What Police Scotland data on hate crime reveals about non-crime hate incidents
Since the Hate Crime and Public Order (Scotland) 2021 (HCPOA) came into force on 1 April, Police Scotland has published weekly data relating to the new law. This includes statistics on the number of reports made to the police, the number of recorded crimes, non-crime hate incident (NHCIs) and other related data.
NHCIs are cases where no evidence criminality is established. The new data suggests a much lower level of these than previously reported by Police Scotland to the Scottish Police Authority (SPA). It brings out that, contrary to the published policy, many cases reported to the police must be being screened out before recording.
We drew the disparity between the new figures and earlier reports to the SPA to the attention of a Telegraph journalist, who sought clarification from the Police Scotland press office. The explanation given by Police Scotland suggests two things. Firstly, that the data on hate crime presented to the SPA fails to meet basic standards of accuracy and clarity, not least those that should be expected of a major public body. Secondly, that the number of NCHIs recorded has trebled in the four years to 2022/23. This finding does not appear to have been flagged at any point by Police Scotland, and no explanation has been published of what sort of cases account for this rise.
This blog unpicks the ongoing confusion in this area, which principally relates to the varying and inconsistent terms and definitions used by Police Scotland in its policies, performance reports, and on its website. We argue that the discrepancy between the new figures and those previously issued by Police Scotland raises questions not only about how information has been handled in this specific area, but wider questions about the quality control of information supplied to the SPA.
Weekly HCPOA reports: non-crime hate incidents.
The new weekly HCPOA reports present daily, weekly, and cumulative data on NCHIs, compared to the same period in previous years. The same numbers are also reproduced as text, beneath the relevant table. Police Scotland has not previously published data in this specific form in the decade since it was created.
Within the first two weeks of the new Act, Police Scotland recorded 55 NCHIs. It is too soon to see reliable patterns and the small numbers will make year-on-year change more erratic. It is nonetheless striking that the year-on-year increase now being reported is much smaller than comparable increases between 2020/21 and 2021/22, and between 2022/23 and 2023/24.
No data is provided on what characteristics are relevant to each incident. Amid ongoing concerns that people may have their details recorded as alleged perpetrators of an NCHI as a result of differences of opinion over questions of sex and gender identity, providing data by characteristic could provide reassurance that NCHIs are not being disproportionately used to police freedom of expression on this specific topic.
Neither is any contextual information provided about NCHIs, which would make it easier to understand what the numbers are showing, such as why some incidents are badged as NCHIs, while others do not meet the recording threshold.
It is clear from the weekly reports to date that a very small proportion of complaints made to Police Scotland (for example, online reports, calls and emails) since the Act commenced have resulted in a recorded crime, and even fewer in an NCHI.
As a key metric of how the demand placed on Police Scotland has changed, it is surprising that data on the total volume of reports is not presented in the weekly reports in a way that makes it straightforward to see what percentage led to any sort of official record, rather than leaving the calculations to stakeholders and journalists.
Police Scotland performance reports
Separately, since 2020/21, Police Scotland has presented data on hate crimes and hate incidents in its quarterly performance reports to the SPA. The data is presented in a single table and includes data for the previous year plus five-year average data.
As revealed in the Police Scotland response to the Telegraph, it transpires that the data on hate crime presented to the SPA is a very poor representation of the actual position.
This becomes clear when this data is properly examined. The extract below is from Police Scotland’s ‘Q4’ 2022/23 report to the SPA. The ‘YTD’ figure is for ‘year to date’: for the Q4 report this means the figures give the full year’s data for each year shown.
The report continues:
Hate incidents have decreased compared to the same period last year but remain above the five year mean. Hate crimes have decreased significantly compared to last year and are also below the five year mean.
The same figures for the previous two years can be seen in the Police Scotland 2020/21 Q4 report.
Incidents and crimes are not defined in the performance report. They are however, set out in Police Scotland National Guidance on hate crime, where they are defined as two distinct categories.
In brief, hate crimes require evidence of a criminal offence. Hate incidents (also referred to by Police Scotland as non-crime incidents and non-crime hate incidents) do not require evidence and are based on person’s perception of prejudice.
Police Scotland will record all hate crimes and hate incidents in terms of the following definitions:
Hate Incident – Any incident which is perceived by the victim or any other person, to be motivated (wholly or partly) by malice and ill-will towards a social group but which does not constitute a criminal offence (non-crime incident)…
Hate Crime – A hate crime is any crime which is perceived by the victim or any other person, to be motivated (wholly or partly) by malice and ill-will towards a social group.
The recently updated Police Scotland website further underscores the difference between the two. It also makes clear that the definition of a ‘hate incident’ in the National Guidance, refers to NCHIs (emphasis added):
A non-crime hate incident is any incident perceived by the victim, or any other person, to be motivated either entirely or partly by malice and ill-will towards a person or group based on the victim’s actual or perceived membership of one or more of the characteristics, but which does not constitute a criminal offence.’
