Impact assessment

Economic note 1040: threatening, abusive or insulting behaviour towards emergency workers (accessible)

Published 22 October 2026

Economic Note: Number: HO EN 1040

Title of regulatory proposal: Threatening, abusive or insulting behaviour towards emergency workers.

Lead Department/Agency: The Home Office

Expected date of implementation: Autumn 2026

Origin: Domestic

Date: 2 April 2026

Lead Departmental Contact: CrimeandPolicingBillTeam@homeoffice.gov.uk

Departmental Assessment: Green

Rationale for intervention, objectives and intended effects

This measure is being introduced to ensure that emergency workers have full protection from abuse, including behind closed doors. It seeks to achieve this by introducing two new offences for threatening, abusive, or insulting behaviour towards emergency workers, where that behaviour is aggravated by the individual’s race, religion, sexual orientation, disability, transgender identity, or sex, including when it occurs within private dwellings. These offences apply where a person either intends their words or behaviour to be threatening, abusive or insulting, or is aware that their conduct may have that effect.

Policy options (including alternatives to regulation)

Option 0 – ‘Do nothing’

Option 1 - Creation of two new offences for threatening, abusive or insulting behaviour towards emergency workers on the basis of the protected characteristics of race, religion, sexual orientation, transgender identity, disability or sex, including in private dwellings. Option 1 is the government’s preferred option.

Costs and benefit summary

The introduction of this new measure is expected to increase Criminal Justice System (CJS) costs, as threatening or abusive behaviour based on protected characteristics within a private dwelling - when directed at an emergency worker, will now constitute a criminal offence.

There are currently no monetised benefits for this measure and so leads to a negative NPSV. This measure is expected to protect emergency workers which may help their retention and morale, as well as act as a possible deterrent.

Risks

Creating bespoke offences for specific professions risks encouraging further lobbying from other groups for the creation of a bespoke offence for their workers. The creation of further bespoke offences would lead to familiarisation costs across the CJS.

Total Cost (£m PV)	Transition Cost (£m Constant)	Cost to Business (£m PV)
1.86	0.00	0.00
Total Benefit (£m PV)	NPSV (£m PV)	BNPV (£m PV)	EANDCB (£m PV)
0.00	-1.86	0.00	0.00
Price Base Year	PV Base Year	Appraisal period (Years)	Transition period (Years)
2026/27	2026/27	2026/27 to 2035/36	2026/27

Departmental sign-off (SCS): Katharine Wilson Date: 19/02/2026

Chief Economist sign-off: Tim Laken Date: 14/04/2026

Better Regulation Unit sign-off: Emma Kirk Date: 14/04/2026

Evidence Base

A. Policy objectives and intended effects

1. This measure introduces two new offences for targeting threatening, abusive or insulting behaviour directed towards emergency workers^{[footnote 1]} on the basis of the protected characteristics of race, religion, sexual orientation, transgender identity, disability or sex. These offences will specifically apply where the individual:

• Intends their words, behaviour, writing, sign or other visible representation to be threatening, abusive or insulting; or
• Is aware that they may be threatening, abusive or insulting.

2. This will address the fact that aggravated public order offences (under sections 4, 4A and 5 of the Public Order Act 1986) cannot be currently committed when both parties are in a private dwelling or in private dwellings. This has a particular effect on emergency workers who may experience abuse in people’s homes. These new offences will apply anywhere, including in private dwellings, and will protect emergency workers from suffering abuse related to their protected characteristics whilst on duty.

3. When introduced in the House of Commons, this provision within the Crime and Policing Bill covered racially and religiously aggravated abuse only; but, in line with the government’s manifesto commitment^{[footnote 2]} to extend the aggravated offence to other protected characteristics, the clause was amended at Lords Report stage to include transgender identity, sexual orientation, disability and sex.

B. Policy options considered, including alternatives to regulation

Option 0 – ‘Do-nothing’

4. This would mean no changes to the existing legislation on threatening, abusive and insulting behaviour in sections 4, 4A and 5 of the Public Order Act 1986. Emergency workers subject to threatening, abusive and insulting words and behaviour within people’s homes would not have any legal protection.

