Consultation outcome

Introducing public health measures impact assessment 2019 (accessible version)

Updated 18 August 2021

Title:	Introducing Public Health Measures IA 2019
IA No:	HO0345
RPC Reference No:	N/A
Lead department or agency:	The Home Office
Other departments or agencies:
Date:	04/07/2019
Stage:	FINAL
Source of intervention:	Domestic
Type of measure:	Primary legislation
Contact for enquiries:
RPC Opinion:	RPC Opinion Status

Summary: Intervention and Options

Cost of Preferred (or more likely) Option (in 2016 prices)

Total Net Present Social Value	Business Net Present Value	Net cost to business per year	Business Impact Target Status
£377m	-£25m	£3m	Non-qualifying

What is the problem under consideration? Why is government intervention necessary?

Serious violence is extremely costly to society and instils fear within communities. It has increased since 2014. For example, knife crime rose by 71 per cent between December 2014 and December 2018. The Government’s Serious Violence Strategy recommended a multi-agency approach involving a range of partners but the success of the existing partnerships is mixed. Government intervention legislating for a public sector duty is necessary to ensure that partnerships prioritise tackling serious violence, increasing public safety and helping to safeguard family life.

What are the policy objectives and the intended effects?

The objectives of the duty are to: 1) improve the current multi-agency approach involving partners and agencies such as education, health, social services, housing, youth and victim services. 2) increase the effectiveness of partnerships at preventing/tackling serious violence to share data, intelligence and knowledge to generate evidence-based analysis of the problem and solutions. The intended effect is two-fold: to increase effective collaboration across agencies and ultimately to reduce the number of serious violence offences.

What policy options have been considered, including any alternatives to regulation? Please justify preferred option (further details in Evidence Base)

Option 0: (Do Nothing) the current partnership arrangements would continue with different levels of success and sometimes poor responses. This does not meet the Government’s objectives.

Option 1: Enact primary legislation to place a statutory duty on specified authorities to ensure organisations collaborate and communicate regularly to take effective coordinated action to prevent, and tackle serious violence. A number of studies have shown marked crime-reduction effects when these kinds of partnerships have been implemented. This is the Government’s preferred option. Non-regulatory and other options were considered but did not meet the objectives.

Will the policy be reviewed?

It will be reviewed.

If applicable, set review date:

10/2021

Does implementation go beyond minimum EU requirements?

N/A

Is this measure likely to impact on trade and investment?

No

Are any of these organisations in scope?

Micro	Small	Medium	Large
Yes	Yes	Yes	Yes

What is the CO2 equivalent change in greenhouse gas emissions? (Million tonnes CO2 equivalent)

Traded	Non-traded
N/A	N/A

I have read the Impact Assessment and I am satisfied that, given the available evidence, it represents a reasonable view of the likely costs, benefits and impact of the leading options.

Signed by the responsible Minister: Date: 15 July 2019

Summary: Analysis & Evidence Policy Option 1

Description: Full economic assessment

Price Base Year:	2016
PV Base Year:	2017
Time Period Years:	10

Net Benefit (Present Value (PV)) (£m)

Low	N/A
High	N/A
Best Estimate	377

Costs (£m)

	Total Transition (Constant Price) Years (1)	Average Annual (excl. Transition) (Constant Price)	Total Cost (Present Value)
Low	N/A	N/A	N/A
High	N/A	N/A	N/A
Best Estimate	0.3	33	281

Description and scale of key monetised costs by ‘main affected groups’

Costs are expected to be incurred by partnership organisations that are not currently prioritising or successfully tackling/preventing serious violence, estimated to be around 71 per cent (223) of current community safety partnerships.

For these organisations ongoing labour costs and initial set-up/equipment costs are monetised over 10 years. Total costs are estimated, over 10 years, to be £281 million (PV).

Other key non-monetised costs by ‘main affected groups’

Possible costs to prisons, youth offending teams and housing organisations were not monetised.

Benefits (£m)

	Total Transition (Constant Price) Years (1)	Average Annual (excl. Transition) (Constant Price)	Total Cost (Present Value)
Low	N/A	N/A	N/A
High	N/A	N/A	N/A
Best Estimate	0	77	657

Description and scale of key monetised benefits by ‘main affected groups’

Monetised benefits arise from the reduction in serious violence offences estimated from published evaluations. These are multiplied by the societal cost, using published cost of crime methodology. Only a small number of eligible partnerships (11) are assumed to be successful and accrue benefits. Total benefits are estimated, over 10 years to be £657 million (PV).

Other key non-monetised benefits by ‘main affected groups’

Evidence suggests successful partnership working can reduce fear of crime as well as numbers of offences. This has not been monetised.

Any impacts on non-violent offending have not been monetised.

Key assumptions/sensitivities/risks

Discount rate

3.5%

It is assumed the duty only affects the proportion of partnerships (and associated organisations) that are not currently tackling/preventing serious violence effectively (71% of the total). It is estimated that only 5 per cent of these will show successful benefits. There is a risk that not all partners are motivated to adopt a multi-agency approach, that crime-reduction benefits are not as large as past evaluations have estimated. Both could reduce the NPV of the preferred option.

Business assessment (Option 1)

Direct impact on business (Equivalent Annual) £m:

Costs:	2.9
Benefits:	0.0
Net:	-2.9

Score for Business Impact Target (qualifying provisions only) £m:

N/A

Evidence Base (for summary sheets)

A. Strategic Overview

A.1 Background

Since 2014, certain types of serious violence have increased markedly in England and Wales. Offences involving knives increased by 71 per cent between December 2014 and December 2018. Homicides (excluding terrorism offences) increased by around 30 per cent and gun crime rose by 34 per cent over the same period.

