The demographics of grooming gangs

Are Muslims overrepresented?

NOTE: I now write for Aporia Magazine. Please sign up there!

In 2011, Times journalist Andrew North published a bombshell investigation of child sexual abuse in the town of Rotherham, England. Thanks to his reporting, and an independent inquiry by Professor Alexis Jay, we know that between 1997 and 2013 more than 1,400 girls were groomed, sexual abused and (in some cases) violently gang-raped by mostly British Pakistani men.

The incident was notable not only because of the scale of the abuse, but also because of the ethnic dimension (most of the victims were white or non-Muslim Asian), and because of evidence that authorities had failed to act for fear of appearing “racist”. In subsequent years, various other examples came to light, and the groups of perpetrators became known as “grooming gangs”.

In 2017, the now-defunct thinktank Quilliam published a report titled ‘Group Based Child Sexual Exploitation – Dissecting Grooming Gangs’. Based on “extensive data mining”, the authors concluded that 84% of those who’d been convicted for grooming gang offences were of South Asian origin – a remarkable figure, given that South Asians comprised only 7% of England’s population at the time.

While the report received favourable coverage in right-wing media, it was heavily criticised by some academics and left-wing activists. Writing in the journal Race & Class, Ella Cockbain and Waqas Tufail described it as “shoddy pseudoscience” that “empowered Islamophobes”. They also disputed the “dodgy 84 per cent statistic”.

In 2020, the Home Office published a report titled ‘Group-based Child Sexual Exploitation Characteristics of Offending’. The accompanying literature review largely agreed with Cockbain and Tufail’s criticisms of the Quilliam report. Citing their paper in Race & Class, it stated that Quilliam’s findings are “not suitable for drawing conclusions about ethnicity of group-based [child sexual exploitation] offenders”.

The literature review did cite studies finding an overrepresentation of “Asians” among child sexual exploitation offenders (though typically much less than 84%). Yet based on the quality of evidence, it concluded that “it is not possible to draw any conclusions as to whether some ethnicities have a greater involvement in group-based offending compared with others”.

What’s going on? Are Muslims merely overrepresented among grooming gang offenders, as opposed to comprising the vast majority? Are they not overrepresented at all?

I believe the best available evidence suggests that Quilliam’s original 84% figure is approximately correct. However, this vast overrepresentation is somewhat less shocking than it initially appears. Let me explain.

The “best available evidence” to which I refer is a paper by two academics, Kish Bhatti-Sinclair and Charles Sutcliffe, titled ‘Group Localised Child Sexual Exploitation Offenders: Who and Why?’ It was published in 2020 in Seen and Heard – the journal of Nagalro (a professional association for social workers). The paper is also available on the Social Science Research Network.

To estimate the ethnic composition of grooming gang offenders, Bhatti-Sinclair and Sutcliffe obtained data on all relevant prosecutions between 1997 and 2017 “of which we are aware” from over two thousand media reports. There were an average of over twenty reports per case, and they exhibited “almost complete agreement, often repeating the same quotes and phrases”.

The researchers acknowledge that their reliance on media reports may have led to bias: “less newsworthy trials, possibly those with white British perpetrators, may have been missed; while more sensational trials involving Asian Muslim perpetrators may have been covered more prominently in the media”. On the other hand, some cases involving Muslims may not have been prosecuted “due to fears of accusations of racial bias”.

Bhatti-Sinclair and Sutcliffe identified 73 relevant prosecutions involving 498 accused perpetrators. They determined the accused perpetrators’ ethnicities from their names (a method that is common in medical and demographic research).

Overall, 83% were Muslim – which is almost identical to Quilliam’s figure of 84%. Based on some reasonable, back-of-the-envelope calculations, the researchers conclude: “it is very unlikely that the high proportion of Muslim names identified in our data is due to the exclusion of a large number of non-Muslim offenders”. So even if the true figure’s less than 83%, it’s unlikely to be much less.

Okay, why do I say “this vast overrepresentation is somewhat less shocking than it initially appears”? The reason is that “grooming gangs” is quite a specific category. As Bhatti-Sinclair and Sutcliffe note, “group-based localised child sexual exploitation” (the technical term for grooming gangs) is a subset of child sexual exploitation, which is in turn a subset of child sexual abuse.

Muslims appear to be massively overrepresented among grooming gang offenders, but it’s not clear that they’re overrepresented among child sexual abusers in general. As the researchers note, in the year 2015, “5,879 mostly white offenders were successfully prosecuted for child sexual abuse”. Yet the number of people prosecuted for grooming gang offences over the entire period from 1997 and 2017 was “only” 498 (which is 25 per year, on average). Needless to say, 25 is a small percentage of 5,879.

Now, there are obviously some caveats. One major reason so few people have been prosecuted for grooming gang offences is that authorities often failed to act for fear of appearing “racist”. There would have been more prosecutions, perhaps many more, in the absence of this fear.

Another caveat is that the average number of victims per grooming gang offender may be higher than for other categories of child sexual abuse. As you’ll recall, there were more than 1,400 victims in Rotherham alone. In addition, the level of abuse to which grooming gang victims were subjected may be higher than for other categories of child sexual abuse. Violent gang rape is clearly more serious than, say, distribution of pornographic images.

It’s possible that weighting by number of victims, Muslims are overrepresented among child sexual abusers in general; or that they’re overrepresented among the most serious abusers.

One reason to think they aren’t is that “Asians” (a broad category that includes Britons of Indian, Pakistani and Bangladeshi origin) are underrepresented among those found guilty of sexual activity with minors. According to the latest data from 2017, only 4.6% of such individuals (for whom ethnicity was given) were Asian. Yet as of 2021, Asians make up 9.3% of the population.

Asians were also slightly underrepresented among those found guilty of all sexual offences, making up 7.9% of such individuals (for whom ethnicity was given). Whether ethnicity was less likely to be given for Asian and other non-white offenders, I don’t know.

In any case, the biggest scandal of the whole grooming gang affair is the one Andrew Norfolk highlighted back in 2011: the fact that authorities turned a blind eye to abuse for reasons that basically amount to political correctness; they allowed vulnerable girls to be victimised so that no one would appear “racist”.

Regardless of what a comprehensive analysis would show, we already know that some individuals in England managed to get away with extremely serious crimes purely because of their ethnicity and that of their victims.

Image: Ben Sutherland, Rotherham town centre, 2010

---

Thanks for reading. If you found this newsletter useful, please share it with your friends. And please consider subscribing if you haven’t done so already.