Option 1  

5. These provisions will create new offences for aggravated threatening, abusive and insulting behaviour towards emergency workers with protected characteristics of race, religion, sexual orientation, transgender identity, disability or sex, including where it takes place in a private dwelling. This is the government’s preferred option.

Non-regulatory options

6. There are no non-regulatory options, the gap in existing legislation can only be filled by legislating via primary legislation.

C.  Past evaluations and rationale for government intervention

7. Current offences under sections 4, 4A and 5 of the Public Order Act 1986 relating to threatening, abusive and insulting words and behaviour do not apply to abuse inside private dwellings.

8. Section 31 of the Crime and Disorder Act 1998 provides for racially or religiously aggravated public order offences. These aggravated offences apply where a person commits an offence under sections 4, 4A or 5 of the Public Order Act 1986 which provide for offences relating to causing fear or provocation of violence, and the causing of harassment, alarm or distress.

9. The racially or religiously aggravated offences under section 31 of the Crime and Disorder Act 1998 attract higher maximum penalties than the base offences in the Public Order Act 1986. The maximum penalty for the aggravated versions of the Public Order Act 1986 section 4 and 4A offences is two years imprisonment, a fine or both (as opposed to six months’ imprisonment or a fine for the base offences) and the maximum penalty for the section 5 offence is a level 4 fine (maximum of £2,500) as opposed to a level 3 fine (maximum of £1,000) for the base offence).

10. Police representatives have highlighted this shortfall in the law and called for changes to the legislation, in particular in relation of officers being subject to racial abuse within people’s homes. The government considers there is a need to protect emergency workers who are subjected to abuse, where that behaviour is aggravated by the individual’s race, religion, sexual orientation, disability, transgender identity, or sex because when performing their professional duty they cannot choose to leave a person’s home. The new offences apply wherever the abuse takes place, to ensure full protection for emergency workers from this type of behaviour.

D. Appraisal

General assumptions and data

11. The following assumptions apply to the measure within this note:

• The appraisal period for measuring impacts of the proposed legislation is 10 years, 2026/27 to 2035/36.
• Annual costs and breakeven analysis are in 2026/27 prices using the HM Treasury GDP deflators (November 2025)^{[footnote 3]}.
• The present value base year is 2026.
• A 3.5 per cent annual social discount rate is implemented in line with the HM Treasury Green Book (2026)^{[footnote 4]}.
• All costs and benefits are relative to the baseline presented in Option 0: ‘Do-nothing’.
• Estimates for how volumes of proceedings will change as a result of these measures are based on the CJS statistics quarterly (June 2025)^{[footnote 5]}  including the Outcomes by Offences tool^{[footnote 6]} and Magistrates Court data tool^{[footnote 7]}.
• The annual cost for the year 2026/27 is adjusted to account for the fact that the implementation date is in Autumn 2026. It is assumed that the date of implementation for this measure is October 2026, which is six months into the year 2026/27 and so the monetised cost for this year is 50 per cent of the total annual cost.

Costs

12. As outlined above, this measure will provide additional protection for emergency workers and address the fact that aggravated public order offences (under sections 4, 4A and 5 of the Public Orders Act 1986) cannot be currently committed when both parties are in a private dwelling.  Only these offences in private dwellings will incur additional costs to the police and to the CJS, in arresting, prosecuting and sentencing the offenders. These costs are estimated in the monetised costs section of the note.

13. To estimate the number of additional cases that are expected for the threatening, abusive or insulting behaviour towards emergency workers offences, the breakdown of the proportions of the protected characteristics (race, religion, sexual orientation, transgender identity, disability and sex) that make up hate crime was first considered. The estimates for all the proportions of hate crime, except for sex, come from a five-year average of police recorded hate crime data^{[footnote 8]}. As gender-based hate crimes are not currently in police recorded crime, a proxy for the proportion of hate crimes that are gender-based was created by using an average proportion across the US, Canada, Germany, Ireland and Finland^{[footnote 9]}. These particular countries were chosen as they are economically comparable to the UK and have similar proportions of hate crime across the protected characteristics of race, religion, sexual orientation, transgender identity and disability. The five countries above were chosen as they are high income countries like the UK (note the hate crime data available is for England and Wales^{[footnote 10]} and have small combined differences between the proportions for each hate crime strand.