Responding to this increase, the Government published the Serious Violence Strategy in April 2018. The strategy made it clear that a successful approach depends not only on the work of law enforcement but also on harnessing the value of cross-sector partnerships including education, health, social services, housing, youth services and victim services in tackling serious violence.

To enshrine this approach, it is proposed to introduce a new duty through new primary legislation. This would place a duty on specific organisations to plan and collaborate to prevent and tackle serious violence.

A.2 Groups Affected

The proposed duty would affect the following groups:

• Police.
• Local authorities (including responsibilities such as housing, youth offending teams and public health).
• Criminal Justice organisations (including probation services and prisons).
• Health and social care bodies (e.g. clinical commissioning groups; primary care trusts and children’s care homes).
• Education authorities (representatives of schools, alternative provision providers and higher education organisations).

The authorities subject to the duty will be similar to those listed in Schedule 6 of the Counter Terrorism & Security Act 2015 (see Annex 1).

A.3 Consultation

Within Government

A number of government departments agencies were consulted as part of the development of the consultation including:

• Department for Education.
• Department for Health & Social Care.
• Ministry of Justice.
• Crown Prosecution Service.
• Ministry of Housing, Communities & Local Government
• Department for Digital, Culture, Media & Sport.
• Department for Work and Pensions.

Public Consultation

A full public consultation opened on 1 April to seek views on the introduction of a new legal duty to support a multi-agency approach to preventing and tackling serious violence. The consultation closed on 28 May 2019. Responses were particularly welcomed from those with expertise in working with young people at risk of criminal involvement and/or re-offending or victimisation, those involved in law enforcement and, more generally, the communities affected by serious violence including the voluntary and community sector. This included relevant professionals, such as those working in social care, education, law enforcement, local government, community safety, youth services, offender management, victims’ services, public health and healthcare. In recognition of multi-agency approaches outside England and Wales, responses were also welcomed from across the UK.

The consultation sought views on three options:

a) A new duty on specific organisations to collaborate and plan to prevent and reduce serious violence.

b) A new duty through legislating to revise Community Safety Partnerships.

c) A voluntary non-legislative approach to preventing and tackling serious violence.

B. Rationale for intervention.

The problem: serious violence is increasing and though partnership working is an evidence-based response, only a small proportion of partnerships are currently tackling serious violence effectively.

Serious violence has increased since 2014/15. For example, knife crime rose by 71 per cent between December 2014 and December 2018. The rise has continued in the latest figures which show a 6 per cent year-on-year increase in knife offences up to December 2018. Evidence shows that partnership working is an evidence-based response. For example, a Home Office review of partnership working^{[footnote 1]} that looked at nine studies, six of which were evaluations of initiatives designed to tackle serious violence, concluded that:

…on balance, the evidence suggests that the principle of applying partnership working as a component of initiatives to tackle complex crime and disorder problems is effective”. (piii).

A recent systematic review tested the effectiveness of one partnership approach known as focussed deterrence or `pulling levers’. This targets serious offenders with offers of help from a range of services with the threat of enforcement if none of these offers are accepted. The review of 24 evaluations found that the approach was associated with an overall statistically significant, moderate crime reduction effect.^{[footnote 2]} As a result of these and other studies, the World Health Organization (WHO) has advocated a ‘public health approach’ to be taken to violence.^{[footnote 3]}

Legislation is needed to ensure that serious violence is prioritised. Evidence suggests the response to serious violence is mixed under the current framework.

The 2016 review of partnership working to tackle community safety and crime conducted by the Local Government Association (LGA)^{[footnote 4]} found that only about 20 per cent of Community Safety Partnerships (CSPs) and County Safety Groups (CSGs) listed serious violence as a priority.^{[footnote 5]} Only 29 per cent of partnerships stated they were `very confident’ in delivering on their priorities. The summary of the report^{[footnote 6]} also highlighted the variable quality of partnership working:

…variations were reported regarding the strength of local authority relationships with other statutory CSP partners. In some areas there are excellent relationships in place; representation at (and chairing of) meetings is consistent, allowing relationships to flourish; communication is good across partners, and information is shared. However it is clear this is not universal; in other places some partners may be more engaged than others, and there continue to be concerns in some areas about silo working and core issues such as data sharing – which are critical in efforts to ensure public safety. Working with partners who do not share coterminous boundaries presents additional challenges. Given limited resources and pressured budgets across partners, it is even more important to address these concerns.” (p9)

C. Policy objective

The aim is for local areas to take a multi-agency approach to understand the causes and consequences of serious violence, focused on prevention and early intervention, and informed by evidence and rigorous evaluation of interventions. This is often referred to as a ‘public health’ approach. It requires a range of bodies and organisations to work together to tackle this issue including law enforcement agencies, education partners, local authorities, offender management services including youth offending services and health care and public health professionals taking joint action. It would also be expected that those partners look for opportunities to draw in support and wider expertise in their communities, including those in the voluntary and charitable sector.

Briefly summarised, the approach advocates (as advocated by WHO) is:

• Focused on a defined population, often with a health risk in common.
• With and for communities.
• Not constrained by organisational or professional boundaries.
• Focussed on generating long-term as well as short term solutions.
• Based on data and intelligence to identify the burden on the population, including any inequalities.
• Rooted in evidence of effectiveness to tackle the problem.

A multi-agency preventative approach is already being taken forward through the Government’s Serious Violence Strategy. The aim of this legislation is to reinforce that approach and ensure that the partnerships that are currently not functioning as they should, take an active role in reducing serious violence via an evidence-based approach.

D. Description of options considered.

Option 0 is to make no changes.

Option 1 is to legislate. This would create a new statutory duty on authorities similar to those set out in Schedule 6 of the Counter Terrorism & Security Act 2015 (see Annex 1) to plan and collaborate to prevent and tackle serious violence. The duty will be underpinned by guidance to relevant authorities issued by the Secretary of State.

