Misreading the poverty map

An article (free registration required) in yesterday’s Financial Times (FT) appeared to show a more conclusive link between London’s recent riots and poverty than previous attempts.  The FT analysed unpublished court papers to find the addresses of 300 suspects and then mapped them to their respective areas of London. According to the newspaper:

“Overall, two-thirds of all suspects live in neighbourhoods with below-average income, and only 3 per cent hail from the wealthiest 20 per cent of areas”.

Others have previously discerned a similar link based simply on the correlation between areas of rioting and poverty. The FT took the analysis a step further by showing that suspects arrested in what are probably relatively richer areas, such as Ealing, came from relatively further away.

Unlike some others the FT article was careful not to claim a causal link between poverty and rioting. It merely noted that there was a “link” between them.

But those who are tempted to argue that poverty causes rioting should note at least two key points:

First, poverty, and indeed inequality, are permanent features of capitalism. If they were on their own sufficient to lead to rioting then London would be in a permanent state of chaos (see 11 August post).

Second, the old and often forgotten maxim that correlation is not the same as causation. Just because two factors are linked does not mean that one necessarily causes the other. For example, blond people often have blue eyes but it would be wrong to argue that hair colour causes eye colour. It is only by investigating the science of genetics that the nature of the link can be understood.

Similarly poverty and the recent London riots may be linked but it does not follow that the relationship is causal. It is much more likely that those people detached from society in various ways (see 3 September post) are statistically more likely to riot. But even though such an association almost certainly exists it should not be forgotten that of the more than seven million people who live in workless households only a tiny minority rioted.

Clearly it is necessary to look at society as a whole, for example the breakdown of traditional forms of authority, to get the whole picture.