14. Although it should be noted that the proxy is not representative of the true proportion of gender-based hate crime and relies on data collected in other countries, which have different recording standards. For instance, in UK as well as the US, Canada and Germany there is a legal definition for hate crime whereas in Ireland, there isn’t hate crime legislation so the approach to recording hate crime was to include offences which had one or more bias motivation^{[footnote 11]}. Also, in the Criminal Code of Finland, there isn’t a hate crime definition but for reporting purposes, hate crime is classed as offences that are motivated by prejudice^{[footnote 12]}. As the true proportion of gender-based hate crime in England and Wales is not known, there is the possibility that the proxy estimate is an underestimation or overestimation. In this analysis, the proportion of gender-based hate crime in England and Wales is assumed to be 1.4 per cent and a sensitivity range is used to account for uncertainty with the low estimate being 0.5 per cent, which is the proportion of gender based hate-crime in Finland, and the high estimate is 4 per cent, which is the average of the five countries mentioned above but also including Spain, Belgium and Austria.

15. The proportions for each hate crime strand are then rescaled to include the proxy estimate for sex. In the central scenario, the final proportions for each hate crimes strand in England and Wales are Race: 66 per cent, Religion: 5 per cent, Sexual orientation: 16 per cent, Disability:  9 per cent, Transgender identity: 3 per cent and Sex: 1 per cent.

16. To understand the total number of prosecutions for (hate-motivated) aggravated public order offences, the number of prosecutions for racially or religiously aggravated public order offences (from Outcomes by offences tool) is uplifted to capture all hate crime strands not just race and religion by using the proportions estimated above. The central estimate for racially or religiously aggravated public order prosecutions is based on the average number recorded over the past three years (2023 to 2025), which is 4,618 cases. The low scenario uses the three‑year minimum (4,176), while the high scenario uses the three‑year maximum (5,215). After applying the uplift to these estimates, the total number of hate‑motivated aggravated public order prosecutions is estimated at 6,546 in the central scenario, with a range of 5,864 to 7,587 across the low and high scenarios.

17. To estimate the number of prosecutions for the new offences, it is necessary to determine the proportion of hate‑motivated public order prosecutions that involve emergency workers. Using the definition of an emergency worker set out in the Assaults on Emergency Workers Act 2018^{[footnote 13]}, and combining the number of professionally registered clinical NHS staff, police officers, firefighters, and prison officers, the total UK emergency‑worker population is estimated at around 956,000^{[footnote 14]}. This represents approximately 1.4 per cent of the UK population, assuming the population is 69.5 million, in accordance with ONS estimates^{[footnote 15]}.

18. Using the 1.4 per cent estimate, the central projection for prosecutions relating to threatening, abusive, or insulting behaviour towards emergency workers is 90 cases (low estimate: 81; high estimate: 104). This assumes that emergency workers face hate‑motivated offences at the same rate as the general population and therefore may be an underestimate.

19. Because the additional costs arise only when the new offences occur in private dwellings, an estimate of the proportion of such cases is required. There are no official statistics on the proportion of incidents against emergency workers that take place in private homes. A range of proxies are used to account for the inherent uncertainty. One possible proxy is the proportion of Independent Police Complaints Commission (IPCC) investigations into police use of force that occurred in a private home^{[footnote 16]}, which is 22 per cent and is used as the low estimate. Another proxy is from a BMJ Open^{[footnote 17]} study on suspected seizures^{[footnote 18]} that states 60 per cent of ambulance callouts for seizures were in private residences. The 60 per cent figure is used as the high estimate while the midpoint of both figures, 41 per cent, is used as the central estimate since there isn’t more certainty with either the low or high estimate so the average of both ends of the range is taken. These figures in the low and high estimates are used as they give some insight into the proportion of incidents where police officers and paramedics (both roles are classed as emergency workers) have entered private dwellings while at work. In absence of exact figures, we can take these figures as the low and high estimate for the proportion of hate-motivated public order offences against emergency workers that occur in private dwellings.