E. Appraisal.

Option 0

This option involves no new activities so there are no additional costs or benefits.

Option 1

General assumptions and data

This policy is appraised over a 10-year period in line with HM Treasury (2018) Green Book guidance. A social discount rate of 3.5 per cent is used to discount future values to give present values (PV) over the period with 2017 used as the present value base year. All other costs and benefits are in 2016 prices unless stated, for example, costs per year. Transition costs are assumed to occur in year 1 only. The other main assumptions are:

1) Of the 305 Community Safety Partnerships, 29 per cent (88) are currently tackling/preventing serious violence successfully. This is taken from the 2016 Local Government Association Report which found that 29 per cent partnerships were very confident in delivering on their priorities^{[footnote 7]}. These currently successful partnerships are assumed to be unaffected by the new duty so incur no costs and benefits.

2) However, the counterpart of assumption 1) is that 71 per cent (223) of partnerships are assumed to be not as effective as they should be. They are therefore assumed to be affected by the duty and hence incur costs.

3) Only 5 per cent of partnerships affected by the duty are assumed to achieve crime reduction benefits. This equates to 11 newly successful partnerships (5 per cent of 223). These partnerships receive benefits as well as costs. The rest just incur costs.^{[footnote 8]}

4) It is assumed that successfully tackling/preventing serious violence does not affect the clearance rate of offences. This implies that there are no costs and benefits to the downstream Criminal Justice System (CJS) other than those caused by the change in the number of crimes.

5) It is assumed that all activity driven by the new duty is additional. In other words, organisations do not cut back resources in other areas in order to tackle/prevent serious violence.

6) It is assumed that non-salary costs are equivalent to approximately 18 per cent of salary costs.^{[footnote 9]}

Costs

The proposed legislation places a duty on five types of organisation. The costs of the policy are indicative of what these organisations may incur if this activity is carried out effectively, in accordance with existing responsibilities. These costs fall into two broad categories:

1. Labour costs: Additional personnel/gross salary costs of individuals required to fulfil the duty. These are generally annually recurring.

2. Non-labour costs (for example, additional equipment, IT, building space, training etc). Some of these recur annually but many are transitional set-up costs.

There is no definitive source of data to estimate these costs. To obtain a best estimate, the following method was used for each different sector:

Local authorities

Cost information was requested as part of the consultation process. Respondents from the organisations on which the duty will fall were asked how much additional cost would be required to fulfil it. There were 55 respondents from Local Authorities, 23 of which provided numerical estimates. This data was cleaned and outliers removed. For labour costs the median value was taken: 1 full-time equivalent (FTE) per local authority. Some respondents also indicated the types of additional roles that would be necessary. This information was combined with earnings values in the 2018 Annual Survey of Hours and Earnings (ASHE) to produce a weighted annual estimate of the salary/overhead for per FTE: £26,169.^{[footnote 10]} An additional 18 per cent was added to account for non-salary costs, see assumption 6. This cost was multiplied by the number of local authorities assumed to be affected by the proposed duty: 247.^{[footnote 11]} This figure was obtained by multiplying the total number of local authorities by the proportion assumed to be affected by the proposed duty (71 per cent, see assumption 2). This gave a total annual labour cost of £7.3 million (in 2016 prices).

It was not possible to monetise the non-labour costs from the consultation responses, so these were taken from an economic evaluation by Florence et al. (2014) of a successful violence-reduction partnership model in Cardiff, published in the British Medical Journal.^{[footnote 12]} The partnership involved police, health and local authority partners and costs for each were split between labour and non-labour and between set-up cost and ongoing costs. The non-labour costs to local authorities were estimated to be £13,600 and were judged to be annually recurring.

The total costs in present value terms over 10 years is estimated to be £57 million.

Police

There were 29 consultation responses from the police, six of which contained numerical estimates of the additional labour cost of the proposed duty on a police force. The median number of FTE required was 18.75. The weighted salary for that FTE, obtained by combining the data from the consultation responses with the unit costs in ASHE 2018, was £29,596. An additional 18 per cent was added to account for non-salary costs, see assumption 6. Multiplying the total by the estimated number of forces affected (71 per cent of the 44 forces, see assumption 2), produces an annual estimate of labour costs of £19.5 million in 2016 prices.

There were not enough responses to the consultation response to accurately estimate non-labour costs to the police. These were taken instead from the Florence et al. (2014) economic evaluation of the partnership model aimed at reducing violence in Cardiff. The study estimated that there would be an first-year set-up cost to police of about £137,000.

The total costs in present value terms over 10 years is estimated to be £152 million.

Criminal Justice organisations

The consultation received responses from three types of criminal justice organisations: prisons, youth offending teams and probation. There was not sufficient data to estimate costs to prisons and youth offending teams, so these costs have not been monetised.

For probation an additional resource estimate was received from the National Probation Service. The additional FTE required (75 per year) was monetised by combining the salary information provided with unit costs from ASHE 2018. An additional 18 per cent was added to account for non-salary costs, see assumption 6. This produced an annual estimate of about £3.0 million (in 2016 prices).

The consultation response indicated there weren’t likely to be any additional non-labour costs falling on the probation sector from the proposed duty.

The total costs in present value terms over 10 years is estimated to be £23 million.

Health and Social Care

There were not enough consultation responses containing numerical information to estimate costs on the health sector. Estimates for the impact on Clinical Commissioning Groups (CCGs) were taken from the Florence et al. (2014) economic evaluation of the partnership model aimed at reducing violence in Cardiff. This partnership was primarily between the police and the health sector so it is an appropriate source for these costs.^{[footnote 13]}

The study estimated that health-related labour costs for the partnership were £10,600 per year. The project operated in Cardiff, a community safety partnership (CSP) area. There are 314 CSPs in England and Wales. The cost was therefore multiplied by the estimated number of partnerships affected by the duty (314 x 71% = 223, see assumption 2). This gave a final annual labour cost of about £2.4 million. The study also provided an estimate for the non-labour costs. This was converted to a national estimate to produce a final estimate of about £23,000 annual non-labour costs.

