be as recent as possible, and if not, should be able to be updated to provide an estimate for 2018/19

Crime Survey for England and Wales (CSEW)

The first option this report considered was the Crime Survey for England and Wales (CSEW), which has been used by previous Home Office publications on the cost of crime. The CSEW (2020) estimates the number of adults who experienced contact CSA as children. While this is a robust estimate, based on a representative survey of adults in England and Wales which asks about all forms of contact sexual abuse, it does not meet the needs of this work. It is not an estimate of contact CSA in a given year and it also does not capture experiences of those aged 16-17, as the CSEW defines children as those under 16. Therefore, it cannot be used in this estimate.

Police recorded crime (PRC) using contact CSA sub codes

Police recorded crime (PRC) is a measure of the number of crimes reported to the police in a certain year. Thus, it will always meet the criteria of providing a yearly estimate for the number of crimes committed and the data provided is for police forces in England and Wales. However, the Office for National Statistics (ONS) advises that PRC is not a good measure of the prevalence of CSA (ONS, 2020). The main issue is that offences that are not reported to, or recorded by, the police are not captured in the data. When considering a so-called ‘hidden crime’ such as CSA where there are known to be low rates of reporting, this is a major issue. There is also a lack of consistency in recording crimes between police forces leading to concerns about the quality of the data. Finally, the PRC contact CSA sub codes (used to identify contact CSA cases) only include victims up to the age of 16, and thus miss a crucial cohort of contact CSA victims aged 16 and 17 years. For these reasons, it was decided that PRC should not be used to provide an estimate of prevalence.

Police recorded sexual offences where victims are under 18

An alternative way to identify sexual offences against children in PRC is to look at all sexual offence sub codes where the victim is aged under 18. The benefit of this approach compared to that outlined above is that it captures crimes committed against victims aged 16 and 17 years. However, as it is PRC, the caveats above relating to under identification and reporting still apply, so this is not considered suitable to provide an estimate of prevalence.

Office of the Children’s Commissioner estimate

The estimate produced by the Office for the Children’s Commissioner (OCC) used a mathematical approach known as multiple systems estimation to estimate the total number of victims of CSA over two years. It is based on PRC and the Department for Education’s Children in Need (CiN) database. This resulted in an estimate of 375,000 – 475,000 victims of CSA between 2012-2014 (Children’s Commissioner, 2015). Although this is a two-year estimate, it can be divided by two to estimate the yearly prevalence. Equally, this estimate uses a mathematical formula to ‘correct’ for the underreporting present in police-based estimates. However, there are two aspects of this estimate that mean it is unsuitable. First, the mathematical approach requires that the two datasets are independent of one another. However, PRC and the CiN datasets are not independent as there can be referrals between the two. Second, at the time this estimate was created a considerable proportion (approximately 30%) of CSA offences recorded by the police were ‘non-recent’ – i.e. they were offences that happened more than one year prior to being recorded. This results in an overestimate of the number of current victims. Finally, this prevalence estimate could not be updated to the appropriate year (2018/19) because of lack of access to both datasets. This estimate was therefore judged not to be appropriate for the current work.

Radford’s estimate for the NSPCC

Radford et al. (2013) looked at the prevalence of child maltreatment through a survey of caregivers, children and young people. In total 6,196 people were asked about theirs and others’ experiences of child maltreatment, including contact CSA, in the previous year.

The sample included 2,160 parents or caregivers of children aged between 2 months and 10 years, 2,275 children aged between 11 and 17 years and their parent or caregiver, and 1,761 young adults aged 18-24 years. There were slightly more female children, young people, and young adults in the sample than males (51.6% female and 48.4% male). Most parents or caregivers were female (84.7%) reflecting the parents’ own definitions of who was the primary caregiver. Most children and young people in the sample had White British ethnicity (82%), while 3% were “Other White”, 4% were Mixed, 5% South Asian, 3% Black British, African or African Caribbean, and 2% were Chinese or from other ethnic groups (Radford et al., 2013)^{[footnote 9]}.

This survey meets all four requirements set out above: it provides a prevalence estimate for a given year in England and Wales, includes all forms of contact CSA including sexual abuse perpetrated by those aged under 18 as well as by adults, and extra- and intra-familiar sexual abuse, and it is possible to update the estimate to give a figure for 2018/19. However, it should be noted that the survey was conducted in 2009, and so it is possible that the prevalence of contact CSA has changed since then. It is also important to be aware that the estimate of the number of children aged under 11 years old who experienced contact CSA relies on caregivers reporting this, which could lead to under reporting because those caregivers are either not aware the child is experiencing abuse, or could be the abuser and therefore are unlikely to report this in the survey.

Nevertheless, the Radford et al. (2013) estimate provides the most appropriate and robust estimate of the prevalence of contact CSA for the current work.

Potential prevalence estimates

As set out in Table 2, different estimates fulfil different criteria. As Radford et al. (2013) is the only estimate that meets all the criteria, this has been chosen to provide the prevalence figure for this report.

Table 2: Features of potential prevalence measures

	Prevalence features
Prevalence option	Yearly data?	Data from England and Wales?	Includes types of CSA and all victims of CSA?	Recent data or able to be updated to 2018/19
Crime Survey for England and Wales		×	×
Police recorded crime (PRC) estimate	×	×		×
Recorded sexual offences where victims are under 18	×	×		×
Office of the Children’s Commissioner estimate	×	×	×
Radford et al. (2013) estimate	×	×	×	×

3.2 Creating an estimate of yearly prevalence

The Radford et al. (2013) estimate is based on 0-17-year olds who have experienced contact CSA in the year prior to the survey, with responses for children 0-10 years coming from caregivers. The survey did not ask 18-24-year olds about their experiences because the majority were not children in the year prior to the survey. The survey found that 0.2% of 0-10 year olds (as reported by caregivers) and 2.1% of 11-17 year olds reported being victims of contact CSA (Radford, et al., 2013).

Table 3: Victimisation rates for contact CSA by age (Radford, et al., 2013)

Age	Victimisation rate
0 – 10 years old (as reported by caregivers)	0.2%
11 – 17 years old	2.1%

While the survey was carried out in 2009, the prevalence estimate can be updated by applying the above figures to 2018 population levels. In doing this the following assumptions were used:

1. The victimisation rate found in Radford, et al. (2013) is representative of the general population in England and Wales.
2. The reported victimisation rates within Radford, et al. (2013) do not change between 2010 and 2018. This assumption is a cautious one as it assumes there has been no increase in victimisation.

By using the population of England and Wales for mid-2018 (Office for National Statistics, 2020b), and the victimisation rates from Radford, et al. (2013), an estimate of the number of victims of contact CSA in a given year can be made. As shown in Table 4 it is estimated that in 2018 there were approximately 113,000 victims of contact CSA. In this report this will be referred to as the prevalence figure or the prevalence.

Table 4: Estimated number of victims of contact CSA in a year

Age	Victimisation rate	Number of victims
0 to 10 years old	0.2%	15,908
11 to 17 years old	2.1%	97,235
Total victims	-	113,144

It is acknowledged that this estimate of the number of yearly victims is likely to be an underestimate of the true scale of contact CSA within England and Wales, due to the underreporting issues laid out in the assessment of Radford et al. (2013) in section 3.1.

4. Estimating the costs in anticipation of contact child sexual abuse

This chapter discusses the methodological approach and outlines the estimates of the costs in anticipation of contact CSA. These include:

1. educational prevention
2. offender prevention
3. CSA training

As set out in Section 2.1, Disclosure and Barring Services (DBS) costs are not included in this estimate. This is because it is not possible to split the costs associated specifically with safeguarding children, from those relating to the safeguarding of adults. In line with the conservative approach throughout DBS costs have therefore been excluded from the model.

4.1 Educational prevention

4.1.1 Approach

Educational prevention captures programmes that are intended to:

• raise awareness of contact CSA
• teach children about the risks of contact sexual abuse
• teach children about consent within relationships

These programmes potentially result in greater levels of disclosure of sexual abuse as well as acting as an aspect of prevention. It is of course not the child’s responsibility to prevent child sexual abuse. These programmes are considered to be costs in anticipation of contact CSA because they would likely not exist without the crime of contact CSA.

Within this report two key areas of spending on educational prevention have been costed: the voluntary sector, and National Crime Agency (NCA) spending on the Child Exploitation and Online Protection team (CEOP). There is additional spending by the Department for Education (DfE) in this area, however this spend is not included here as it is not possible to isolate the element specific to CSA.

A survey was used to collect data from the voluntary sector on their spending on educational prevention. Charities were asked about their typical annual spending on “educational programmes in schools, or targeted support for those at risk of child sexual abuse and programmes such as youth engagement”. Responses were grouped into three categories: first, charities working with offenders who also provide programmes focused on educating children about safety online, how to spot the signs of grooming and reporting CSA and inappropriate activity. All three such charities identified and contacted provided data. The second category is Rape Crisis centres, of which eight responded, and the third is ‘other charities working with victims, of which 12 responded. As many charities working in this space did not respond to the survey, the data provided by Rape Crisis centres and ‘other charities’ was used to scale up to provide an estimate of costs to the whole sector, using the median spend for each category.

The data showed that the spend reported by different Rape Crisis centres was broadly similar while the spending of other charities varied considerably. The median spend of Rape Crisis centres was therefore applied to all Rape Crisis centres, while a separate median spend for ‘other charities’ was applied to the rest of the sector. Combined with the data from the three offender-focused charities, this provided an estimate of the total third sector spend on educational prevention.

The NCA provide educational prevention programmes via CEOP, with spending split into two categories: online resources available on the CEOP website that can be used by, for example, schools, and spend on the staff that develop, deliver and update these resources. The NCA provided data on this spend for this report.

The DfE also spend money on educational prevention programmes through ‘Relationship and Sex Education’ (Department for Education, 2019a), which includes the aim of teaching pupils to “recognise and to report abuse, including emotional, physical and sexual abuse”. However, it is not possible to assign a proportion of the cost of these materials specifically to contact CSA. Thus, this area of cost has not been included in this report.

This report acknowledges that there is additional spending on educational prevention that has not been identified or estimated by this report. It is for this reason that this estimate of spending is considered to be an underestimate.

The educational prevention costs presented here are financial costs (as there is a transfer of budget between organisations) which are realised over one year (the year ending 31st March 2019). The contribution to the unit cost is a ‘top-down’ cost. It has been estimated by dividing the total educational prevention cost by the prevalence estimate.