20. As the figures only come from data on police officers and paramedics, it is likely that they are not representative of the true proportion of incidents against emergency workers that take place in private homes. There are further limitations to these proxies as the estimate from the IPCC only looks at incidents where officers used force rather than all police callouts. Similarly, the estimate from the BMJ Open study only looks at callouts for suspected seizures rather than all callouts that ambulance services attend and so it may be that the true proportion of callouts in homes is lower as seizure incidents may be more likely to require ambulance attendance.

21. Because the estimated hate-related prosecutions for threatening, abusive, or insulting behaviour exclude offences committed in private dwellings, using the percentages above can help estimate the number of prosecutions occurring in private dwellings. Using this approach, the low scenario suggests around 23 prosecutions, the central scenario around 63, and the high scenario around 157.

22. Given the uncertainty around this method, the estimates were cross-checked using data from the Crime Survey for England and Wales (CSEW) year ending March 2025. It was found that there was a very low prevalence of hate-motivated threat incidents in and around the workplace (including the car park and other adjacent areas), when focusing on respondents who fell under the occupational categories corresponding to emergency workers^{[footnote 19]}. The low prevalence seen in the CSEW data suggests that the estimates for the number of prosecutions in private dwellings of between 23 and 157 could be an overestimate, which would mean lower costs to the CJS. However, while the CSEW provides supporting evidence that such hate-motivated threats have a very low prevalence, it cannot produce statistically reliable precise estimates due to the very small number of relevant incidents captured within the sample.

Monetised Costs

Ongoing Costs

Ongoing Costs 1: Court Costs

23. An estimated 9 per cent of these cases are expected to be heard in the Crown Court, with the remaining 91 per cent proceeding in the Magistrates’ Court. This breakdown is based on data from the Magistrates’ Court data tool and reflects the average distribution of racially or religiously aggravated public order offences across court types over the past three years. Because outcomes data are not available for aggravated public order offences covering all protected characteristics, the distribution for racially or religiously aggravated offences is used as a proxy throughout the ongoing‑cost estimates.

24. Applying this court breakdown means in a central scenario, 57 cases in the Magistrates’ Court, with a range of 21 to 142 cases. The estimated number of Crown Court cases ranges from 2 to 14, with a central estimate of 6.

25. From assumptions provided by the MoJ (internal correspondence, 2025), the unit cost of a Crown Court sitting is £3,563 while for the Magistrates’ Court, it is £389, in 2026/27 prices. The estimated court cost is £0.02 million in the low scenario; £0.04 million in the central scenario and £0.11 million in the high scenario.

Ongoing Costs 2: Legal Aid Costs

26. Based on MoJ guidance (internal correspondence, 2025), it is assumed that the proportion of legal aid take-up is 60 per cent for the police; 55 per cent for the Magistrates’ court and 100 per cent for the Crown court. These figures are held constant across the low to high scenarios.

27. According to assumptions provided by the MoJ, the unit cost of legal aid is £394 for the police station, £650 for the Magistrates’ Court and £2,412 for the Crown court, in 2026/27 prices. Using these figures, the legal aid costs are estimated to be £0.02 million in the low scenario, £0.05 million in the central scenario and £0.12 million in the high scenario.

Ongoing Costs 3: Prison Place Costs

28. It is assumed that offenders will serve 40 per cent of their sentence in 2026/27 (Year 1 in the 10-year appraisal period) and 33 per cent in Year 2 to 10.

29. Prison release rates may change, and it should be considered that the offenders will likely be released in several tranches in Year 1, whereby the release rate approaches 33 per cent over the year rather than remain at 40 per cent for the whole year. In this case, the prison place costs are likely to be lower although the difference with the current costs would be minimal.