In addition, the proposed duty is also expected to affect children’s homes. Children in care, or who have been in care, are disproportionately likely to become involved in serious violence either as victims or perpetrators. So closer partnership working between children’s care homes and the other sectors, particularly through sharing of knowledge and data is an important aim of the duty. Costs associated with this are estimated to be similar to the data/knowledge sharing achieved in the Cardiff model (which was between police and A&E departments). This was estimated (above) at around £2.4 million per year. Because 81 per cent of children’s care homes are privately run, 81 per cent of the total cost is assumed to fall on business^{[footnote 14]}.

The total costs in present value terms over 10 years is estimated to be £37 million.

Education

There were not enough consultation responses containing numerical information to estimate costs on the Education sector and the Florence et al. (2014) study could not be used as the Cardiff partnership did not include education organisations.

Therefore, whilst the activity expected of the education sector under this duty will be different, we have estimated the impact based on that of the `Prevent’ duty introduced in 2014^{[footnote 15]}. The estimated labour costs of this were about £1.2 million annually (in 2016 prices), consisting of the FTE required to co-ordinate a suitable response from across the sector for tackling/preventing serious violence. There was also an estimated £0.2 million initial set-up cost and a recurring non-labour cost of about £0.3 million. The total costs in present value terms over 10 years is estimated to be £12 million.

Some of these costs are likely to fall on private schools, colleges, universities and also on privately-run alternative provision providers. Statistics for the Department of Education show that in 2017/18 there were 581,875 children attending independent schools (those that don’t receive funding from DfE)^{[footnote 16]}. Dividing this by the total number of children in schools or alternative provision gives 6.7 per cent, which is used as an estimate for the split of costs between the public and private sector. To estimate the cost to business, the total costs for schools were therefore multiplied by 6.7 per cent. In addition, all the costs to higher and further education were assumed to fall on privately managed organisations. Taken together, this implies a total cost to business for the education sector of around £1.2 million per year.

A summary of these costs is presented in Table 1.

Table 1: Estimated costs for Option 1, by sector, £ million. (2016 prices).

Year	2020	2021	2022	2023	2024	2025	2026	2027	2028	2029	Total
Year of Programme	1	2	3	4	5	6	7	8	9	10
Organisation
Police (labour)	19.5	19.5	19.5	19.5	19.5	19.5	19.5	19.5	19.5	19.5	195.3
Police (non-labour)	0.1	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.1
Total police costs (PV)											151.8
Local Authorities (labour)	7.3	7.3	7.3	7.3	7.3	7.3	7.3	7.3	7.3	7.3	73.4
Local Authorities (non-labour)	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.1
Total local authorities costs (PV)											57.1
Criminal Justice: Probation (labour)	3.0	3.0	3.0	3.0	3.0	3.0	3.0	3.0	3.0	3.0	29.8
Criminal Justice: Probation (non-labour)	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Total criminal justice costs (PV)											23.1
Health (labour)	4.7	4.7	4.7	4.7	4.7	4.7	4.7	4.7	4.7	4.7	47.5
Health (non-labour)	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.2
Total health costs (PV)											37.0
Education (labour)	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	11.7
Education (non-labour)	0.5	0.3	0.3	0.3	0.3	0.3	0.3	0.3	0.3	0.3	3.1
Total education costs (PV)											11.6
Total Costs (2016 prices)	36.4	36.1	36.1	36.1	36.1	36.1	36.1	36.1	36.1	36.1	361.4
Present Value of Costs											280.6

Source: Home Office, own calculations.

Note: Some figures may not sum due to rounding.

Total costs are estimated to be £361 million, which equates to £281 million in present value (PV) terms over the 10-year appraisal period.

Benefits

There is good evidence to show that successful multi-agency approaches can reduce violence, including the most serious types of street violence that have been rising since 2014 in England and Wales, like knife/gun crime. For example:

• A Home Office review of partnership initiatives concluded that the approach was effective. Six out of nine studies, on which this conclusion was based, were aimed at a serious violence.^{[footnote 17]}
• A systematic review of one particular multi-agency approach, known as `focused deterrence’ showed that it has a moderate crime reduction effect.^{[footnote 18]}
• Home Office analysis of the multi-agency approach to violence taken in Scotland in the mid-2000s finds that it is likely to partly explain the lower levels of violence in Scotland relative to England and Wales since 2005. See Annex 2 for details.

None of the above studies used cost-benefit analysis. However, another partnership approach to violence reduction, followed in Cardiff, did monetise the costs and benefits of that partnership approach. These are used to proxy the benefits from the proposed option.^{[footnote 19]}

The basis of the Cardiff model was to bring the police and the health professionals responsible for treating those injured through violence together to examine prevention. This was done by sharing of anonymised police and emergency department data to devise prevention strategies and interventions.

The model was evaluated using a quasi-experimental design where 14 control cities were generated using a Home Office-approved mechanism. It employs a basket of indicators to match areas to similar control areas. Following implementation of the partnership model, the evaluation estimated that serious violence was reduced by 38 per cent in Cardiff compared with the control areas. ^{[footnote 20]} A separate cost-benefit study found that the partnership was highly cost effective. An economic evaluation of the Cardiff partnership approach has been published in the online BMJ journal, Injury Prevention.^{[footnote 21]} This study contains economic details of the programme’s costs and benefits. The method used to estimate benefits was to examine police recorded crime and compare monthly offences before and after the start of the programme in Cardiff and 14 matched control areas. Estimates of effectiveness were then calculated using a difference-in-difference approach with additional controls for unemployment and changes in police numbers across the areas.