4.1.2 Calculating the cost of educational prevention

As above, all three offender-focused charities provided data on spend. Of the victim-focused charities 12 provided data (out of a total of 56), and eight Rape Crisis centres provided data (out of a total of 41). This gave a response rate of approximately 20% for both^{[footnote 10]}.

As above, because all offender-focused charities identified provided data, the total spend can be used as-is. This reported spend was £263,000.

The total spend from the rest of the charities that responded to the survey, including Rape Crisis centres and other charities, was £837,400.

As set out above, to estimate the spend for those charities that did not respond, median values were used. The median spend for each Rape Crisis centre was £15,000 in the year ending 31st March 2019, compared to £19,000 in the year ending 31st March 2019 for each of the other charities. Based on these median spending values, the spending for the 44 remaining other charities was therefore estimated to be £836,000 and for the remaining 33 Rape Crisis centres, £495,000.

By combining the reported spending with the extrapolated spending, it is calculated that the total spending by charities on educational prevention is £2,431,400. While a large number of charities were invited to complete the survey, it was of course not possible to identify and contact every one that may work in this area. Therefore, this is likely to be an underestimate of the full cost to the voluntary sector.

The NCA data on spending by CEOP on educational programmes in a year showed that spending on both staff costs and the resources produced by CEOP was £842,000. However, as CEOP covers online CSA as well as contact CSA this is a broader level of spending than just contact CSA, which this report is costing. It is not possible to break this spend down between online and contact offending. Therefore, this is likely to be an overestimate of CEOP’s education spend in respect of contact CSA.

4.1.3 Contribution to the total cost and the unit cost

By adding the spending by the voluntary sector and CEOP, it is estimated that the total spending on educational prevention was £3.3 million in the year ending 31st March 2019. This is a contribution to the unit cost of £30 per victim of contact CSA.

4.2 Offender prevention

4.2.1 Approach

Offender prevention describes any spending on services that work with offenders or potential offenders in order to reduce the risk of offending against children. This can include programmes to reduce reoffending amongst convicted CSA offenders, or programmes that offer support to individuals concerned about their own or others’ behaviour. Spending in this area is by charities, and thus the cost was estimated using a survey sent to three charities that offer support to offenders and individuals concerned about their behaviour. Charities were asked about spending on “counselling and support for those with an interest in children, support for those engaged in harmful sexual behaviour, management of those convicted of offences”.

Charities were also asked about the number of volunteer hours used. From an economic perspective these volunteer hours are a cost to society as they are hours that the volunteer could otherwise use for their leisure. The number of volunteer hours in the year ending 31st March 2019 as reported by the survey were multiplied by the value of non-working time. The value of non-working time is worked out by the Department for Transport and is used as a proxy for leisure time.

The offender prevention costs presented here consist of both financial costs and non-financial costs. Spending on support programmes represents a financial cost (as there is a transfer of budget between organisations). The cost of volunteer time represents a non-financial (monetised) cost as this is the ‘notional’ estimate of the cost in monetary terms. These costs are realised over one year (in the year ending 31st March 2019). The contribution to the unit cost is a ‘top-down’ cost. It has been estimated by dividing the total offender prevention cost by the prevalence estimate.

4.2.2 Calculating the offender prevention cost

Total reported spending on offender prevention work by all three charities came to approximately £2,181,000 for the year ending 31st March 2019. All 3 charities responded to the survey giving a good indication of spending in this area. The charities reported that they received approximately 30,000 volunteer hours in the year ending 31st March 2019. This was multiplied by the value of non-working time which was £5.74 per hour in 2018 prices (Department for Transport, 2020). Thus, the value of volunteer time in this area is estimated to be approximately £172,330 in the year ending 31st March 2019. Adding the cost of volunteer time to the spending on offender prevention, it is estimated that the total cost of offender prevention is approximately £2,353,330 in the year ending 31st March 2019.

4.2.3 Contribution to the total cost and unit cost

Through the survey responses it is estimated that total spending on offender prevention is £2.4 million in the year ending 31st March 2019 and this is a contribution to the unit cost of £20 per victim.

4.3 CSA training

4.3.1 Approach

CSA Training covers spending on measures to train adults to be able to respond to contact CSA, as well as equipping them with the skills to identify contact CSA. While there is a wide range of spending in this area this report has only been able to estimate spending on two aspects: spending by charities and spending by the police. It is acknowledged that there is additional spending on CSA training that has not been included in this report^{[footnote 11]}, and it is for this reason that this is considered to be an underestimate.

As set out in Section 4.1, a survey was used to gather data on spending in this area by the voluntary sector. This went to the same three offender-focused charities, 56 victim-focused charities, and 41 Rape Crisis centres. The survey asked about spending on “programmes delivered to adults rather than children, for example this could be training teachers to identify the signs of CSA, training police officers on supporting CSA victims, or training social workers to identify at risk individuals. Training to parents and other non-professionals would also be included”. As with educational prevention, the median values of spending were used to estimate the spending for victim-focused charities and Rape Crisis centres that did not respond. Separate median values were calculated for each due to the differences in spending between these two categories.

Data on the cost of CSA training by the police was provided by the College of Policing^{[footnote 12]}. The police receive a variety of training packages related to contact CSA including as part of the basic training to become a police officer. However, it is not possible to calculate how much of this training relates specifically to contact CSA so this type of training has not been included in this costing. Spending on three programmes could be directly attributed to the crime of contact CSA (College of Policing Internal Communication, 2020):

• child abuse specialist investigators course
• e-learning on CSA offending in the home
• e-learning on CSA offending in missing person cases

Estimates of spending on these programmes are based on the delivery costs of the programmes, multiplied by the number of people undertaking the training in a year. Delivery costs were found by calculating the number of hours of police time the training took and the cost of that time. This time could be used elsewhere if there was no contact CSA and thus represents an economic cost. It should be noted that whilst the College of Policing provides these packages to police forces, forces can choose to expand on this training, for example in the form of external specialist speakers or additional materials. The estimate for spending on training within this report is therefore likely to be an underestimate of the true police spending on training.

The estimate for the cost of training does not include spending on training specifically for those working in education, and health and social care that is provided centrally/by Government. Thus, this is an underestimate of the true training cost in anticipation of contact CSA.

The CSA training costs presented here consist of both financial costs and non-financial costs. Spending by charities represents a financial cost (as there is a transfer of budget between organisations). The cost of police time represents a non-financial (monetised) cost as this is the ‘notional’ estimate of police time in monetary terms. These costs are realised over one year (the year ending 31st March 2019). The contribution to the unit cost is a ‘top-down’ cost. It has been estimated by dividing the total CSA training cost by the prevalence estimate.

4.3.2 Calculating the cost of CSA training

As set out in Section 4.1, survey responses were received from all offender-focused charities contacted, and approximately 20% of the victim-focused charities and Rape Crisis centres contacted^{[footnote 13]}.

Again, as in Section 4.1, the data provided by Rape Crisis centres and the other victim-focused charities was used to calculate a median spend for each, which was then used to provide an estimate of costs to the whole sector.

The total reported spend from offender-focused charities was £152,000. The total reported spend of victim focused charities and Rape Crisis centres was £288,225.

The median spending by victim focused charities on training was £15,050 in the year ending 31st March 2019, and the median spend by Rape Crisis Centres on training in the year ending 31st March 2019 was £4,000. Using this to scale up to represent all charities provides an estimated further spend of £662,200 for victim-focused charities, and £132,000 for Rape Crisis centres.

Adding together reported spending with extrapolated spending, it is estimated that spending by charities on training is £1,234,425.

The cost of police training was estimated as follows:

• the time-cost of the child abuse specialist investigators course was £1,415 per person, and 430 people completed the course in 2018 (College of Policing Internal Communication, 2020), leading to a total cost of £608,740^{[footnote 14]}
• the E-learning on CSA offending in the home had a time cost of £14.05 to deliver and was accessed by 10,521 people in 2018 (College of Policing Internal Communication, 2020), leading to a cost of £147,760
• the E-learning on CSA offending in missing persons cases had a time cost of £16.85 per person and was accessed by 10,235 police officers in 2018 (College of Policing Internal Communication, 2020), leading to a cost of £172,495

By combining the cost of all three training programmes, the total cost of police training on contact CSA is estimated to be £928,995.

4.3.3 Contribution to the total cost and unit cost

By combining the spending by charities on training and the spending by police on training, it is estimated that the total cost of training in anticipation of contact CSA is £2.2 million in the year ending 31st March 2019. This is a contribution to the unit cost of £20 per victim.

4.4 Contribution of cost in anticipation to the total cost and unit cost

The total contribution of the costs in anticipation of contact CSA to the overall cost is £7.8 million in the year ending 31st March 2019. This area of cost contributes £70 per victim to the average unit cost of a contact CSA victim.

5. Estimating the costs as a consequence of contact child sexual abuse

This chapter discusses the methodological approach and outlines the estimates of the costs as a consequence of contact CSA. These include:

1. physical and emotional harm
2. healthcare costs
3. lost output
4. victim services.

5.1 Physical and emotional harm

5.1.1 Approach

The physical and emotional harms that victims of contact CSA experience vary widely, both in the type of harm (be it physical injuries, or emotional harms such as anxiety and depression) and in the duration of that harm (see Section 5.1.2). The quality adjusted life year (QALY) approach used in Dolan et al. (2005) and Heeks et al. (2018) is used here. This approach determines the negative percentage impact on a person’s quality of life from different injuries, regardless of how they are incurred. For example, if a person breaks their rib, they are judged to suffer approximately a 10% reduction in their quality of life whilst they are recovering (Salomon et al., 2015).

The cost of the injury is based on the likelihood of sustaining physical and emotional injuries (LIKE) multiplied by the percentage reduction in quality of life (REDUCEQL) multiplied by the duration of the injury (DUR) as a fraction of a total year. This is then combined with the value of a year of life at full health (VOLY) to give an estimate of the average cost associated with contact CSA. For this aspect of the cost estimate contact CSA is split into two types – rape and sexual assault – each with its own harms. The QALY formula is as follows:

LIKE×REDUCEQL×DUR×VOLY=Average physical and emotional cost

The physical and emotional harms are adjusted if they run over multiple years, through economic discounting. This discounting follows the guidance laid out in ‘The Green Book’ (HM Treasury, 2018). Each of the components of the QALY approach will be discussed in more detail in this section to show how the estimates have been developed.