30. In this analysis, the Immediate custody rate estimate comes from the Outcomes by Offence tool. The estimate across low to high scenarios is 15 per cent, which is the average immediate custody rate for racially or religiously aggravated public order offences in the last three years. Thus, the number of custodial sentences in the low scenario is 3; in the central scenario it is 9 and the high scenario it is 23.

31. The Outcomes by Offences tool shows that the average custodial sentence length (ACSL) for racially or religiously aggravated public order offences is 2.9 months. In Year 1, the central estimate for the number of prison places required for these offences is 0.87 (Low: 0.32 and High: 2.18) and in Year 2 to 10, the central estimate is 0.72 (0.26 to 1.80). According to MoJ estimates, the unit cost of a prison place is £56,700 in 2026/27 prices, and so, in Year 1, the annual prison place cost in the central scenario is £0.05 million (£0.01 million to £0.12 million). In Year 2 to 10, the central estimate for annual prison place cost is £0.04 million (£0.02 million to £0.10 million).

Ongoing Costs 4: Probation Costs

32. Since, in Year 1, it is assumed offenders only serve 40 per cent of their custodial sentence in prison, it is assumed they would serve the remainder of the sentence on licence (60%). In Year 2 to 10, it is assumed that 67 per cent of the custodial sentence is served on licence.

33. Based on MoJ guidance, the marginal cost of a person on licence probation without Electronic Monitoring (EM) is assumed at £4,305, in 2026/27 prices. In Year 1, the central estimate for the licence caseload would be 1.31 (0.48 to 3.27) and in Year 2 to 10, the central estimate is 1.46 (0.53 to 3.65). The licence caseload calculated by multiplying the number of custodial sentences by the proportion of the custodial sentence is served on licence.

34. To estimate the licence probation cost, the licence caseload is multiplied by the unit cost of licence probation In Year 1, The estimated licence probation cost is £2,000 in the low scenario; £6,000 in the central scenario and £14,000 in the high scenario. For clarity, here is an example of how the central estimate for licence probation cost was calculated in Year 1: Licence caseload (1.31) x Unit cost of licence probation (£4,305) = Total licence probation cost (£6,000, to the nearest £1000). In Year 2-10, the estimated licence probation cost is £2,000 in the low scenario; £6,000 in the central scenario and £15,000 in the high scenario.

35. Using the Outcomes by offences tool, the estimated proportion of cases leading to a community sentence is 25 per cent. This gives the central number of community sentences 16 (6 to 39) and the average length of a community sentence is 13.6 months. According to assumptions provided by MoJ, the cost of community probation without EM is £3,396, which is used for both community sentence and suspended sentence costs. The central estimate for community sentence costs is £0.06 million (£0.02 million to £0.15 million).

36. The estimate proportion of cases leading to a suspended sentence is 10 per cent. This means the central number of suspended sentences is 6 (2 to 15) and the average suspended sentence length is 18.1 months. The central estimate for suspended sentence costs is then £0.03 million (£0.01 million to £0.08 million). The total probation cost including both community and licence probation costs is £0.10 million (£0.04 million to £0.24 million) for Year 1 to 10.

Table 1: Yearly Ongoing Costs by Type, 2026/27 prices (£) (Year 1)

	Court Costs	Legal Aid Costs	Probation Costs	Prison Place Running Costs	Total Running Costs (yearly)	Total Running Costs (yearly) - implementation date
Low	15,000	18,000	35,000	18,000	86,000	43,000
Central	42,000	48,000	97,000	49,000	237,000	119,000
High	105,000	121,000	243,000	124,000	593,000	297,000

Source: Home Office, own estimates, 2026; numbers may not sum due to rounding. Rounded to the nearest £1,000

Table 2: Yearly Ongoing Costs by Type, 2026/27 prices (£) (Year 2 to 10)

	Court Costs	Legal Aid Costs	Probation Costs	Prison Place Running Costs	Total Running Costs (yearly)
Low	15,000	18,000	36,000	15,000	83,000
Central	42,000	48,000	98,000	41,000	229,000
High	105,000	121,000	245,000	102,000	573,000

Source: Home Office, own estimates, 2026; numbers may not sum due to rounding. Rounded to the nearest £1,000

Unmonetised costs

37. One potential cost is that the introduction of these new offences specifically for emergency workers may lead to an increase in the reporting of hate-motivated threats towards emergency workers. This would increase the number of offences that the police and CJS process and so increase the costs. This is not accounted for in this cost-benefit analysis as it is not possible to predict whether there will be any change in the number of offences reported, particularly as these are new offences.