Effects, in terms of reductions in violence offences, were monetised using the Home Office cost of crime methodology. This process generated the results below (Table 2 is re-produced directly from the study)^{[footnote 22]}:

Table 2: Costs and benefits from the Cardiff Model.

Year	Total benefits (2003 £)	Health service benefits (2003 £)	Criminal justice benefits (2003 £)	Programme costs (2003 £)	Total benefit cost ratio	Health services benefit cost ratio	Criminal justice benefit cost ratio
2003	2,085,978	409,355	489,884	46,972	44.41	8.71	10.43
2004	5,663,925	1,000,878	1,309,431	58,081	97.52	17.23	22.54
2005	5,272,716	939,580	1,220,456	69,341	76.04	13.55	17.6
2006	5,406,452	970,706	1,252,777	60,035	90.06	16.17	20.87
2007	6,986,631	1,252,780	1,618,628	72,495	96.37	17.28	22.33
Total	23,445,262	4,220,741	5,434,818	285,143	82.22	14.8	19.06

Source: Florence et al. (2014).

For the purposes of this impact assessment the benefits in the first column were used as the estimated benefits for the first five years of any newly successful partnerships. The Cardiff model operated at the community safety partnership level, of which there are 314 in England and Wales^{[footnote 23]}. Assumption 1 states that 29 per cent of these are already tackling serious violence successfully. This means only 223 (71 per cent of 314) are assumed to be eligible for benefits. It is likely that the proposed duty will not transform all of these partnerships into models as successful as Cardiff. It is therefore assumed that just 5 per cent of eligible partnerships achieve the same benefits as the Cardiff initiative. These calculations are summarised in Table 3.

Table 3: Estimated benefits for Option 1, England and Wales (£ million, volume and %).

Year	2020	2021	2022	2023	2024	2025	2026	2027	2028	2029	Total
Year of Programme	1	2	3	4	5	6	7	8	9	10
Benefits per partnership (2003 prices)	2.1	5.5	4.9	4.9	6.1	4.5	4.2	3.9	3.6	3.4	43.1
Benefits per partnership (2016 prices)	2.7	7.6	7.3	7.8	10.4	8.3	8.3	8.3	8.3	8.3	77.0
PV (base year 2017, 2016 prices)	2.4	6.6	6.2	6.3	8.2	6.3	6.1	5.9	5.7	5.5	59.0
Eligible partnerships	223	223	223	223	223	223	223	223	223	223
Success rate	5%	5%	5%	5%	5%	5%	5%	5%	5%	5%
Total benefits (2016 prices)	30.1	84.5	81.5	86.5	115.6	92.0	92.0	92.0	92.0	92.0	858.3
Total benefits (PV, 2016 prices)	27.1	73.7	68.6	70.3	90.9	69.9	67.5	65.2	63.0	60.9	657.2

Source: Home Office, own calculations.

Benefits are estimated to be £858 million, which equates to £657 million (PV) over 10 years. This is equivalent to a reduction of around 20 homicides per year (there are around 600-700 per year currently).

Non-monetised costs and benefits

Some costs and benefits of the policy have not been monetised. These are listed below:

• It has not been possible to monetise the expected costs on some types of organisation expected to be affected by the proposed duty, notably prisons, youth offending teams and housing organisations.
• Evidence generally suggests that partnership working can reduce fear of crime as well as actual numbers of offences.^{[footnote 24]} This benefit has not been not monetised.
• Given that most offenders are non-specific in their offending and commit a wide range of different offences^{[footnote 25]}, it is possible that the duty may have crime reduction benefits outside of violence. These have not been monetised.

Total costs / benefits, NPV, BNPV and EANDCB

The total cost of Option 1 is estimated to be £361 million and the benefit is estimated to be £858 million (both in 2016 prices). These figures equate to a cost of £281 million and a benefit of £657 million in present value terms (PV) over 10 years.

The concept of present value (PV) takes into account that a set sum of money is not as valuable to people at some point in the future as it would be today, reflecting in part, time preference and in part, an assumption that the economy is expected to grow over time. Therefore, the marginal utility of each additional pound diminishes in the future. Future costs and benefits therefore must be discounted in order to be comparable to today’s costs and benefits.

The Net Present Value (NPV) of a policy is the present value of benefits minus the present value of costs and so the overall current monetary value of a policy option. Therefore in this case, the estimated NPV of Option 1 is £377 million over 10 years.

The total monetised costs to business of Option 1 are estimated to be £25 million (PV over ten years), see section H. No benefits to business have been monetised so the estimated Business NPV is -£25 million over 10 years. The Equivalent Annual Net Direct Cost to Business (EANDCB) is estimated to be £2.9 million (PV).

F. Proportionality.

The analysis in this impact assessment contains best estimates for the cost and benefits of the proposed policies. It was not possible to monetise, but every effort has been made to ensure that the analysis presents the best possible estimates of the likely impact of the preferred option, given the time, resource and data available. Therefore, this is a proportionate effort to appraise the proposed policy change.

G. Risks.

Economic impacts of policy risks:

There is a risk that not all partners are motivated to adopt a multi-agency approach. The consultation showed that certain sectors were less supportive of the preferred option than other sectors. If not all sectors engage, this could result in the benefits being lower than estimated and hence a lower NPV. However, the method used to estimate benefits only incorporated partnership working between three sectors: health, police and local authorities. So it is also possible that the benefits will be greater if all partners engage. In addition, the estimated benefits are currently more than twice as large as the costs. Therefore, even a 50 per cent reduction in benefits will result in a positive NPV.