In some cases, victims of CSA are murdered. It has not been possible to include these in this estimate because there is not sufficient data to isolate those child deaths that are attributable to CSA. Thus, this is an underestimate of the cost of the physical harms that result from CSA.

The physical and emotional harm costs presented here are non-financial (monetised) costs as these costs use notional (non-market) values which represent estimated harm in monetary terms. These costs are realised over a period of up to 7.17 years, with the duration of harm differing for different types of physical and emotional harm (see section 5.1.4 below for further detail). The contribution to the unit cost is a ‘bottom-up’ cost as it is a direct cost per victim measure.

5.1.2 Estimating the likelihood of physical and emotional harm

The physical and emotional harms to children due to rape and sexual assault are likely to be considerable. For example, Fisher et al. (2017) found that experiencing CSA is associated with a wide range of adverse physical health outcomes. Penetrative sexual abuse can result in acute physical injuries to the genital area, as can sexually transmitted infections. These physical impacts may be long-lasting: research suggests that individuals with a history of CSA have a greater number of doctor and hospital contacts than those who have not experienced CSA. There can also be considerable emotional impacts, from those such as depression and anxiety, but also intimacy issues within relationships, trust issues, and feelings of shame and internalised self-blame (Fisher, et al., 2017). These physical and emotional harms will of course vary depending on the type of sexual abuse, the frequency and the duration, and the impacts will differ for adults compared to children. The age of the child and their developmental stage will also affect the impact of the sexual abuse, both in the short and long term.

The first step in estimating the average physical and emotional harm is to estimate the likelihood of a victim of rape and or sexual assault experiencing physical and or emotional injuries. However, despite the wide body of literature in this area, the available data is not in the appropriate format for incorporating into this economic work. Data from the CSEW on adult likelihood of injury as a result of rape and or sexual assault is used as a proxy. It is therefore likely that this approach will result in an underestimate of the costs of physical and emotional harms to victims of contact CSA.

The proxies for likelihood of sustaining injuries due to rape and or sexual assault came from research on victims of domestic abuse, who experienced rape and or sexual assault (Oliver et al., 2019)^{[footnote 15]}. Tables 5 and 6 (below) show the assumed likelihoods of victims experiencing physical and emotional harms, as a result of both rape and sexual assault. For example, victims of rape are deemed to have a 29% likelihood of minor bruising, while victims of sexual assault have a 3% likelihood.

Table 5: Likelihood of suffering physical injury as a result of rape or sexual assault (Oliver, et al., 2019)

Injury	Rape	Sexual assault
Minor bruising	29%	3%
Severe bruising	5%	3%
Scratches	11%	3%
Cuts	24%	0%
Stabbed	0%	0%
Broken bones	0%	3%
Nose bleed	0%	0%
Broken nose	0%	0%
Lost teeth	0%	0%
Chipped teeth	0%	0%
Dislocation	0%	0%
Concussion	6%	0%
Internal injury	0%	0%
Facial injury	4%	0%
Eye injury	0%	0%
Other	0%	0%

Source: CSEW data for year ending 31 March 2017 for rape and sexual assault

Table 6: Likelihood of sustaining an emotional injury due to rape and sexual assault (Oliver, et al., 2019)

Injury	Rape	Sexual assault
Fear	66%	25%
Depression	24%	9%
Anxiety/panic attacks	46%	20%
Drug abuse	5%	0%
Alcohol abuse	5%	0%
Obesity/eating disorder	5%	0%

Source: CSEW data for year ending 31 March 2017 for rape and sexual assault

5.1.3 Quality-adjusted life loss associated with physical and emotional harm

To calculate the percentage reduction to a person’s quality of life (QALY loss), ‘disability weights’ are used. These weights estimate the percentage loss in quality of life that is caused by an injury. For example, a broken rib leads to a 10% reduction in their quality of life whilst they are recovering (Salomon et al., 2015). The most recent disability weights come from Salomon et al. (2015). Table AP1 in Appendix A shows the assumed QALY losses associated with the various physical and emotional injuries experienced as a result of rape and sexual assault. Where the injuries set out by Salomon et al. (2015) do not exactly match the injury suffered, an alternative source has been used.

5.1.4 Duration and frequency of physical and emotional harm

In estimating duration of the emotional and physical harm, there are three aspects to be considered. The first is the time needed to recover from physical and emotional injuries and therefore how for how long the percentage QALY loss should be applied. The second is the duration that the average contact CSA case lasts, as emotional harms are applied throughout the sexual abuse period as well as the recovery period. The final aspect is the frequency with which victims experience different physical and emotional injuries.

Data on the average duration of a contact CSA case is drawn from the CSEW. In the survey, length of sexual abuse can range from ‘less than one year’ to ‘14 years or more but less than 16 years’. For inclusion in this report, mid-points were used in all categories except ‘less than one year’. For this category, a value of one day was used. This conservative assumption was made in the absence of evidence to suggest a longer period. The duration of an average case is therefore estimated at 2.17 years, although because of the conservative assumption regarding offences lasting ‘less than one year’ this is likely to be an underestimate.

Most data on the lengths of the recovery periods from the different physical and emotional injuries were then calculated, using research from Dolan et al. (2005). For other injuries where Dolan et al. (2005) did not provide the duration, other sources are used, or assumptions made. Table AP2 in Appendix 1 outlines the specific sources and durations used within this report.

The frequency with which victims experience physical injuries also needs to be calculated, for both rape and sexual assault. This is used in the Healthcare costs section (section 5.2.1) where it is combined with the likelihood of sustaining an injury to give the average number of injuries that occur in a contact CSA case. Data on the number of rapes and/or sexual assaults experienced was taken from the CSEW, which split responses into four categories: once, 2 to 5 times, 6 to 19 times, or 20 times or more. While midpoints were used for the first three categories, for ‘20 times or more’ a conservative assumption was used that capped this at a value of 20. The average number of rapes was therefore estimated at 8.76 rapes in an average case, and an average 6.46 sexual assaults in an average case. Due to the conservative assumption on the highest threshold, this is an underestimate of the true frequency of rapes and sexual assaults within contact CSA cases. This underestimate of the frequency of rapes and sexual assaults leads to an underestimate of the true cost of physical and emotional harms suffered by victims.

5.1.5 Value of a life year

The value of a life year (VOLY) represents the value of one year of good health. HM Treasury sets this value at £60,000 in 2012 prices (HM Treasury, 2018). To make comparable to other data sources the VOLY is uprated to 2018 prices giving a value of approximately £72,000. Additionally, to account for the impact of contact CSA on underage victims, the physical and emotional harms suffered by victims have been weighted using a multiplier of 1.21. The multiplier is based on evidence from a study by the Minnesota Department of Health (2007) which compared the adult and child costs for rape and sexual assault and showed that the physical and emotional costs are relatively worse for children than adults.

5.1.6 Calculating the cost of physical and emotional harms

As set out in Section 5.1.1, the cost of physical and emotional harms is calculated by:

LIKE×REDUCEQL×DUR×VOLY×1.21=Average physical and emotional cost

To calculate the cost of each of the harms the likelihood of sustaining an injury is multiplied by the reduction in the quality of life, the duration of the injury, the value of a life year and the 1.21 multiplier (see Section 5.1.5). The specific costs of each individual injury have been summed to give the cost of the physical and emotional harms sustained due to rape and sexual assault. When calculating the harm for cases that include both rape and sexual assault, the report uses the methodology of Oliver et al. (2019) in that the physical harms are summed as usual but only the emotional harms with the highest likelihood (rape) are used. The results are shown in Table 7.

Table 7: Costs of physical and emotion harms for rape, sexual assault, and contact CSA overall

Type of victimisation	Emotional	Physical	Total unit cost
Contact CSA – rape	£90,450	£470	£90,915
Contact CSA – rape and sexual assault	£90,450	£695	£91,145
Contact CSA – sexual assault	£23,190	£230	£23,420
Contact CSA – average harm	£45,610	£375	£45,985

The costs of contact CSA were calculated using the data from CSEW on the proportions of victims of contact CSA who experienced rape only, rape and sexual assault, and sexual assault only. Of contact CSA victims, 3% experienced rape only, 30% experienced rape and sexual assault and 67% experienced sexual assault only (Office for National Statistics, 2020a). The costs are multiplied by these proportions and then added together to form an estimate for the average physical and emotional harm of a contact CSA case.

5.1.7 Contribution to the total cost and unit cost

Physical and emotional harm contributes £45,985 per victim to the unit cost. When this is multiplied by the prevalence figure, it is estimated that the total physical and emotional harm is £5,203.1 million (£5.2 billion) over the victims’ lifetime.

5.2 Healthcare costs

5.2.1 Approach

This area captures the cost to the NHS and other healthcare providers of responding to and providing care for the physical and emotional injuries associated with contact CSA. The cost is captured by multiplying the likelihood of sustaining an injury by the cost to the NHS of treating said injury. The total cost of healthcare is found by combining the cost of treating physical injuries with the cost of treating emotional injuries.

This relies on the data showing the likelihood of sustaining injuries set out in Section 5.1.2, in order to calculate the likelihood that victims sustain injuries and thus need treatment. The likelihood data gives the likelihood of sustaining a physical injury within one rape or sexual assault. In order to estimate the number of physical injuries that occur within an average contact CSA case the likelihood of injury is multiplied by the frequency of rape and sexual assault in an average case of contact CSA (as described in section 5.1.4.). This is shown in Table 8 (below). Where the number of injuries exceeds one, the victim is likely to sustain this injury multiple times.

The likelihood of sustaining emotional injury is assumed to remain the same regardless of the number of incidents of rape or sexual assault in a case of contact CSA. Therefore, no multiplier is used, and the original data is used to estimate the emotional injuries sustained. This is a highly conservative assumption as being raped or sexually assaulted multiple times is likely to have a cumulative effect on the emotional harm sustained.

Table 8. Estimate of the number of physical injuries a contact CSA victim sustains

Injury	Number of injuries sustained due to rape	Number of injuries sustained due to sexual assault
Minor bruising	2.6	0.2
Severe bruising	0.4	0.2
Scratches	1.0	0.2
Cuts	2.1	0.0
Stabbed	0.0	0.0
Broken bones	0.0	0.2
Nose bleed	0.0	0.0
Broken nose	0.0	0.0
Lost teeth	0.0	0.0
Chipped teeth	0.0	0.0
Dislocation	0.0	0.0
Concussion	0.5	0.0
Internal injury	0.0	0.0
Facial injury	0.3	0.0
Eye injury	0.0	0.0
Other	0.0	0.0

Within this report it has been assumed that all injuries are treated by a healthcare provider, although it is of course likely that this is not the case, given what is known about the under reporting and under identification of contact CSA. When calculating the costs of treating injuries, separate methods were used for physical injuries and emotional injuries. For physical injuries, the treatment costs are from the NHS reference costs (Department of Health and Social Care, 2020). Using data from Oliver et al. (2019), the cost of an ambulance journey was included for injuries that required ambulance attendance.