38. Familiarisation costs have not been included for the police or CJS because there is only a small difference between the two new offences and the aggravated public order offences that now cover the listed protected characteristics.

Benefits

Monetised benefits:

39. There have not been any monetised benefits identified in this appraisal, with unmonetised benefits discussed below. The potential benefits of a deterrent effect and improved retention rates for emergency workers cannot be monetised as the new offences are currently not separately recorded and so there is insufficient data for any forecasting of decreases in the number of offences or increases in the retention rates.

Unmonetised benefits

40. One potential benefit of the new offences is a deterrent effect: clearer legal consequences may discourage individuals from engaging in aggravated abuse against emergency workers on the basis of protected characteristics, including in private dwellings where such behaviour would not result in an arrest.

41. The recording of these offences in a bespoke offence will allow for a better understanding of the scale of the issue and track whether these offences against emergency workers are decreasing over time.

42. The policy may also support improved retention rates for emergency workers as this measure may reduce the amount of abuse they face. In a Metropolitan Police Exit Survey over two years (2021 to 2023)^{[footnote 20]} , 4.6 per cent of staff put ‘Discrimination, harassment and bullying’ as the main reason for leaving. This displays that preventing harassment plays a role in retaining staff.

Breakeven analysis:

43. Using the violence without injury value from the Economic and Social Costs of Crime (ESCC)^{[footnote 21]} as a proxy for threatening or abusive behaviour towards emergency workers and adjusting to exclude defensive expenditure and insurance administration costs, the economic and social cost of threatening or abusive behaviour is estimated to be £8,185 (2026/27). The ESCC unit cost was uprated using GDP deflators^{[footnote 22]} .The breakeven analysis indicates preventing between 10 and 70 offences would offset the annual steady costs. The breakeven values were calculated by dividing the total ongoing costs (Year 2-10) by the economic and social cost of the offence. The central estimate in the breakeven analysis of 28 offences was calculated as follows: Total ongoing costs (£229,158)/ Economic and social cost of threatening and abusive behaviour (£8,185) = 28. The total ongoing costs for Year 1 are not applicable for entire year of 2026/27 due to the implementation date.

NPSV, BNPV and EANDCB

Table 3: Summary CBA, NPSV, BNPVC and EANDCB, £ million (PV) over 10 years, 2026/27 prices:

Summary	Low	Central	High
Total Transition Costs	0	0	0
Total Ongoing Costs (Year 1) - accounting for implementation date	0.04	0.12	0.30
Total Ongoing Costs (Year 2-10)	0.08	0.23	0.57
Total Costs	0.68	1.86	4.66
Ongoing Benefits	0	0	0
Total Benefits	0	0	0
NPSV	- 0.68	-1.86	-4.66
BNPV	0	0	0
EANDCB	0	0	0

Source: Home Office, own estimates, 2026, numbers may not sum due to rounding. Rounded to the nearest £1000. Note: CBA: Cost Benefit Analysis, NPSV = Net Present Social Value, BNPV = Business Net Present Value and EANDCB = Equivalent Annual Net Direct Cost to Business.

E. Wider impacts

44. There is a positive effect for emergency workers with protected characteristics as people can no longer harass emergency workers for their race, religion, sexual orientation, transgender identity, disability or sex while in a private dwelling, without facing the legal consequences. There is no anticipated environmental or regional impact and there should be no impact on income and wealth distribution; households; smaller businesses and trade and investment.

45. The new offences are strictly for intentional use of threatening or harassing language against emergency workers so this should not disproportionately affect groups such as those with Tourette’s Syndrome or groups more likely to be homebound either due to medical reasons or age.