The understanding that hate crimes and hate incidents1 are separate categories is also reflected in a report on hate crime by HM Inspectorate of Constabulary (HMICS) published in 2021. This states (emphasis added):
In 2019-20, Police Scotland recorded 6,448 hate crimes. In addition, over the same period, Police Scotland recorded 6,745 hate incidents.
It is therefore very clear that in ordinary use the terms ‘hate incidents’ and ‘hate crimes’ describe two mutually exclusive categories. The Police Scotland report to the SPA at no point says it is making any different use of these terms.
The numbers don’t add up
The data presented to the SPA suggests a much higher level of NCHIs than recently reported by Police Scotland in its new report. Looking at the five-year average figure for ‘hate incidents’ (6,634) reported to the SPA in 2022/23, this would be around 130 such reports a week.
Police Scotland’s more recent report giving figures for the first two weeks of the year, quoted above, however implies an average in early April around ten times lower, of just 12 NCHIs per week, over the four years to 2022/23. Even looking just at this year, it shows just 55 NCHIs recorded in the first fortnight of this year.
As noted above, we drew this marked dispartity to the attention of a Telegraph journalist, who was able to ask for clarification of the numbers from the Police Scotland press office.
Astonishingly, Police Scotland confirmed that for the purposes of the data presented to the SPA, ‘incidents’ has been used as a catch-all umbrella term for everything; for both incidents determined to be crimes and those not. In other words, contrary to its own National Guidance and website, for the purposes of reporting to the SPA, Police Scotland do not treat hate incidents and hate crimes as mutually exclusive categories. Instead, crimes are regarded as a subset of all incidents.
The tables presented to the SPA are an exceptionally unhelpful and convention-flouting way to present data. Lacking any specific advice to the contrary, and especially given the ordinary use of the terms, any reader would reasonably assume they were being presented with two distinct and separate categories of information.
Police Scotland’s clarification to the Telegraph further explained that NCHIs in fact account for the difference between hate incidents and hate crimes in its performance reports. These figures are not shown but can be easily calculated by subtracting the number of crimes from the number of incidents.
In the table below, the annual NCHI figures have been obtained by deducting the figure for ‘hate crimes’ from that for ‘hate incidents’ given for each of the four most recently available years, as shown in Q4 reports from Police Scotland to the SPA above.
This reveals that whilst the number of hate crimes has remained broadly consistent, the number of NCHIs has trebled since 2019/20.
Police Scotland recorded hate crimes and NCHIs, 2019/20 to 2022/23
| Year | Hate Crimes | NCHIs | TOTAL (‘Hate incidents’) |
| 2019/20 | 6,452 | 311 | 6,763 |
| 2020/21 | 6,701 | 319 | 7,020 |
| 2021/22 | 6,925 | 651 | 7,576 |
| 2022/23 | 6,233 | 978 | 7,211 |
| % change from 2019/20 to 2022/23 | -4% | +214% | +7% |
Sources:
Police Scotland Quarter 4 YTD Performance Report: April 2022 to March 2023
Quarter 4 Performance Report January to March 2020/21
Given the ongoing controversy around NCHIs throughout the UK, including a recent court case in England and Wales (see footnote 2), failing to identify these figures, and the sharp upward trend, is a significant omission in reporting to the SPA. As noted above, the commentary on the figures for 2022/23 stated only that ‘Hate incidents have decreased compared to the same period last year but remain above the five year mean’.
Hate reports, hate incidents, or non-crime hate incidents?
The ongoing confusion over NCHIs in Scotland is principally a result of how Police Scotland records and describes its polices and performance, with key terms either defined or used differently in different places. Taking an overview, it appears that:
- Police Scotland National Guidance defines ‘hate incidents’ and ‘hate crimes’ as separate categories. The same understanding is also shared by HMICS. In contrast, ‘hate incidents’, as reported to the SPA include hate crimes.
- While Police Scotland National Guidance and its recently updated website treat ‘hate incidents’ and NCHIs as interchangeable, in reports to the SPA, NCHIs are in fact a small, undisclosed, subset of ‘hate incidents’.
- The small number of NCHIs means that the threshold for recording in practice cannot conceivably match that described in the National Guidance, or that on the recently updated Police Scotland website, which is perception-based.
- But though the number of NCHIs is much smaller than implied in Police Scotland policies, the figure has more than trebled in the four years to 2022/23. Police Scotland has not flagged this to the SPA, nor shared any information about what sort of incidents, and which characteristics, have driven this rise.
There are several possible explanations as to what is going on here. The first is that Police Scotland does not, or at least not until recently, did not, understand its own data. Like HMICS, and as reflected in its own policies, it may have viewed the figures for hate incidents as separate to hate crimes, and believed the hate incident line in its tables to be providing information on NCHIs. This helps explain why, despite increasing public interest on the issue, it did not report data specifically on NCHIs by name to the SPA; it thought it was already doing so.
If this is the case, against the backdrop of the 2021 Miller v College of Policing case,2 it suggests a surprising lack of curiosity by both Police Scotland and the Scottish Police Authority, as to why the single force was apparently recording more NCHIs than actual recorded hate crimes.