The treatment provided for emotional injuries was assumed to be counselling, in line with previous Home Office costs of crime reports. The cost was estimated using the average number of hours of counselling needed to recover (Dubourg et al., 2005) with the cost per hour of counselling (Curtis and Burns, 2016, uprated to 2018/19 prices).

The healthcare costs presented here are financial costs (as there is a transfer of cash in exchange for treatment) which are realised over the 2.17-year abuse period. The contribution to the unit cost is a ‘bottom-up’ cost as it is a direct cost per victim measure.

5.2.2 Calculating the healthcare cost

The healthcare costs for physical injuries arising from contact CSA were found by multiplying the unit costs of the treatment for each injury (Table AP3) by the estimated number of injuries that occur within an average contact CSA case (Table 8). Where victims experienced both rape and sexual assault, the healthcare costs were calculated by combining the physical injuries for rape and sexual assault, to represent the fact that the victim sustains injuries from both rape and sexual assault.

For emotional injuries, the cost of an hour of counselling was multiplied by the estimated hours of counselling needed to recover from the emotional injury. The estimated number of hours of counselling needed are based on that needed for the emotional injury in general, rather than being specific to victims of contact CSA. For example, Dubourg et al. (2005) estimates that 20 hours of counselling are needed to treat depression. This is likely an underestimate of the true amount of time needed to treat emotional injuries due to contact CSA, however more specific data is not available. The costs for treating each emotional injury are then multiplied by the likelihood of sustaining the injury (Table 6, above), giving an average cost of treating the emotional injuries of each type of contact CSA (rape only, rape and sexual assault, sexual assault only). Where a victim experiences both rape and sexual assault only the likelihoods for rape are used, to avoid double-counting.

The results of these calculations can be found in Table 9 (below). The calculation of the healthcare costs of contact CSA is a weighted average based on the proportion of victims that experience rape only, rape and sexual assault and sexual assault only, as with physical and emotional harm.

Table 9: Healthcare costs of rape, sexual assault and contact CSA

Type of harm	Emotional injury cost	Physical injury cost	Total healthcare cost
Contact CSA – rape only	£1,190	£1,260	£2,450
Contact CSA – rape and sexual assault	£1,190	£1,750	£2,940
Contact CSA – sexual assault only	£380	£490	£870
Contact CSA – average harm	£650	£895	£1,545

As discussed previously, the estimated healthcare costs are likely to be an underestimate of the healthcare costs due to contact CSA. This is in part due to the use of adult injury likelihoods, which may underestimate the harm caused to children by contact CSA (and therefore the healthcare costs of treating such harms), as well as due to the omission of certain specialist healthcare services such as Sexual Assault Referral Centres (SARCs)^{[footnote 16]}.

5.2.3 Contribution to the total cost and unit cost

The costs associated with treating the injuries of contact CSA contribute £1,545 per victim to the unit cost. When multiplied by the prevalence this is a healthcare cost of £174.6 million in the year ending 31st March 2019.

5.3 Lost output

5.3.1 Approach

Lost output costs aim to estimate the reduced earnings or potential future earnings of an individual, due to the increased likelihood of unemployment experienced by victims of contact CSA. Disrupted schooling and emotional harms as a consequence of CSA may have a long-term impact on the likelihood of stable employment over the victim’s working life (Fisher, et al., 2017).

A different approach has been used to that in previous Home Office costs of crime reports. Other reports, such as Heeks et al. (2018), calculate lost output on the basis of time lost at work and the reduction in productivity as a consequence of being a victim of crime. However, as victims of CSA are not of working age, this is not an appropriate or feasible approach.

Instead, lost output is calculated by estimating the difference between unemployment rates of CSA victims and non-victims, showing the increased likelihood of unemployment. The average UK wage is multiplied by the rate of unemployment, and then multiplied by the increased likelihood of unemployment due to CSA. This equals the loss of output per year, as shown below:

Average Wage×Rate of Unemployment×Increased Risk of Unemployment due to CSA
=Loss of a years output due to CSA

The ONS estimate for average yearly earnings for an employee is used (Office for National Statistics, 2019) to provide the average yearly income for those who are employed. We also use Eurostat’s estimate of the working life of an average UK worker in 2019 (EuroStat, 2020).

The effect of CSA on unemployment was calculated using analysis of the Adult Psychiatric Morbidity Survey (APMS)^{[footnote 17]} (McManus, 2020). The APMS is a survey about the prevalence of both treated and untreated psychiatric disorder in the English adult population (aged 16 and over). It includes questions about sexual victimisation in childhood, enabling data analysis on different outcomes, such as unemployment, between those who reported experiencing sexual victimisation and those who did not. After controlling for demographic and socioeconomic differences, CSA was found to increase the likelihood of unemployment. The exact reasons for this are unknown, however Fisher et al. (2019) suggest “CSA has been associated with an overall reduction in educational engagement and attainment at school and in higher or further education. In some individual cases, however, it has also been linked to increased attainment” (p. 99).

It is assumed that the income and length of working life do not differ for CSA victims compared to the rest of the population, because income while in employment was not significantly different for those who reported experiencing CSA compared to those who did not within the AMPS analysis. This assumption does mean that this could be an underestimate or an overestimate of the true lost output due to contact CSA depending on the relationship between contact CSA and lifetime income.

The lost output costs presented here are non-financial costs (as these costs are not direct cash and budgetary costs associated with CSA). These costs are realised over a period of 41.4 years. This includes a 4-year time lag between the assumed average age of a CSA victim (14) and the assumed beginning of work upon turning 18. At this time, costs are realised over the subsequent 37.4 years (the average length of a working life from age 18). See section 5.3.2 below for more detail. The contribution to the unit cost is a ‘bottom-up’ cost as it is a direct cost per victim measure.

5.3.2 Calculating the cost of lost output

To estimate the lost output caused by CSA, a representative average victim age of 14 is taken^{[footnote 18]}. It is assumed that this representative victim will begin working at age 18 and will earn the average annual earnings of a UK employee in 2018/19 of £24,937 (Office for National Statistics, 2019). Wages were uplifted by 1.5% each year to reflect lifetime pay increases in line with the OBR estimate of long-run labour productivity growth^{[footnote 19]}. Wages were also adjusted to reflect discounting, as laid out in the HM Treasury’s ‘The Green Book’ (2018).

The rolling average UK unemployment rate over the last 40 years was 7.43%, (Office for National Statistics, 2020c). This is multiplied by the greater likelihood of CSA victims being unemployed. This was found to be 25% higher than the total UK population (which includes both CSA victims and non-CSA victims) by McManus (2020)^{[footnote 20]}. It is assumed that this leads to an average 2% rise in lifetime unemployment rate for contact CSA victims. By multiplying the average annual earnings by the average unemployment rate and then the 25% increase for contact CSA victims, the loss of a year’s output can be calculated.

This was then multiplied by the length of a person’s working life which is estimated to be 37.4 years from the age of 18 (EuroStat, 2020) to calculate the loss over the victim’s working life^{[footnote 21]}. It is estimated that the lost output due to contact CSA is £11,550, over a victim’s lifetime.

5.3.3 Contribution to the total cost and unit cost

The costs associated with lost output contribute £11,550 per victim to the unit cost. When multiplied by the prevalence this is a total lost output cost of £1,306.6 million (£1.3 billion) over the victims’ lifetime.

5.4 Victim services

5.4.1 Approach

Victim services costs are the costs of providing support to victims of contact CSA, as well as to their friends and family. From an economic perspective, in the absence of contact CSA, resources used for victim services could be used for other productive activities. Victim service costs therefore represent a cost to society as a result of contact CSA.

As in Heeks et al. (2018), there are two types of costs associated with victim services. The first is the total expenditure on victim services and the second is the opportunity cost of volunteer time. As with educational and offender prevention costs (Section 4.1, 4.2), a survey was used to collect data from the voluntary sector on the number of volunteers they have supporting victims, the number of hours volunteered per month, and their spending in this area. This was sent to 56 victim-supporting charities, and 41 Rape Crisis centres. Responses were received from 12 of the victim-focused charities, and eight Rape Crisis centres.

As many charities did not respond to the survey, the data provided by the Rape Crisis centres and the other charities was used to scale up the data, to provide an estimate of costs to the whole sector, using the median figures for spending and volunteer hours.

The data showed that the amount of spend and volunteer numbers/hours reported by different Rape Crisis centres was fairly similar, whereas the spending of the other charities varied considerably. Thus, the decision was taken to treat Rape Crisis centres and ‘other charities’ separately within this report when extrapolating spend and volunteer numbers/hours for the sector.

The cost of volunteer time was estimated using the Department of Transport value of non-working time. This is a proxy for the time the volunteers would otherwise have at their leisure. This value of non-working time was then multiplied by the number of volunteer hours to reach an estimate for the total cost of volunteer time.

The two types of cost are calculated separately and then added together to give an estimate of the total spending on victim services due to contact CSA. Some of the support given to victims will not be captured within these surveys as victims may also be supported by non-CSA charities. For example, a charity that offers support more broadly, such as Samaritans, may provide valuable support to CSA victims but this cannot be costed. Thus, the estimated cost of victim services is considered to be an underestimate of the true cost to the economy.

It should also be noted that although the contribution to the unit cost presented in this section divides the total cost of these victim services by the prevalence of contact CSA, not every contact CSA victim has received support from victim services. The contribution to the unit cost represents the average (mean) cost over all CSA victims, rather than the cost per victim that actually received support from victim services.

The victim services costs presented here consist of both financial costs and non-financial costs. Spending on support programmes represents a financial cost (as there is a transfer of budget between organisations). The cost of volunteer time represents a non-financial (monetised) cost as this is the ‘notional’ estimate of the cost in monetary terms. These costs are realised over one year (in the year ending 31st March 2019). The contribution to the unit cost is a ‘top-down’ cost. It has been estimated by dividing the total victim services cost by the prevalence estimate.