46. This measure could also affect vulnerable women, in similar way to the bespoke Assault on Emergency Workers (AEW) offence. The AEW offence was introduced in 2018 and applies to an offence of common assault or battery committed against an emergency worker, which differs from the threatening, abusive and insulting behaviour towards emergency workers offences in this measure as these are aggravated public order offences rather than assault offences. The AEW offence has a higher maximum penalty of two years compared to a maximum penalty of six months for the standard common assault and battery offence. The AEW offence accounted for a large proportion of female prosecutions (52% in 2023)^{[footnote 23]} for violence against the person offences, and many of those prosecuted were women who were themselves victims of domestic abuse or experiencing mental health difficulties^{[footnote 24]} . It is possible that the two new offences may also lead to a disproportionate number of prosecutions of women. These new offences do not carry higher maximum sentences than their equivalent aggravated public order offences so will not lead to more women receiving harsher sentences, which was one of the issues with the AEW offence. Moreover, these new offences are being implemented to ensure that emergency workers are covered within private homes and due the expected number of cases in private dwellings of between 23 and 157, any impact will involve only a small number of new cases.

F. Sensitivity

47. Under the Better Regulation Framework, this measure is not a regulatory provision, and the preferred policy option does not have an Equivalent Annual Net Direct Cost to the Public Sector exceeding +/- £20 million, so a full IA is not necessary. This Economic Note has been produced instead and contains proportionate analysis outlining the key costs, benefits, and NPSV calculations.

48. Sensitivity analysis was conducted on assumptions that were held constant in the analysis, including Proportion of population that are emergency workers, Custody rate, ACSL (in years) and Legal Aid take-up. The changes to the central NPSV, in the analysis, due to changes in these assumptions are examined. The current central NPSV is -£1.86 million.

49. The assumption for custody rate is 15%. To test the sensitivity of this assumption, a low and high rate of 12% and 17% is used. The NPSV is -£1.78 million and -£1.94 million in the low and high scenarios, respectively. This indicates that the NPSV is not highly sensitive to reasonable variation around the assumed custody rate.

50. The current assumption for the ACSL for these new offences is 0.24 years. A low and high scenario is tested at 0.19 years and 0.29 years. In the low scenario, the NPSV is -£1.78 million and in the high scenario, the NPSV is -£1.94 million. The NPSV is also not overly sensitive to a change in the ACSL.

51. The current assumptions for Legal aid take-up are 60% for Police; 55% for Magistrates’ court and 100% for Crown court. Low and high scenarios were modelled to account for an uncertainty of ±20%, respectively. In the low scenario, the NPSV is -£1.78 million and in the high scenario, the NPSV is -£1.92 million. This suggests that changing the legal aid take-up proportions would have minimal impact on the NPSV.

52. The current assumption for the proportion of the population of England and Wales (UK population is used as a proxy) who are emergency workers is 1.4%. For sensitivity this is modelled with low and high scenarios of 1.1% and 1.7%. The NPSV is -£1.49 million in the low scenario compared to -£2.23 million in the high scenario. The NPSV is more sensitive to changes to the proportion of the population who are emergency workers than the other assumptions.

G. Risks

53. One significant risk of this analysis and several of the assumptions underpinning it, is that it is largely based on the current system and historic patterns. The analysis may not fully account for the sentencing review, particularly assumptions linked to sentence length, custody rates, or suspended sentences for shorter sentences. Where possible assumptions have been adjusted; however, there may be changes arising from the implementation of the sentencing review which are not fully reflected. For example, the Sentencing Act 2026 may increase the chances of suspended sentence rather than a custodial sentence which would lower prison related costs.

54. Important to note, that there is a risk that other professions may request similar bespoke offences for the protection of their staff which could create a more complicated crime recording system. These measures could produce further pressure to introduce more new offences which can lead to further familiarisation costs. However, the government considers emergency workers in need of additional protection due to the nature of their profession.