This would also suggest a concerning detachment from on the ground policing. If ‘hate incidents’ were understood to mean NCHIs, it would mean that staff processed and signed off figures that were more than twenty times too high for 2019/20, and still more than six times too high for 2022/23, compared to the front-line reality.
Another related explanation is that Police Scotland did not understand its own policy, as operated in practice. On paper, Police Scotland lifted its perception-based policy from the older College of Policing, since overturned in England and Wales by the ruling in Miller v College of Policing.2 Its public-facing policy continues to tell people that ‘Police officers and staff must ensure a hate concern is raised in relation to every hate crime or hate incident reported to police’ (2021: 14).
Yet in practice, the new weekly data provided explicitly on the number of NCHIs recorded both prior to and since the Act, strongly suggests this is not what has been happening. It certainly is not what happened to most of the thousands of reports made in the first two weeks of April this year.
This was recently confirmed in a letter from DCC Alan Spiers to the Criminal Justice Committee:
There are certain circumstances where an incident that does not meet the criminal threshold provided for by statute, will be recorded by the police as a non-crime hate incident. Such incidents will be recorded only where it is necessary and proportionate to do so, and where keeping a record of the circumstances of the incident meets a particular policing purpose.
Whilst this may provide some reassurance to those worried about over-recording, it remains at odds with published Police Scotland policy and provides no clear indication of what the policy in operation actually is, and most importantly, what it takes to reach the recording threshold.
The bigger picture here relates to the sprawling policy agenda around ‘hate crime’. Just as Police Scotland’s website says that it is offering ‘legal’ definitions of hate crime that in fact muddy the boundary between law and policy, NCHIs have occupied a space where incidents, data, reports, and people are readily subsumed under a ‘hate’ umbrella, absent evidence of criminality. In this ‘anything goes’ landscape, which has seen Police Scotland officers investigating feminist stickers as potential hate crimes, a lack of conceptual clarity and linguistic slippage is perhaps unsurprising.
In short it appears that it has taken the commencement of the HCPOA to shed at least some light on a longer-standing confusion over language, policy and law, and practice, and the relationship between all of these.
Police Scotland needs to improve the clarity of its reporting to the SPA. It also now needs to explain what accounts for a threefold increase in recorded in NCHIs over a four-year period, when the number of recorded hate crimes has remained constant. It needs to provide data on the underpinning characteristics associated with those reports and make clear what the threshold for recording actually is.
Given that a high proportion of hate crime charges are brought with police officers as victims, as the First Minister has recently emphasised, it would also be useful to understand how far these cases relate to victims who contact the police, or to police officers themselves.
While doing all this, Police Scotland also still needs to clearly explain how it records details on alleged suspects and whether, when, or how these could ever be disclosed under an enhanced disclosure, or otherwise be referred to again in any later context.
But there is an issue here that goes beyond the specific topic and raises a wider question about how data presented to the SPA is quality controlled and sense-checked.
Police Scotland needs to understand how it could have provided the SPA with data that was so poorly presented as to be substantially misleading. Did those involved in submitting the data to the SPA understand that ‘hate incident’ was being used with a meaning quite different from its usual one? If so, why did they not make that clear in the explanation and presentation of the figures?
If, instead, those signing off the SPA reports thought the figures matched the official definition, at what point in their handling within the organisation did their meaning become lost? And in that case, how was such a large difference from the reality on the ground not picked up by anyone involved? Lastly, did anyone in Police Scotland actually notice the discrepancy between the SPA reports and the new weekly data, before it was raised with the Press Office.
We suggest this specific case raises some wider issues the SPA may want to pursue.
Notes
1 As an insight into the hitherto paucity of information in this area, the HMICS report notes that its data on hate incidents was based on freedom of information responses.
2 Mr Miller posted several tweets which a member of the public had complained were “transphobic”. Humberside Police recorded this as a NCHI and warned Mr Miller that he may face criminal prosecution if he continued to post similar tweets. In the first case (Harry Miller v The College of Policing and The Chief Constable of Humberside 2020) the judge ruled Humberside Police had disproportionately interfered with the rights of free speech of Mr Miller. A subsequent appeal (Miller v College of Policing 2021) focused on 2014 College of Policing guidance on NCHIs, as followed by Humberside Police, which provided for perception-based reporting. The Miller appeal ruling (both successful and unsuccessful grounds) is summarised here.
Whilst awaiting the outcome of the appeal, in a striking turn of phrase, the Interim College of Police CEO told the National Police Chief Council that there is ‘widespread concern the Hate Crime Operational Guidance has been weaponised’ (O’Reilly, 2021: 46).
The appeal judge ruled that perception-based recording of NCHIs per was not unlawful per se; but additional safeguards were needed for the incursion into freedom of expression to be no more than strictly necessary.
In response, in July 2022 the College of Policing issued interim guidance on recording NCHIs. Departing from the previous perception-based approach, this required officers to make to a judgment as to the gravity & validity of complaints, and whether recording was justified, before taking recording the complaint.