5.4.2 Calculating the cost of victim services

As above, of the victim-focused charities contacted 12 provided data (out of 56) along with eight Rape Crisis centres (out of 41). This represents a response rate of approximately 20%^{[footnote 22]}.

The total reported spend of victims charities and Rape Crisis centres that responded to the survey was £6,351,325.

As above, to estimate the spend for charities that did not respond, median values were used. The median spending by victim-focused charities on victim services for a year was £222,500, and the median spend by Rape Crisis Centres on victim services for a year was £242,500. Based on these values, the spend for the remaining victim-focused charities was estimated to be £9,790,000, and an estimated £8,002,500 for the remaining 33 Rape Crisis centres. Adding together reported spending with extrapolated spending, it is estimated that total spending by charities and Rape Crisis centres on victim services is £24,143,825.

The same survey was used to estimate the number of volunteer hours dedicated to contact CSA. The reported hours of victims’ charities and Rape Crisis centres was approximately 54,800 hours a year. Using the median values for the charities (1,818 hours in the year ending 31st March 2019) and Rape Crisis centres (1,878 hours in the year ending 31st March 2019), it is estimated that non-responding charities had 79,992 volunteer hours in the year ending 31st March 2019 and Rape Crisis centres had 61,974 hours. Adding extrapolated and reported hours, it is estimated that approximately 196,800 hours of volunteer time were devoted to contact CSA in the year ending 31st March 2019.

Using the estimate of the costs of non-working time, which the Department for Transport estimated to be £5.74 per hour in 2018 in 2018 prices. Multiplying this cost per hour by the number of hours, the cost of volunteer time is estimated to be £1.1 million in the year ending 31st March 2019.

While a large number of charities were invited to complete the survey, it was of course not possible to identify and contact every one that may work in this area. In addition, as not all charities responded, this estimate of spending is likely an underestimate of the true level of spending by voluntary sector organisations. Therefore, this is likely an underestimate of the full cost to the voluntary sector.

5.4.3 Contribution to the total cost and unit cost

The cost associated with victim services is estimated to be £25.3 million in the year ending 31st March 2019 and when this is divided by the prevalence this gives a contribution to the unit cost of £225 per victim.

5.5 Contribution of the costs as a consequence to the total cost and unit cost

The total contribution of the costs as a consequence of contact CSA to the overall cost is £6.7 billion over the victims’ lifetime. This area of cost contributes £59,320 per victim to the average unit cost of a contact CSA victim.

6. Estimating the costs as a response to contact child sexual abuse

This chapter discusses the methodological approach and outlines the estimates of the costs as a response to contact CSA. These include:

1. Police costs
2. Court and trial costs
3. Prison and probation costs
4. Safeguarding costs

6.1 Police costs

6.1.1 Approach

This section captures the cost to the police of investigating contact CSA. This is done by multiplying the cost of an individual investigation into contact CSA by the number of investigations. Therefore, the first thing that needs to be estimated is the cost of an individual investigation.

To estimate the cost of an investigation the approach of Heeks et al. (2018) was used. This uses the activity-based costing (ABC) system, which uses a comparison between the time spent on different crimes (most notably burglary and all other crimes) to estimate the share of the police budget they take up and thus estimate the cost of an individual crime. The ABC gives costs of investigations for sexual offences which are split into two categories: rape, and other sexual offences (OSO).

There are then three steps to adjust from the ABC cost of investigations to the costs of contact CSA investigations: an update to reflect increases in police time spend dealing with these crimes, wage costs uprating, and an adjustment to take into account the longer investigation time of CSA offences.

To reflect increases in the time police spend dealing with sexual offences, data on the time share of each crime a police force deals with was used. This allowed a comparison between the share of time spent dealing with sexual offences in 2006/7 and 2016/17 and showed the time share of sexual offences had increased. Therefore, the costs of rape and OSO were adjusted accordingly. These updated costs were included within Oliver et al. (2019).

As these costs are still in 2016/17 levels, these costs were uprated using police wage changes between 2016/17 and 2018/19 to reach an estimate of the cost of adult rape and OSO investigations in 2018/19.

These 2018/19 adult rape and OSO investigation costs were then amended to represent contact CSA cases using data on how much longer CSA cases take than adult cases. This gives an estimate of the cost of contact CSA rape and OSO investigation costs that can be used to estimate police costs.

To calculate the total police costs, the cost of a rape/OSO investigation is multiplied by the number of police recorded contact CSA offences in the year to 31st March 2019.

The police costs presented here are non-financial (monetised) costs as they are the ‘notional’ estimate of the value of police time in monetary terms. These costs are realised over one year (in the year ending 31st March 2019). The contribution to the unit cost is a ‘top-down’ cost. It has been estimated by dividing the total police costs by the prevalence estimate.

6.1.2 Calculating the police costs

Taking the ABC costs from Heeks et al. (2018) and adjusting for the increased time share of sexual offence investigations, the cost of an adult rape and OSO investigation is estimated to be £20,580 and £8,860 in 2016/17.

These costs then need to be uprated to 2018/19 prices. This was done using the Annual Survey of Hours and Earnings (ASHE) (Office for National Statistics, 2019) to show increases in police wage costs. After uprating, the cost of a police investigation into rape is estimated to be £21,170 and an OSO investigation is estimated to be £9,115.

In order to convert the adult investigation costs into costs for contact CSA cases, the length of time from offence to completion was used. This showed that CSA cases took 17% and 29% longer for rape and OSO respectively. Thus, the costs were increased by 17% and 29% to give costs of £24,795 and £11,760.

The estimated police costs are calculated by combining the cost of rape investigations and the cost of OSO investigations. In the year to 31st March 2019, police recorded 63,621 contact CSA offences. Of these 17,478 were rape offences against children (Home Office, 2020). Multiplying these by the investigation cost of a rape (£24,800) estimates the total cost of police investigations into CSA rape offences to be £433.4 million.

In the year to 31s March 2019, there were 46,143 OSO offences against children (Home Office, 2020). Multiplying these by the investigation cost of OSO (£11,760) estimates the total cost of police investigations into CSA OSO offences to be £542.5 million.

6.1.3 Contribution to the total cost and unit cost

Police investigations into contact CSA have a total cost of £975.9 million in the year to 31st March 2019. When divided by the prevalence, this gives a contribution to the unit cost of £8,625 per victim.

6.2 Court and trial costs

6.2.1 Approach

This section estimates the cost to the criminal justice system (CJS) due to holding trials for those proceeded against for contact CSA. The cost of trials can be split into four major parts:

• jury service costs
• Crown Prosecution Service (CPS) costs
• Crown Court costs
• legal aid costs

This trial cost is then multiplied by the number of cases proceeded against for contact CSA offences.

The approach taken by Heeks et al. (2018) was used to estimate the cost of jury service. This multiplies the number of hours in Crown Court for the jury by the cost of that jury. The cost of holding a jury is made up of jurors’ lost time while the court case is in progress. There was no data available for the hours at Crown Court for contact CSA offences specifically, so the median number of hours for adult sexual offences was used as a proxy^{[footnote 23]}. The lost time cost of serving on a jury was calculated by multiplying the average hourly wage by the employment rate. To reflect that some jurors will not be in employment, the cost of leisure time was used instead of average hourly wage and multiplied by the percentage of the population not in employment. These costs for employed and non-employed jury members were added to give the cost of the average member of jury and then multiplied by 12 to reach the cost of a jury in a contact CSA case.

CPS costs reflect the cost of prosecuting a case in the Crown Court. This is based on the cost of prosecuting any crime at the Crown Court level uprated from 2012/13 to 2018/19 prices, rather than contact CSA specifically (Home Office, 2014).

There are also Crown Court costs associated with holding a trial. These are the judicial and staff (non-CPS) costs, as well as the cost of maintaining the court. The costs in Home Office (2014) were uprated to 2018/19 prices and then multiplied by the average hours in Crown Court for sexual offences (as above, these are used as a proxy for contact CSA).

Legal aid costs represent the cost of representation for the defendant on trial. As there is no data on how many contact CSA defendants receive legal aid, it is assumed that all do. Legal aid has lower costs attached than non-legal aid representation, and so this is likely to result in an underestimate of the true costs.

There are additional trial costs associated with trials for vulnerable victims, such as video links and registered intermediaries, that have not been included within this costing. Additionally, much of the trial cost is based around costs for average offenders rather than contact CSA offenders. This means that the trials and court cost is likely an underestimate of the true cost of proceedings for contact CSA.

The court and trial costs presented here consist of both financial costs and non-financial costs. The costs of holding a trial and prosecuting a case represent a financial cost (as there is a transfer of budget between organisations). The lost time cost of serving on a jury represents a non-financial (monetised) cost as this is the ‘notional’ estimate of the lost time cost in monetary terms. These costs are realised over one year (in the year ending 31st March 2019). The contribution to the unit cost is a ‘top-down’ cost. It has been estimated by dividing the total court and trials cost by the prevalence estimate.

6.2.2 Calculating the court and trials cost

Total trial costs are calculated by combining the four areas outlined in the previous section.

The jury cost is found by multiplying the length of a trial by the value of the lost time for the jury (wages missed or leisure time foregone). The median length of a Crown Court trial for a sexual offence was 6.1 hours (Ministry of Justice, 2019a). The lost time cost for working jurors is found by multiplying the UK employment rate for 2019 (76.1%) (Office for National Statistics, 2020c) by the average hourly wage (£13.27) (Office for National Statistics, 2019). The lost time cost of a non-employed juror was found by multiplying the non-employment rate (23.9%) by the average cost of non-working time (£5.74) (Department for Transport, 2020). These costs were added to find the cost per hour of the average jury member, which was multiplied by 12 to give a whole jury. This was then multiplied by the time in court (6.1 hours) to give an estimate of the jury cost of a contact CSA trial, which is £840.

Home Office (2014) estimated the cost to the CPS of an average Crown Court trial was £2,600 (in 2012/13 prices). When uprated to 2018/19 prices, this was estimated to be £2,885. This is likely an underestimate as it is for any Crown Court cases and not contact CSA offenders specifically.

The Crown Court cost is related to the judicial and staff costs of a Crown Court trial. Home Office (2014) estimates the cost to be £320 per sitting hour (2012/13 prices). This is multiplied by the median hours in Crown Court for sexual offence cases (6.1 hours) (Ministry of Justice, 2019a) and uprated to 2018/19 prices. This estimates the Crown Court costs to be £2,165. This may be an underestimate due to the use of median trial length for sexual offences, and that this is a proxy rather than the length for contact CSA offences specifically.

