This month, the government announced that Baroness Casey will be undertaking a three-month audit of “gang-based exploitation”, which will include an examination of the ethnicity of victims and offenders.
The announcement referred to data gaps identified by the Independent Inquiry into Child Sexual Abuse led by Alexis Jay in 2022. Since then, the National Police Chiefs’ Council (NPCC) has, through the Child Sex Exploitation (CSE) Taskforce, begun to address those gaps. Last year the CSE Taskforce released its analysis of data from 2023. The report showed that, where ethnicity data was available, the majority of CSE suspects (83 per cent) and victims (85 per cent) were White.
In response to recent political and media attention, the NPCC released data from 2024 that showed only 3.9 per cent of group-based abusers were Pakistani and 85 per cent were White. The same data also appear to show that this small proportion of Pakistani suspects was more concentrated in the narrowly defined cases not involving institutions or family settings. That data is only drawn from three quarters of 2024 and only 30 per cent of the records included ethnicity information.
This piece is not intended to be yet another analysis of the limited data available on one specific kind of offence. The negative effect such a focus has on victims is highlighted by Assistant Chief Constable Becky Riggs in that same press release. This piece rather aims to address some wider discussions that have surfaced about why ethnicity data is incomplete and what the data can really tell us.
Why ethnicity is not recorded
There have been suggestions that ethnicity is not recorded for fear of stoking community tensions and concerns about being perceived as racist. This conclusion is partly drawn from the 2014 Rotherham inquiry, which connected frontline staff confusion to perceptions that senior police and councillors were downplaying ethnicity. However, there may be more practical barriers to collecting ethnicity information in general, both by the police and by the courts.
Aside from discomfort around asking about ethnicity, research from one of the police forces in the Jay Inquiry revealed that practical reasons for not recording ethnicity included: little understanding about why ethnicity data matters; lack of clarity on who should be recording it; and that some of the reporting systems did not make it easy to record. In its inspection of race disparity in police criminal justice decision-making in 2023, HM Inspectorate of Constabulary and Fire & Rescue Services outlined similar reasons that police may not record ethnicity for offences more generally.
Meanwhile, the recording of ethnicity data in court has partly been limited by the level of interaction with a defendant (in summary offences the defendant may not have even appeared in court).
What ethnicity data tells us and where the limits are
Ethnicity data, if complete, can show whether a particular demographic is disproportionately represented among victims or offenders, or both. This provides a starting point for further investigation and understanding.
In investigating disproportionality in the criminal justice system, properly recorded ethnicity data can be (and has been) used to evidence ethnic disparity in charging decisions and sentencing outcomes. As the 2017 Lammy Review noted, disparity in the criminal justice system is caused by inequalities both within and beyond the system, but it is the responsibility of institutions within the system to find the explanations that lie behind the data.
If local agencies notice that offences are affecting and involving a particular community, not necessarily disproportionately, this can only be addressed if police and local services work together with community groups and charities. The 2014 Rotherham inquiry found that agencies had not properly worked with the local community, focusing only on a handful of religious and council leaders. Women’s groups argued that this also disenfranchised women and girls, which contributed to under-reporting of ethnic minority victims.
A warning label
The most important limitation on what ethnicity data can tell us is that ‘ethnicity’ itself is only ever a shorthand. The categories are necessarily loosely defined because they are subject to change and the boundaries between them are not clear. In the 2011 census, for example, Chinese ethnicity was recategorized from Other to Asian, and two more ethnic groups were added to White and Other categories (Gypsy or Irish Traveller and Arab respectively).
Ethnicity is not innate, it tells us nothing about how a person is wired to behave. It is but one tool we have to summarise one factor that influences how we see each other and how we see ourselves.
Whatever further investigation ethnicity data can lead to, it is crucial that the data are as complete as possible and explained. When data are incomplete and unexplained, it leaves a vacuum to be filled by claims that risk being, at best, misleading and, at worst, incendiary.