The duty will require the specified authorities to collaborate and plan to prevent and tackle serious violence and it will require local agencies and bodies to determine how best to organise themselves to address local need effectively. There is a risk that this could result in different approaches, where some areas are still performing better than others, resulting in lower than estimated benefits. However, guidance and support will be provided to mitigate this risk and, as stated above, sensitivity analysis shows that benefits would have to reduce by more than 57 per cent to cause the costs to outweigh the benefits.^{[footnote 26]}

There is also a risk that rather than obtaining additional resources to fulfil the serious violence duty, organisations will reduce activity in other areas and outcomes in those areas will decline, or that having a specific duty for serious violence will reduce the prioritisation of other crime types. This would result in costs to the policy that have not been captured here, driven by worse outcomes in the activities reduced. However, as shown above costs would have to rise markedly for the costs to outweigh the benefits.

New legislation around data sharing, particularly the General Data Protection Regulation (GDPR) may have compounded the perception that organisations are prevented from sharing data and information (or make that more costly to achieve). The Information Commissioner’s Office (ICO) have confirmed with similar information sharing, that the GDPR and the Data Protection Act 2018 do not prevent data sharing for law enforcement purposes and provides mechanisms to achieve this (this can include broader safeguarding schemes to stop vulnerable people falling into crime), but it does require organisations to use those mechanisms appropriately. If they do not, this would likely reduce the estimated benefits. The ICO are currently working on an updated data sharing code of practice which should provide greater clarity to the organisations involved.

Analytical risks:

The evidence base for partnership working has few high quality evaluations. Given that randomised control trials generally show smaller effects than quasi-experimental designs, and this impact assessment relies heavily on the latter, it is possible that the benefits could be over-stated. However, arguably this is compensated by assuming a very high failure rate of 95 per cent.

There is a risk that certain partnerships could pursue an investigative rather than preventative approach to serious violence. This would lead to more offenders entering the CJS. This would increase downstream costs, irrespective of the effect on crime.

H. Direct costs and benefits to business calculations

As described in section E, some of the costs for the proposed option fall upon businesses. These are summarised in Table 4.

Table 4: Estimated costs to business of the preferred option (£ million, 2016 prices)

Year	1	2	3	4	5	6	7	8	9	10	Total
Privately owned children’s homes	1.9	1.9	1.9	1.9	1.9	1.9	1.9	1.9	1.9	1.9	19.2
Higher and Further Education	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	12.2
Schools	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.2
Total costs	3.2	3.2	3.2	3.2	3.2	3.2	3.2	3.2	3.2	3.2	31.6

The total cost to business is estimated at £32 million in 2016 prices. This equates to £24.6 million in present value terms over the 10 year appraisal period. The average cost to business per year is estimated at £3.2 million, which equates to £2.9 million (PV).

Some costs to business were not monetised. For example, none of the costs to prisons were monetised. Some of these would fall on business given that, of the 122 prisons in England and Wales, 14 are privately managed.^{[footnote 27]}

No benefits to business have been monetised. This is because benefits arising from reduced offences fall to public sector organisations like police and the health sector as well as to the physical and mental health/wellbeing of victims. However, it is possible that there are non-monetised benefits to business, for example through a reduction in fear of crime leading to improved business conditions in previous hot-spot areas.

I. Wider impacts

A New Burdens Assessment is currently being completed.

Small and Medium-Sized Business Assessment (SaMBA)

A number of different types of business will be subject to the new duty. These include private schools, private colleges, private alternative provision providers and private prisons. Some of these will be small and medium-sized business, for example, many of the alternative provision providers. It is important that these organisations are included within the scope of the proposed duty to ensure that individuals vulnerable to involvement with serious violence are identified and supported, regardless of the size of the educational (or other) establishment they are associated with.

J. Trade Impact.

This policy will not have any impact upon trade.

K. Implementation date, monitoring and evaluation (PIR if necessary), enforcement principles.

Implementation would occur as soon as the legislation has been passed in Parliament and receives Royal Assent, likely to be in 2020.

The policy will be monitored and evaluated and planning is underway to develop the framework for this.

L. Annexes.

Annex 1: Schedule 6 of the Counter Terrorism & Security Act 2015.

Annex 2: Further evidence on the effectiveness of the partnership approach in Scotland.16

Annex 1: Schedule 6 of the Counter Terrorism & Security Act 2015

Schedule 6 Specified authorities

Local government

• A county council or district council in England.
• The Greater London Authority.
• A London borough council.
• The Common Council of the City of London in its capacity as a local authority.
• The Council of the Isles of Scilly.
• A county council or county borough council in Wales.
• A person carrying out a function of an authority mentioned in section 1(2) of the Local Government Act 1999 by virtue of a direction made under section 15 of that Act.

Criminal justice

• The governor of a prison in England and Wales (or, in the case of a contracted out prison, its director).
• The governor of a young offender institution or secure training centre (or, in the case of a contracted out young offender institution or secure training centre, its director).
• The principal of a secure college.
• A provider of probation services within the meaning given by section 3(6) of the Offender Management Act 2007.

Education, child care etc

• A person who is authorised by virtue of an order made under section 70 of the Deregulation and Contracting Out Act 1994 to exercise a function specified in Schedule 36A to the Education Act 1996.
• A person with whom arrangements have been made for the provision of education under section 19 of the Education Act 1996 or section 100 of the Education and Inspections Act 2006 (cases of illness, exclusion etc).
• The proprietor of—

a) a school that has been approved under section 342 of the Education Act 1996,

b) a maintained school within the meaning given by section 20(7) of the School Standards and Framework Act 1998,

c) a maintained nursery school within the meaning given by section 22(9) of that Act,

d) an independent school registered under section 158 of the Education Act 2002,

e) an independent educational institution registered under section 95(1) of the Education and Skills Act 2008, or

f) an alternative provision Academy within the meaning given by section 1C of the Academies Act 2010.