Home Office (2014) is also used to estimate the cost of legal aid within the average Crown Court case. This was estimated to be £5,300 in 2012/13 prices and has been uprated to 2018/19 prices, giving a value of £5,885. As with the cost to the CPS and Crown Court costs, this is not specific to contact CSA and therefore may be an underestimate of the true cost.

By adding all these costs together, the cost of a contact CSA trial is estimated to be £11,775 per trial. This cost is applied to the 3,500 individuals proceeded against in 2018. This gives a total trial cost of £41.2 million in the year ending 31st March 2019.

6.2.3 Contribution to the total cost and unit cost

The cost of courts and trials due to contact CSA was estimated to be £41.2 million in the year ending 31st March 2019, with this area contributing £365 per victim to the unit cost when the total cost was divided by the prevalence.

6.3 Prison and probation Costs

6.3.1 Approach

Prison and probation costs are made up of the cost of those sentenced to custodial and non-custodial sentences respectively. This section does not take into account the cost of licence periods, or the cost of offenders being under notification requirements (commonly referred to as being on the ‘sex offender register).

The cost of custodial sentences is the cost of custody for all those who are sentenced for contact CSA offences within a given year. The total cost of prisons due to contact CSA was found by multiplying the annual cost per prisoner by the average sentence length for CSA offenders in years, and the number of offenders sentenced to custodial sentences in a given year. Cost per prisoner was used to be consistent with previous Home Office publications. As the cost was over multiple years costs were discounted. This ensures the cost remains in 2018/19 prices and was done in accordance to Treasury guidance (HM Treasury, 2018).

Non-custodial sentences refer to suspended sentences and community sentences. The cost for both is borne by the Ministry of Justice in the form of the HM Prison and Probation Service (HMPPS). The cost of non-custodial sentences was calculated using estimates for the average offender (Ministry of Justice, 2013)^{[footnote 24]}. This assumes that the cost of non-custodial sentences for the average offender is the same as that for a contact CSA offender. These were uprated to 2018/19 prices and applied to all those who were given a non-custodial sentence in 2018.

As above, the custodial sentences costed here do not include any periods spent on licence, because data was not available in a format that could be included. It is therefore likely that the prison costs presented here are an underestimate of the total costs.

The prison and probation costs presented here are financial costs (as there is a transfer of budget between organisations to pay for managing and holding a prisoner). These costs are realised over a period of 5.47 years, the average custodial sentence length for an offender sentenced for a contact CSA offence. The contribution to the unit cost is a ‘top-down’ cost. It has been estimated by dividing the total prison and probation costs by the prevalence estimate.

6.3.2 Calculating the prison and probation costs

In the year to December 2018, 1,657 offenders received custodial sentences for contact CSA offences (Ministry of Justice, 2019b). The average sentence length for these 1,657 offenders was 65.6 months or 5.47 years (Ministry of Justice, 2019b). The annual cost per prisoner in 2018/19 was £41,135 (Ministry of Justice, 2019c). The cost has been discounted over the 5.47 year sentence according to ‘The Green Book’ (HM Treasury, 2018). Thus, by multiplying the sentence length (5.47) by the number sentenced (1,657) and the annual cost per prisoner (£41,135) and adjusting for discounting, the total cost of custodial sentences is estimated to be £345.5 million for the year to December 2018.

The estimated cost of a non-custodial sentence is £4,305 in 2013 prices (Ministry of Justice, 2013), and when uprated to 2018/19 prices, this becomes £4,780. This is multiplied by the 293 community sentences and 401 suspended sentences given to CSA offenders in 2018 to give a total cost of £3.3 million.

As both these costs are for the average offender, rather than contact CSA offenders specifically, this may be an underestimate of the true cost of prisons and probation. Furthermore, as above, this does not take into account the cost of licence periods or notification requirements.

6.3.3 Contribution to the total cost and unit cost

The cost of prisons due to contact CSA was estimated to be £348.8 million in the year ending 31st March 2019, with this area contributing £3,085 per victim to the unit cost when the total cost was divided by the prevalence.

6.4 Safeguarding costs

6.4.1 Approach

While the previous costs in response have focused on the response to the offender, this area of cost focuses on the response to victims of contact CSA. Two areas of safeguarding have been costed within this report. The combination of these two areas has been judged to best represent the costs of safeguarding victims. All are part of the Department for Education’s (DfE) and the Welsh Government’s safeguarding role. These areas are:

• Children in Need (CiN) in England or Children Receiving Care and Support (CRCS) in Wales
• Looked After Children (LAC)

CiN or CRCS is a form of safeguarding which aims to help children who are considered “unlikely to achieve or maintain, or to have the opportunity of achieving or maintaining, a reasonable standard of health or development without the provision for them of services by a local authority” (Department for Education, 2018b). This form of safeguarding is the basic level of safeguarding that this report has costed, and the other category is a subset of this category.

LAC is a subset of CiN that details children that are in the care of the local authority. It is defined as “A child is looked after by a local authority if he or she has been provided with accommodation for a continuous period of more than 24 hours” (Hampshire County Council, 2019).

A similar method was used to cost these two areas. First, the number of children in each safeguarding category due to contact CSA was multiplied by the average length of time a child spent in the safeguarding category. This was then multiplied by the cost of a year in each safeguarding category to reach the overall cost of each safeguarding category.

Safeguarding data does not use ‘contact CSA’ as a category, and so to estimate the number of children in each safeguarding category due to contact CSA, an assumption is made. The ONS (2020a) found that of all children experiencing abuse, 29% experienced sexual abuse specifically. This proportion is therefore applied to the data on children experiencing abuse or neglect, to estimate how many experience CSA.

For Welsh LAC cases, the proportion due to abuse or neglect on 31st March 2019 was assumed to reflect same proportion as the proportion starting a plan due to abuse and neglect (Statistics for Wales, 2019a).

The median time spent on each plan was used to estimate the average length of time spent on each plan. This was available for LAC and calculated using interpolation from DfE data for CiN (Department for Education, 2019b).

The annual cost per person of each safeguarding plan was calculated by applying the methodology outlined by Aldaba (2017) to 2018/19 data. Actual spending by local authorities, based on section 251 expenditure lines in Department for Education outturn data, have been aggregated to estimate expenditure on these safeguarding plans for each local authority in England in 2018/19^{[footnote 25]}. This expenditure is adjusted by relative local authority labour costs to obtain real term spending estimates that are less biased by differences in salaries across England.

To calculate the contribution to the unit costs for Wales, the Aldaba methodology was applied to the Welsh safeguarding expenditure lines (based on StatsWales social services expenditure data) which have the same names as the English section 251 expenditure lines. As such, it is assumed that these English and Welsh expenditure lines are equivalent.

The safeguarding costs presented here are financial costs (as there is a transfer of budget between organisations to safeguard children). These costs are realised over a period of 2.21 years, the average length of time spent as a looked after child. The contribution to the unit cost is a ‘top-down’ cost. It has been estimated by dividing the total safeguarding cost by the prevalence estimate.

6.4.2 Calculating the safeguarding cost

On the 31st of March 2019 there were 216,290 children who were categorised as CiN with a primary category of abuse and neglect in England (Department for Education, 2019b). After adjusting for contact CSA as a proportion of abuse and neglect, it is estimated there were 63,628 CiN due to contact CSA.

In Wales, on the 31st March 2019 there were 8,850 children categorised as CRCS (StatsWales, 2020). After adjusting for contact CSA as a proportion of abuse and neglect, it is estimated there were 2,589 CRCS due to contact CSA.

In the same period there were 49,570 LAC in England where abuse was the initial category (Department for Education, 2019c). After adjusting for contact CSA as a proportion of abuse, it is estimated that there were 14,500 children considered LAC due to being at risk of contact CSA.

In Wales, for the same period, there were 4,051 LAC (Statistics for Wales, 2019a). After adjusting for contact CSA as a proportion of abuse and neglect, it is estimated there were 1,185 CRCS due to contact CSA.

The costs of safeguarding are set out in Table 10 below. The total cost has been calculated by multiplying the number of children on the plan due to contact CSA by the annual cost per person on the plan, and the average length of time on the plan, and discounting appropriately.

Table 10: Costs of safeguarding summary

Category of safeguarding	Number of children on plan due to abuse or neglect	Number of children on plan due to contact CSA [1]	Time on plan (years) [2]	Annual cost per person (£, 2018/19 prices) [3]	Total cost (£ million, 2018/19 prices)
Children in Need (CiN) and Children Receiving Care or Support (CRCS) England (CiN)	216,290	63,268	0.88	£8,640	£479.7
Children in Need (CiN) and Children Receiving Care or Support (CRCS) Wales (CRCS)	8,850	2,589	0.88	£11,070	£25.1
Looked After Children (LAC) England	49,570	14,500	2.21	£45,085	£1,413.5
Looked After Children (LAC) Wales	4,051	1,185	2.21	£37,230	£95.4
Total					£2,013.7

[1] Estimated by applying the proportion of victims of child abuse that are victims of contact CSA from ONS (2020a) data.
[2] Estimated using interpolation of Department for Education (2019b) plan length data.
[3] Unit costs estimated by applying the methodology used in Aldaba (2017) to 2018/19 section 251 safeguarding expenditure data. Equivalent Welsh expenditure lines were used to calculate the Welsh unit costs.

It should be noted that children who are experiencing CSA may have significantly different lengths on plans than the averages set out here, and the costs associated with those plans may be different. The assumptions made here to enable the calculation of safeguarding costs may therefore be an underestimate of the cost of safeguarding children as a result of CSA.

6.4.3 Contribution to the total and unit cost

Combining the cost of the two aspects of safeguarding set out above (the cost of Children in Need in England; Children Receiving Care or Support in Wales; and Looked After Children in both England and Wales) gives a total safeguarding figure of £2,013.7 million (£2.0 billion) to the total cost of contact CSA over the victims’ lifetime. When divided by the prevalence, safeguarding contributes £17,800 per victim to the unit cost of contact CSA.

6.5 Contribution of the costs in response to the total cost and unit cost

The total contribution of the costs in response to contact CSA to the overall cost is £3.3 billion. This area of cost contributes £29,295 per victim to the average unit cost of a contact CSA victim.

