Why is data on grooming gangs so bad?

Episode Summary

Executive Summary: This episode examines three fact-checkable claims: misleading missing data in child sexual exploitation ethnicity figures, what uranium enrichment percentages actually mean in nuclear terms, and whether Adam Curtis accurately linked Thatcher’s immigration rhetoric to the 1979 election. It also argues, via Viking history, that “being pillaged” can sometimes stimulate trade through tribute and stationary bandit dynamics.

Main Topics: Child sexual exploitation ethnicity data: The show explains why official ethnicity figures for group-based child sexual exploitation are badly limited by missing data and by how cases are recorded, making headline claims unreliable. What the missing data means: Reporter Josephine Cassidy details how 'perpetrator' is defined early in investigations, why ethnicity is often unknown or unrecorded, and why no one back-fills the data later. More complete local police datasets: Greater Manchester, West Yorkshire, and Rotherham datasets contain fuller suspect ethnicity data and show a high proportion of Asian or Pakistani suspects, though these are not treated as nationally representative. Uranium enrichment and nuclear weapons: A nuclear engineering expert explains the technical difference between low-enriched uranium for power and high-enriched uranium for weapons, and why 60% enrichment is far closer to weapon-grade than civilian fuel. Adam Curtis, Thatcher, and the 1979 election: Sir John Curtis challenges documentary claims that Thatcher was behind in 1979 and made a campaign speech on immigration that won swing voters, arguing the chronology and polling context are wrong. Vikings and the economics of conflict: Duncan Weldon describes how Viking raiding evolved into stationary banditry, with tribute payments and ongoing settlement potentially increasing trade and production rather than purely destroying wealth.

Key Arguments: Official ethnicity statistics on group-based child sexual exploitation are too incomplete to support confident conclusions about perpetrator ethnicity. The 83% white figure is misleading if two-thirds of cases lack ethnicity data; the denominator changes the story dramatically. Many records are created before suspects are identified, so ethnicity is often absent because no arrest, no suspect, or no reliable description exists. The nationally published dataset covers all group-based child sexual exploitation, not just 'grooming gangs,' which may be only about 20% of that total. Local datasets with better recording show very different patterns, including a high share of Asian or Pakistani suspects, but they reflect heavily investigated areas and are not national averages. Uranium must be enriched by separating isotopes; civilian reactors need only low enrichment, while weapons require much higher enrichment, typically around 90%. Iran’s reported 60% enrichment is already much closer to weapons-grade than to natural uranium, because the hardest separation work has already been done. Adam Curtis’s episode overstates the 1979 Thatcher-immigration link: she was not behind in the polls, did not make the cited speech during the campaign, and the key poll shift came later with the winter of discontent. The Viking case suggests coercion can create economic activity when raiders become stationary extortionists who spend tribute locally rather than simply destroying wealth.

Data Points: Missing ethnicity data in group-based child sexual exploitation cases: 66% - England and Wales, 2023 official data discussed in the child sexual exploitation segment White perpetrators in published data: 83% - Reported headline figure before accounting for missing data Asian perpetrators in published data: around 7% - Reported headline figure before accounting for missing data Black perpetrators in published data: 5% - Reported headline figure before accounting for missing data White perpetrators including missing cases: 28% - Recalculated share when all cases are included Asian perpetrators including missing cases: 2% - Recalculated share when all cases are included Black perpetrators including missing cases: 2% - Recalculated share when all cases are included Share of all child sexual abuse accounted for by group-based exploitation: about 5% - The segment stresses this is a small subset of total child sexual abuse Estimated share of group-based exploitation that may be 'grooming gangs': around 20% - Approximate share suggested within the broader category Natural uranium-235 content: 0.7% - Composition of uranium as it comes out of the ground Natural uranium-238 content: 99.3% - Composition of uranium as it comes out of the ground Civilian reactor enrichment: about 4% - Typical enrichment level needed for civil nuclear fuel UK enrichment cited by David Lammy: about 6% to 8% - Comparison made with U.K. sites at Urenco and Sellafield Iran enrichment discussed: 60% - Level cited as unacceptable in the nuclear segment Practical nuclear weapon enrichment: 90% - Approximate weapons-grade level explained by the expert Rough threshold for a nuclear weapon: above 20% - In principle possible, but large and less practical Centrifuge stages for reactor fuel: about 25 - To go from natural uranium to about 4% enrichment Greater Manchester suspect ethnicity data: 54% Asian suspects - High-completeness local dataset compared with 21% Asian population share Greater Manchester Asian population share: 21% - Comparison point for suspect data Rotherham Stovewood suspects: 64% Pakistani - Specific investigation-linked dataset compared with about 4% of Rotherham’s population Rotherham Pakistani population share: around 4% - Comparison point for the Stovewood dataset Conservative polling status in 1979 campaign: ahead by one on average and in all polls but one - Sir John Curtis’s assessment of Thatcher’s position before the election Winter of discontent polling shift: late 1978 to early 1979 - Period when polls moved significantly against Labour Turkey population quiz answer: 85 million - Duncan Weldon’s pop-quiz on country populations South Africa population quiz answer: 63 million - Duncan Weldon’s pop-quiz on country populations Vietnam population quiz answer: 101 million - Duncan Weldon’s pop-quiz on country populations Colombia population quiz answer: 52 million - Duncan Weldon’s pop-quiz on country populations Saudi Arabia population quiz answer: 35 million - Duncan Weldon’s pop-quiz on country populations UK population quiz answer: 68 million - Duncan Weldon’s pop-quiz on country populations

Pivotal Quotes: "For God's sakes, don't half-collect it. That's a bloody disaster." — Baroness Louise Casey: Criticizing incomplete ethnicity data in official child sexual exploitation records "Mrs Thatcher was behind in the 1979 election campaign." — Adam Curtis (on-screen claim cited in the programme): The documentary claim later challenged by Sir John Curtis "I think you know where this is going, but the final statement is that, immediately, she took the lead in the polls. No, I mean, the speech to which reference is made ... is not followed by any significant change in the Conservative position in the opinion polls." — Sir John Curtis: Fact-checking the Thatcher-immigration-polls claim

Implications: Listeners should be skeptical of simplified headline statistics and documentary narratives: missing data can invert conclusions, technical percentages matter in nuclear policy, and historical claims need precise chronology. Local exceptions may be real, but they are not automatically national trends.