Mr R. Liddle

So Rod Liddle is on Radio 5 Live just now, and he's repeating his claim that 'most' gun crime in London is committed by 'the Black community' (didn't know communities committed crime, but still). In fact he seemed to be suggesting a figure of 85 per cent.

Which sounds pretty unbelievable. And it fact, it is. Although the figures are pretty hard to come by, this (PDF) suggest that in London in 2005, 47 per cent of those proceeded against for gun enabled crime were Black (and 38 per cent White). Which, given the suspicion that Black people are more likely to be proceeded against than White, given the same crime, is probably about parity. So Black people (well, men) are probably over-represented, based on raw population proportions, but not by nearly so much as Liddle claims.

There are a number of other points to make. For one thing, there were 14,250 offences in which firearms were used in England and Wales in 2008/09 (that includes air rifles and imitation firearms). Of these, 38 per cent took place in London (around 5,400). These numbers have been going down in recent years, too (all figures from this PDF). So we're talking about a rather small number of crimes, probably committed in large part by an even smaller number of men - i.e. repeat offenders - of whom less than half were apparently Black. So to blame an entire 'community' seems a bit harsh, to say the least.

The other thing is definition of the denominator. If raw population figures are used then, as I said above, Black people probably are somewhat over-represented in the gun crime data. But surely class and economic status is a confounding factor? If nearly 50 per cent of young Black men are unemployed, is it really surprising that their involvement in crime is somewhat greater? Simply treating 'Black' and 'White' as homogenous blocks means all any subtlety is missed and, essentially, like is not being compared with like. I've not seen any work which looks at this in detail, but someone must have done it.....