• A person who is specified or nominated in a direction made in relation to the exercise of a local authority’s functions given by the Secretary of State under section 497A of the Education Act 1996 (including that section as applied by section 50 of the Children Act 2004 or section 15 of the Childcare Act 2006).
• A person entered on a register kept by HM Chief Inspector of Education, Children’s Services and Skills under Part 2 of the Care Standards Act 2000.
• The governing body of a qualifying institution within the meaning given by section 11 of the Higher Education Act 2004.
• The provider of education or training—

a) to which Chapter 3 of Part 8 of the Education and Inspections Act 2006 applies, and

b) in respect of which funding is provided by, or under arrangements made by, the Secretary of State or the Chief Executive of Skills Funding.

• A person registered under Chapter 2, 2A, 3 or 3A of Part 3 of the Childcare Act 2006 or under section 20 of the Children and Families (Wales) Measure 2010 (nawm 1).
• A body corporate with which a local authority has entered into arrangements under Part 1 of the Children and Young Persons Act 2008.
• A person who is specified in a direction made in relation to the exercise of a local authority’s functions given by the Welsh Ministers under section 25 of the School Standards and Organisation (Wales) Act 2013 (anaw 1) (including that section as applied by section 50A of the Children Act 2004 or section 29 of the Childcare Act 2006).
• The governing body of an educational establishment maintained by a local authority in Wales.
• The governing body or proprietor of an institution (not otherwise listed) at which more than 250 students, excluding students undertaking distance learning courses, are undertaking—

a) courses in preparation for examinations related to qualifications regulated by the Office of Qualifications and Examinations Regulation or the Welsh Government;

b) courses of a description mentioned in Schedule 6 to the Education Reform Act 1988 (higher education courses).

Health and social care

• An NHS Trust established under section 25 of the National Health Service Act 2006 or under section 18 of the National Health Service (Wales) Act 2006.
• An NHS foundation trust within the meaning given by section 30 of the National Health Service Act 2006.
• A Local Health Board established under section 11 of the National Health Service (Wales) Act 2006.
• A Community Health Council in Wales.
• The Board of Community Health Councils in Wales or Bwrdd Cynghorau Iechyd Cymuned Cymru.

Police

• A chief officer of police for a police area in England and Wales.
• The British Transport Police Force.
• A Port Police Force established under an order made under section 14 of the Harbours Act 1964.
• The Port Police Force established under Part 10 of the Port of London Act 1968.
• A Port Police Force established under section 79 of the Harbours, Docks and Piers Clauses Act 1847.
• The Common Council of the City of London in its capacity as a police authority.
• A police and crime commissioner established under section 1 of the Police Reform and Social Responsibility Act 2011.
• The Mayor’s Office for Policing and Crime established under section 3 of that Act.
• The Civil Nuclear Police Authority.

Annex 2: Scotland’s approach to violence reduction – details, evidence and comparisons with England and Wales

Scotland introduced a multi-agency violence reduction programme in Strathclyde in 2005 and extended it nationally in 2006. The unit’s broad-based approach involved three strands – criminal justice enforcement measures, short and long-term preventative work, and efforts to change attitudes to violence in Scottish society. This paper examines the effectiveness of the Scottish approach in light of increases in serious violence in England and Wales since 2014.

Violent crime trends in Scotland fell after the introduction of the programme:

• Homicide rates, police recorded serious assault numbers, robberies, and knife possession offences all fell by around a half between 2006/07 to 2014/15.
• Common assaults declined by 26% from 2006/07 to 2014/15.
• The Scottish Crime Survey recorded a 46% fall in violent crime against adults from 2008/09 to 2017/18.
• Hospital admissions for assault with a sharp object fell by 28% from 2007/08 to 2017/18.

This cannot automatically be taken as a sign of effectiveness for the programme for three reasons. Firstly, many other nations, including England and Wales, that did not implement a violence reduction programme at the same time, also experienced marked drops in violence in the decade up to 2014. Secondly, there has been relatively little robust evaluation of the Scottish approach. Thirdly, like England and Wales, there is evidence that the downward trend has ended for some (but not all) of the violent offence-types in Scotland:

• Police recorded serious assaults increased 32% from 2014/15 to 2017/18,
• Police recorded robberies increased by 17% from 2015/16 to 2017/18,
• Police recorded knife possession increased by 24% from 2014/15 to 2017/18.
• Hospital admissions for assault with a sharp object have risen by 4% from 2014/15 to 2017/18.

However, while Scotland’s trends share the same overall shape as England and Wales, the magnitude of the violent crime decline in Scotland was larger up to 2014, and the magnitude of the violent crime rise since then has been smaller. Furthermore, some offence types, including homicide, have not risen at all since 2014 in Scotland. In other words, based purely on an analysis of crime trends, there is some evidence of a potential crime reduction effect for the Scottish programme of activity.

These findings are in line with a recent academic report, which took a different approach to studying effectiveness. Rather than comparing to England and Wales, the authors found that the falls in different types of violent crime were particularly concentrated in the types of violent crime and homicides targeted by policies. They concluded that “it is clear that the largest declines in homicide and violence are in line with the policies and intervention strategies put in place during the mid-2000s to tackle gang violence and knife crime” (Skott and McVie, 2019, p. 5).^{[footnote 28]} The authors still urged caution, given the lack of robust evaluation evidence, and it is important to recognise that there were other differences between Scotland and England and Wales that could explain the better trends in Scotland.

For example, while England and Wales had a 22 per cent rise in fixed-term exclusions from school and a 29 per cent rise in permanent school exclusions from 2014/15 to 2016/17, Scotland’s trend in exclusions has continued downwards. It more than halved the number of temporary exclusions between 2006/07 and 2016/17 (44,546 to 18,376) and permanent exclusions fell from 248 to just 5 over the same period. Scotland also had a different trend in police numbers. For example, whereas officer numbers have fallen 15 per cent between 2010 and 2018 in England and Wales, they fell only by less than 1 per cent in Scotland over the same period.