55. Another potential limitation of the analysis is that emergency workers may be exposed to a disproportionately high level of hate-related threatening or abusive behaviour because of the nature of their roles. If this is the case, the assumption that the proportion of aggravated public order offences against emergency workers is equal to their share of the population may lead to an underestimation of CJS costs.

H. Annex

Mandatory specific impact test - Statutory Equalities Duties

The Offences for the Abuse of Emergency Worker makes it a criminal offence to intentionally direct threatening, abusive, or insulting behaviour on the basis of race and religion, including in a private dwelling. This measure will be extended to other protected characteristics including sexual orientation, transgender identity, disability and sex. This measure may positively discriminate against those who face abuse due to these characteristics.

It is also expected that this measure will encourage good relations and increase mutual respect between emergency workers and the public as inclusive enforcement practices are being promoted in this measure and protecting emergency workers in the course of their duties based on protected characteristics.

The SRO has agreed these findings

Complete: Yes

1. As defined under the Assaults on Emergency Workers Act 2018: https://www.legislation.gov.uk/ukpga/2018/23 ↩

2. Government’s manifesto commitment: https://questions-statements.parliament.uk/written-questions/detail/2024-10-08/HL1414 ↩

3.    HM Treasury GDP deflators (2025): https://www.gov.uk/government/statistics/gdp-deflators-at-market-prices-and-money-gdp-november-2025-budget-2025 ↩

4. https://assets.publishing.service.gov.uk/media/698dbcd17da91680ad7f4308/The_Green_Book_2026.pdf ↩

5.   https://www.gov.uk/government/statistics/criminal-justice-system-statistics-quarterly-june-2025 ↩

6. https://assets.publishing.service.gov.uk/media/68f93024ec6267c615ed8fe9/outcomes-by-offence-data-tool-June-2025.xlsx ↩

7. https://assets.publishing.service.gov.uk/media/68f92f9d0794bb80118bb7b9/magistrates-court-data-tool-june-2025.xlsx ↩

8. https://www.gov.uk/government/statistics/hate-crime-england-and-wales-year-ending-march-2025 ↩

9. Gender-based hate crime data: https://hatecrime.osce.org/gender-based-hate-crime?year=2024 ↩

10. https://www.gov.uk/government/statistics/hate-crime-england-and-wales-year-ending-march-2025 ↩

11. https://hatecrime.osce.org/national-frameworks-ireland#dataCollection ↩

12. https://hatecrime.osce.org/national-frameworks-finland#dataCollection ↩

13.  As defined in Assaults on Emergency Workers (Offences) Act 2018:  https://www.legislation.gov.uk/ukpga/2018/23 ↩

14.   [1] https://digital.nhs.uk/data-and-information/publications/statistical/nhs-workforce-statistics/november-2025 ↩

15. https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/provisionalpopulationestimatefortheuk/mid2025 ↩

16. https://www.policeconduct.gov.uk/sites/default/files/documents/IPCC_Use_Of_Force_Report.pdf ↩

17. BMJ Open is an online, open access journal, dedicated to publishing medical research from all disciplines and therapeutic areas. ↩

18. https://bmjopen.bmj.com/content/6/2/e010573 ↩

19. https://www.legislation.gov.uk/ukpga/2018/23 ↩

20. https://www.met.police.uk/foi-ai/metropolitan-police/disclosure-2023/december-2023/rates-of-police-leavers/#:~:text=The%20most%20selected%20reason%20for,a%20career%20change%27%20for%2063 ↩

21. https://assets.publishing.service.gov.uk/media/5b684f22e5274a14f45342c9/the-economic-and-social-costs-of-crime-horr99.pdf ↩

22.   GDP deflators: https://www.gov.uk/government/statistics/gdp-deflators-at-market-prices-and-money-gdp-november-2025-budget-2025 ↩

23. https://assets.publishing.service.gov.uk/media/682d8d995ba51be7c0f45371/independent-sentencing-review-report-part_2.pdf ↩

24.  Independent Sentencing Review: https://assets.publishing.service.gov.uk/media/682d8d995ba51be7c0f45371/independent-sentencing-review-report-part_2.pdf ↩