7. Conclusion

This report estimates the financial and non-financial (monetised) costs relating to all children who began to experience contact CSA, or continued to experience contact CSA, in England and Wales in the year ending 31st March 2019. The total cost is estimated to be at least £10.1 billion (in 2018/19 prices). This estimate includes the costs of this cohort being victimised in previous and future years, in addition to lifetime consequences as a result of experiencing CSA. The unit cost of contact CSA is an estimated £89,240, based on dividing the total cost by the number of victims (113,144). This unit cost represents the average economic and social cost of a single victim of contact CSA.

These estimates illustrate the harm that contact CSA presents to society and the economy. As Table 11 shows the majority of the unit cost is for costs as a consequence of contact CSA, with costs in anticipation being the smallest area. Most of the harm is borne by the victim who bears the cost of physical and emotional harm (£45,985) and lost output (£11,550). Government also bears significant elements of the cost, including costs of safeguarding (£17,800), and police costs (£8,625).

Table 11: Unit costs of contact CSA by each individual cost area (£, 2018/19 prices)

Costs
Costs in anticipation	Educational prevention program costs	£30
	Offender prevention program costs	£20
	CSA training costs	£20
	Subtotal	£70
Costs as a consequence	Physical and emotional harm costs	£45,985
	Healthcare cost	£1,545
	Lost output costs	£11,550
	Victim services costs	£225
	Subtotal	£59,300
Costs in response	Police costs	£8,625
	Court costs	£365
	Prison costs	£3,085
	Safeguarding costs	£17,800
	Subtotal	£29,870
	Total	£89,240

Further to the estimates produced by Saied-Tessier (2014) and Conti, et al., (2017), this report has expanded to include the QALY cost of contact CSA and includes a robust estimate of the costs in response to contact CSA. This report also extends the scope by including a wide range of costs in anticipation of contact CSA, including the cost borne by the third sector particularly in supporting victims, and providing training and preventative education. The addition of the QALY methodology for physical and emotional harms and the lost output estimate leads to the estimate by this report (£10.1 billion) being far higher than the previous estimate of £3.2 billion by Saied-Tessier (2014). Interestingly the estimated unit cost of contact CSA (£89,240) is very similar to the unit cost of child maltreatment in Conti et al. (2017) (£89,390)^{[footnote 26]}.

It is likely that the true cost of contact CSA is higher than is estimated in this report. This is due to the use of conservative assumptions throughout, particularly on the duration of sexual abuse and the number of rapes and sexual assaults in a contact CSA case.

8. Future developments

This economic and social cost of contact CSA report provides a methodological framework for future costings, as well as a method for estimating yearly prevalence. However, there are still a number of improvements that could be made, which could contribute to a future update of this report. Potential areas for development are listed below.

• separate injury data for children

The physical and emotional injuries within this publication are based on CSEW data on the harm suffered by adults during domestic abuse. In future, any update of this report would benefit from data on the likelihood of injuries suffered during contact CSA. This would allow the physical and emotional injuries to be child-specific and thus reflect the harm suffered by contact CSA victims more accurately.

• better data on the length of contact CSA cases

One key part of the physical and emotional harm calculations is the duration of sexual abuse as this is used to estimate the length of time a victim suffers emotional injuries such as depression and anxiety. Within this report, conservative assumptions were used to estimate the length of sexual abuse. Further research in this area could allow for a more accurate estimate that would reflect the true length of contact CSA cases.

• better data on the number of rapes and sexual assaults in a contact CSA case

In this report, conservative assumptions were made for the number of rapes and sexual assaults in a contact CSA case. This is a crucial component in estimating the physical and emotional harm as well as the healthcare costs.

Future research and particularly more granular data within this area would allow for a more accurate estimation of the number of rapes and sexual assaults experienced by contact CSA victims. This would allow for more accurate costing of the harm caused by contact CSA.

• better data on the expenditure of charities, in particular for victim services

For this estimate a survey was used to gauge charity expenditure. However, as a number of charities did not respond, total costs were estimated based on those that did provide data. When updating these costings, it would be useful to have increased data on expenditure to estimate charity costs more accurately.

• expand to consider the long-term health costs of contact CSA

The long-term health costs of contact CSA were one area that this report wished to cost. However, this was not possible due to inconclusive data on the direct relationship between contact CSA and long-term health. Any future data that helps prove direct links between long-term health outcomes and CSA could possibly allow this area to be costed. There is also a lack of information on the impact of contact CSA on sexual and reproductive health in the long term and so this has not been considered in this cost.

• expand to include the cost of non-contact CSA

Within this report contact CSA has been costed, but without exploration of the non-contact side of CSA. Therefore, the true cost of contact CSA remains incalculable due to the potential of crossovers in cost between the two forms of CSA. It would be helpful to develop cost estimates for non-contact CSA in the future, to get an understanding of the true cost of CSA.

• updating the prevalence estimate

Within this report, prevalence was estimated using the work of Radford et al. (2013), which provided an estimate for past year prevalence of contact CSA. An updated prevalence estimate would allow for a more accurate estimation of the number of victims in a year which improves the accuracy of both the total and unit cost.

Appendix A – Tables

Table AP1: QALY losses of physical and emotional injuries

	Injury	Corresponding Global Burden of Disease (GBD) injury[footnote 27]	QALY loss
Physical harms	Minor bruising or black eye	0.25 of broken bones (Dolan et al., 2005)	2.58%
	Severe bruising	0.5 of broken bones (Dolan et al., 2005)	5.15%
	Scratches	0.25 of cuts (Dolan et al., 2005)	0.15%
	Cuts	Open wound: short term, with or without treatment	0.60%
	Puncture or stab wounds	No associated injury in GBD. Broken bones used as an appropriate proxy	5.15%
	Broken/cracked/fractured bones	Fracture of sternum or one or two ribs: short term, with or without treatment	10.30%
	Nose bleed	Open wound: short term, with or without treatment	0.60%
	Broken nose	Fracture of face bone: short or long term, with or without treatment	6.70%
	Broken/lost teeth	0.5 of fracture of face bone: short or long term, with or without treatment (Dolan et al., 2005)	3.35%
	Chipped teeth	0.5 of broken/lost teeth (Dolan et al., 2005)	1.68%
	Dislocation of joints	Dislocation of shoulder: long term, with or without treatment	6.20%
	Concussion or loss of consciousness	0.5 of disability weight for intracranial injury (short term) (Dolan et al., 2005)	11.00%
	Internal injuries	No associated injury in GBD. Severed bruising has been taken as an appropriate proxy	5.15%
	Facial/head injuries (no mention of bruising)	No associated injury in GBD. Cuts has been taken as an appropriate proxy	0.60%
	Eye/facial injuries	Injury to eyes: short term	5.40%
	Other	Other injuries of muscle and tendon (includes sprains, strains and dislocations other than shoulder, knee or hip)	0.80%
Emotional harms	Fear	Anxiety disorders: mild (Ohman, 2008)	3%
	Depression	Major depressive disorder: mild episode	39.6%
	Anxiety/panic attacks	Anxiety disorders: moderate	13.3%
	Drug abuse	Moderate cocaine dependence	47.9%
	Alcohol abuse	Moderate alcohol use disorder	37.3%
	Obesity/eating disorder	Anorexia nervosa/bulimia nervosa	22%

Table AP2: Duration of physical and emotional harms

	Injury	Duration (years)	Duration (days)	Source
Physical harms	Minor bruising or black eye	0.0288	10.5	Dolan et al. (2005)
	Severe bruising	0.0575	21.0	Dolan et al. (2005)
	Scratches	0.006	2.2	Dolan et al. (2005)
	Cuts	0.024	8.8	Dolan et al. (2005)
	Puncture or stab wounds	0.0575	21.0	3 weeks (Advanced Tissue, 2014)
	Broken/cracked/fractured bones	0.115	42.0	Dolan et al. (2005)
	Nose bleed	0.0027	1.0	No source available, assumed 1 day
	Broken nose	0.059	21.5	Dolan et al. (2005)
	Broken/lost teeth	0.0192	7.0	Dolan et al. (2005)
	Chipped teeth	0.0192	7.0	Dolan et al. (2005)
	Dislocation of joints	0.154	56.2	8 weeks (Drukin et al., 2008)
	Concussion or loss of consciousness	0.0335	12.2	Dolan et al. (2005)
	Internal injuries	0.0575	21.0	No source available, assumed the same as severe bruising
	Facial/head injuries (no mention of bruising)	0.024	8.8	No source available, assumed the same as cuts
	Eye/facial injuries	0.0192	7.0	1 week – traumatic iritis (Root, 2010)
	Other	0.0192	7.0	Dolan et al. (2005)
Emotional harms	Fear	1.25	456.3	Norris and Kaniasty (1994) show that fear from crime is still evident after 15 months
	Depression	1	365.0	Dolan et al. (2005): Victims of violent crime who suffer short-term depression do so for 1 year.
	Anxiety/panic attacks	3	1095.0	Dolan et al. (2005): Victims of violent crime who suffer anxiety/panic attacks do so for 3 years.
	Drug abuse	5	1825.0	Dolan et al. (2005)
	Alcohol abuse	5	1825.0	Dolan et al. (2005)
	Obesity / eating disorder	5	1825.0	Dolan et al. (2005)

Table AP3: Average cost of medical requirement per injury

Injury	Average cost of medical requirement per injury
Minor bruising	£0
Severe bruising	£505
Scratches	£0
Cuts	£90
Stabbed	£1,015
Broken bones	£2,410
Nose bleed	£0
Broken nose	£780
Lost teeth	£150
Chipped teeth	£55
Dislocation	£125
Concussion	£1,670
Internal injury	£650
Facial injury	£90
Eye injury	£0
Other	£75
Fear	£110
Depression	£1,090
Anxiety/ panic attacks	£1,360
Drug abuse	£2,720
Alcohol abuse	£2,720

Table AP4: Method of inflating/deflating

Cost area	Cost category	Method of indexation
Costs in anticipation	Educational prevention programme	Survey data in financial year 2018/19 prices
	Offender prevention programmes	Survey data in financial year 2018/19 prices; other data in 2018 calendar year prices
	CSA training costs	Survey data in financial year 2018/19 prices; other data uprated by HMT GDP deflator
Costs as a consequence	Physical and emotional harm	VOLY uprated by GDP per capita to 2018 calendar year prices
	Healthcare cost	Data in financial year 2018/19 prices
	Lost output	Annual wage growth assumed at OBR long-run labour productivity growth rate
	Victim services	Survey data in financial year 2018/19 prices; other data in 2018 calendar year prices
Costs in Response	Police costs	Police costs uprated to 2018/19 prices using ASHE earnings by sectors growth
	Court costs	HMT GDP deflator
	Prison costs	HMT GDP deflator
	Safeguarding costs	Data in financial year 2018/19 prices

References

Advanced Tissue, 2014. ‘The signs and stages of wound healing’.