Changes in drug markets or use, by contrast, show more similarity to England and Wales. Crack-cocaine use in particular has increased since 2014 in Scotland.^{[footnote 29]} However, there is some evidence that ‘County Lines’ models for selling drugs are less developed in Scotland than England, and London in particular, possibly calling in to question the transferability of interventions from Scotland to the English context.^{[footnote 30]}

Finally, while Scotland’s ‘public health approach’ to violence has been widely praised, the breadth of the approach and lack of robust evaluation makes it difficult to assess the impact of the individual strands.

1. Berry, G., Briggs, P., Erol, R., & Van Staden, L. (2011). The effectiveness of partnership working in a crime and disorder context. A rapid evidence assessment, 1. Accessed on 19/06/19. ↩

2. Braga, A. A., Weisburd, D., & Turchan, B. (2018). Focused deterrence strategies and crime control: An updated systematic review and meta‐analysis of the empirical evidence. Criminology & Public Policy, 17(1), 205-250. ↩

3. See The public health approach: Violence Prevention Alliance ↩

4. See: LGA review of the future of community safety services: Local Government Association ↩

5. The survey found that 21 per cent of partnerships stated that reducing violence was a priority and a (non-mutually exclusive) 18 per cent stated that ` gangs/county lines/youth violence’ was a priority. County Lines is a method of drug dealing that has been linked to the increases in serious violence in the Serious Violence Strategy. Read the survey results. ↩

6. See: LGA review of the future of community safety services: Local Government Association ↩

7. The survey results for the report, including the 71% figure, can be read in the community safety survey 2016 by the Local Government Association. ↩

8. There is no known data to inform the above `success rate’ assumption so it has been made deliberately conservative. It is hoped the actual number of successful partnerships will be much higher. ↩

9. The share of non-wage costs (%) is taken from Euro stat, Code: lc_lci_lev . ↩

10. Annual Survey of Hours and Earnings, 2018, Table 14 ↩

11. There are 348 local authorities in England and Wales according to ONS lookup table. ↩

12. Florence, C., Shepherd, J., Brennan, I., & Simon, T. R. (2014). An economic evaluation of anonymised information sharing in a partnership between health services, police and local government for preventing violence-related injury. Injury prevention, 20(2), 108-114. ↩

13. The Cardiff model was a data-sharing partnership between A&E departments, police and local authorities. This makes it an appropriate proxy as CCGs now commission A&E services. This is not to say that CCGs must discharge respond to the duty in this way. CCGs also commission other types of services including mental health services, drugs services etc. So they may choose a different partnership model. The Cardiff example is just used as our best estimate for costs of whatever model is ultimately adopted. ↩

14. Ofsted: developments in children’s social care. ↩

15. See ‘Prevent’ Duty impact assessment. ↩

16. National tables of Schools, Pupils and their Characteristics, Table 2a ↩

17. Berry, G., Briggs, P., Erol, R., & Van Staden, L. (2011). The effectiveness of partnership working in a crime and disorder context. A rapid evidence assessment, 1. ↩

18. Braga, A. A., Weisburd, D., & Turchan, B. (2018). Focused deterrence strategies and crime control: An updated systematic review and meta‐analysis of the empirical evidence. Criminology & Public Policy, 17(1), 205-250. ↩

19. Not all partnerships are likely to follow the Cardiff model exactly. Many will likely follow a new and different approach. The Cardiff costs and benefits are just used as an estimation. ↩

20. Florence, C., Shepherd, J., Brennan, I., & Simon, T. (2011). Effectiveness of anonymised information sharing and use in health service, police, and local government partnership for preventing violence related injury: experimental study and time series analysis. British Medical Journal, , 342, d3313. ↩

21. Florence, C., Shepherd, J., Brennan, I., & Simon, T. R. (2014). An economic evaluation of anonymised information sharing in a partnership between health services, police and local government for preventing violence-related injury. Injury prevention, 20(2), 108-114. ↩

22. Note that the costs from the Cardiff study are not used in this Impact Assessment because the Cardiff model only includes police, health and local authority partners. For the proposed duty costs are assumed to fall on a wider set of partnership organisations: all those that are subject to the duty. ↩

23. The population in Cardiff is approximately 360,000 people. This is in comparison to the CSP average of 186,000. It could therefore be argued that the benefits are overestimated. That is a further reason why we have used a conservative 5 per cent success rate. ↩

24. Berry, G., Briggs, P., Erol, R., & Van Staden, L. (2011). The effectiveness of partnership working in a crime and disorder context. A rapid evidence assessment, 1. ↩

25. Brame, R., Mulvey, E. P., Piquero, A. R., & Schubert, C. A. (2014). Assessing the nature and mix of offences among serious adolescent offenders. Criminal Behaviour and Mental Health, 24(4), 254-264. ↩

26. The sensitivity analysis involved calculating the amount that the benefits would have to reduce by to equal the costs, which are estimated to be £280.6 million over 10 years. The benefits are estimated to be £657.2 million. (£280.6m / £657.2m) – 1 = 57.3%. So the benefits would have to decrease by more than 57% to produce a negative NPV. ↩

27. Performance tracker 2018, Prisons. Institute for Government. ↩

28. Skott, S., & McVie, S. (2019). Reduction in homicide and violence in Scotland is largely explained by fewer gangs and less knife crime. ↩

29. For a more detailed analysis, see Annex: other factors. ↩

30. See for example Robinson, Grace, McLean, Robert and Densley, James (2019) Working County Lines: Child Criminal Exploitation and Illicit Drug Dealing in Glasgow and Merseyside. International Journal of Offender Therapy and Comparative Criminology, 63 (5). pp. 694-711. ISSN 0306-624X ↩