Aldaba, 2017. Children’s services: spending, 2010-11 to 2015-16, London: Department for Education.

Children’s Commissioner, 2015. Protecting children from harm, London: Office for the Children’s Commissioner.

Children’s Commissioner, 2019. Childhood vulnerability in numbers. [Online] [Accessed 06 July 2020].

Conti, G., Morris, S., Melnychuk, M. and Pizzo, E., 2017. The economic cost of child maltreatment in the UK, London: NSPCC and UCL.

Curtis, L. and Burns, A., 2016. Unit Costs of Health and Social Care 2016, Canterbury: Personal Social Services Research Unit.

Department for Education, 2018a. Statutory guidance: Working Together to Safeguard Children, London: Department for Education.

Department for Education, 2018b. Children in need of help and protection, London: Department for Education.

Department for Education, 2019a. Relationships Education Relationships and Sex Education (RSE) and Health Education, London: Department for Education.

Department for Education, 2019b. Characteristics of children in need: 2018 to 2019, London: Department for Education.

Department for Education, 2019c. Children looked after in England (including adoption), year ending 31 March 2019, London: Department for Education.

Department for Transport, 2020. TAG Data Book, London: Department for Transport.

Department of Health and Social Care, 2020. National Cost Collection for the NHS. [Online] [Accessed 02 July 2020].

Disclosure and Barring Service, 2019. Disclosure and Barring Service Annual Report and Accounts, London: House of Commons.

Disclosure and Barring Service, 2020. DBS dataset 1: Disclosure Progress, Update Service and DBS Basics Information, London: Disclosure and Barring Service.

Dolan, P., Loomes, G., Peasgood, T. and Tsuchiya, A., 2005. Estimating the intangible victim costs of violent crime. British journal of criminology, pp. 958-976.

Drukin, R., Burkhalter, W. and King, J., 2008. ‘A guide to dislocated shoulder’. The Sports Medicine Patient Advisor.

Dubourg, R., Hamed, J. and Thorns, J., 2005. The economic and social costs of crime against individuals and households 2003/04, London : Home Office .

EuroStat, 2020. Duration of working life - statistics. [Online] [Accessed 1 July 2020].

Fell, E., James, O., Dienes, H., Shah, N., Grimshaw, J., 2019. Understanding organised crime 2015/16, London: Home Office.

Fisher, C., Goldsmith, A., Hurcombe, R. and Soares, C., 2017. The impacts of child sexual abuse: A rapid evidence assessment, London: Independent Inquiry into Child Sexual Abuse.

Hampshire County Council, 2019. Children in Care (Children Looked After). [Online] [Accessed 07 July 2020].

Heeks, M., Reed, S., Tafsiri, M. and Prince, S., 2018. The economic and social costs of crime: Second edition, London: Home Office.

HM Treasury, 2018. The Green Book: Central Government Guidance on Appraisal and Evaluation, London: HM Treasury.

Home Office, 2014. Serious Crime Bill - Improvements to the Serious Crime Impact Assessment (lA) Prevention Order, London: Home Office.

Home Office, 2020. Police recorded crime and outcomes open data tables. [Online] [Accessed 13th January 2021].

Lipscombe, S., 2012. Sarah’s law: the child sex offender disclosure scheme, London: House of Commons Library.

McManus, S., 2020. Personal Communication

Ministry of Justice, 2013, Probation Trust Unit Costs Financial Year 2012-13, London: Ministry of Justice.

Ministry of Justice, 2019a. Criminal court statistics quarterly: January to March 2019, London: Ministry of Justice.

Ministry of Justice, 2019b. Criminal justice system statistics quarterly: December 2018, London: Ministry of Justice.

Ministry of Justice, 2019c. Costs per place and costs per prisoner by individual prison, London: Ministry of Justice.

Ministry of Justice, 2020. Criminal justice system statistics quarterly: December 2019. [Online] [Accessed 30th June 2020].

Minnesota Department of Health, 2007. Cost of Sexual Violence in Minnesota, St Paul: Minnesota Department of Health.

Norris, F.H. and Kaniasty, K., 1994. ‘Psychological distress following criminal victimization in the general population: Cross-sectional, longitudinal, and prospective analyses’. Journal of Consulting and Clinical Psychology, vol. 62(1) pp. 111-123.

NSPCC, 2020. Sexual Abuse. [Online] [Accessed 14 August 2020].

Office for National Statistics , 2019. Annual Survey of Hours and Earnings time series of selected estimates. [Online]

Office for National Statistics, 2020a. Child Sexual Abuse in England and Wales. [Online]

Office for National Statistics, 2020b. Analysis of population estimates tool. [Online] [Accessed 23 June 2020].

Office for National Statistics, 2020c. Employment in the UK: July 2020, London: Office for National Statistics.

Oliver, R., Alexander, B., Stephen, R. and Wlasny, M., 2019. The economic and social costs of domestic abuse, London: Home Office.

Ohman, A., 2008. Fear and Anxiety: Overlaps and Dissociations. In: M. Lewis, J. Haviland-Jones and L. Feldman Barret, eds. The Handbook of Emotions . New York : The Guilford Press , pp. 709-729.

Radford, L., Corral, S., Bradley, C. and Fisher, H., 2013. The prevalence and impact of child maltreatment and other types of victimization in the UK. Central Lancashire Online Knowledge, pp. 1-29.

Reed, S., Roe, S., Grimshaw, J. and Oliver, R., 2018. The economic and social costs of modern slavery, London: Home Office.

Root, T., 2010. ‘Eye Trauma’. Timroot.com.

Saied-Tessier, A., 2014. Estimating the costs of Child Sexual Abuse in the UK, London: NSPCC.

Salomon, J. et al., 2015. Disability weights for the Global Burden of Disease 2013 Study. The Lancet, 3(11), pp. e712-e723.

Statistics for Wales, 2019a. Children looked after by local authorities, 2018-19, Cardiff: Welsh Government.

Statistics for Wales, 2019b. Social Services activity 2018-19, Cardiff: Welsh Government.

StatsWales., 2020. Children receiving care and support by local authority and category of need, Cardiff: StatsWales.

1. See the ‘The Green Book’ (HM Treasury, 2018), page 75, for non-market valuation techniques. ↩

2. Note that some of this lost output will represent a cost to the government through lost tax revenue. However, in line with the Heeks et al. (2018) framework, all costs as a result of ‘lost output’ are attributed as costs to the victim. ↩

3. Note that this unit cost should not be multiplied by any other prevalence estimate. ↩

4. In some areas, full costs across the victim’s lifetime were unable to be calculated such as long-term health issues or suicide risks and these costs are thus not estimated in this report. ↩

5. Sums to more than 6% as participants could select more than one category and could have experienced both rape and sexual assault. ↩

6. Child Sexual Exploitation is a subset of Child Sexual Abuse that involves an offender exploiting an imbalance in power and the element of exchange and financial advantage to sexually abuse children. A full definition of CSE can be found in the 2017 Child Sexual Exploitation Guidance. ↩

7. CSODS is a scheme that allows parents, carers and guardians to formally ask the police to tell them if someone has a record for child sexual offences. ↩

8. For example, screens in courts, video links and registered intermediaries. ↩

9. Does not sum to 100% due to rounding. ↩

10. It is important to note that this response rate of 20% is low, and therefore this estimate may not be an accurate reflection of the third sector spend on educational prevention programmes. In particular, three large charities did not respond. However, this is the best available evidence on the costs for charities, although it is acknowledged that it is likely to be an underestimate. ↩

11. For example, because it was not possible to isolate the specific spending/element focused on contact CSA. ↩

12. The College of Policing is responsible for knowledge, education and standards within the police. ↩

13. It is recognised that this may not be a fully accurate estimate. The low response rate means that the estimate may skewed by the small sample who responded. However, these responses are the best available evidence on the costs for charities and Rape Crisis centres. ↩

14. Some who receive this training may work to combat online CSA as opposed to contact CSA. This is therefore likely to be a slight overestimate of the cost of contact CSA-related spending on the child abuse specialist investigators course. ↩

15. This work does not include the likelihood of victims experiencing pregnancy or sexually transmitted infections, because of a lack of data. The costs of these harms are therefore not included. ↩

16. SARCs provide forensic and medical support to those who have experienced contact CSA. ↩

17. More information about the APMS study can be found here: Adult Psychiatric Morbidity Survey: Survey of Mental Health and Wellbeing, England, 2014. ↩

18. The average age of a CSA victim is calculated as 13.8 and rounded to 14. This is based on the Radford (2013) prevalence estimate. ↩

19. Fiscal sustainability report – July 2020. ↩

20. It is recognised that here the baseline (average 40-year unemployment rate) includes CSA victims, when ideally the baseline would be one in which there were no victims of contact CSA. However, it is not possible to estimate a baseline unemployment rate of non-contact-CSA victims. Since the proportion of CSA victims relative to the total population is small, it has been judged that this represents an appropriate proxy for the impact of contact CSA above the baseline of no contact CSA. ↩

21. The EuroStat estimated length of working life is 39.4 for a 15-year-old who is about to enter the workforce at age 16. In this report, it is assumed that victims begin working at age 18, and so this length working life has been adjusted to 37.4 years. ↩

22. It is important to note that this response rate of 20% is low, and therefore this estimate may not be an accurate reflection of the third sector spend on educational prevention programmes. In particular, three large charities did not respond. However, this is the best available evidence on the costs for charities, although it is acknowledged that it is likely to be an underestimate. ↩

23. Median hours were used to prevent the average being distorted by extreme values in the data. ↩

24. The cost is the cost for an average offender over the length of a sentence of average sentence length, rather than a cost per year of sentence length as for custodial sentence costs. ↩

25. ‘Total gross expenditure, excluding capital expenditure’ from section 251 expenditure lines identified in Aldaba (2017) as relevant to these safeguarding plans. Data taken from 2018/19 Department for Education outturn data. ↩

26. Note the different price years here, although the unit cost is still similar. ↩

27. Salomon et al. (2015). ↩