In a recent issue of Poverty & Public Policy, Jana Morgan and Nathan J Kelly have a really interesting piece about the relationship between crime and support for redistribution in Latin America. Put simply, they argue that people in poor countries who are worried about crime are much more likely to support redistribution, compared to similar people who are less worried about crime. This effect decreases as countries become richer, disappearing at about the level of Argentina.
(The diagram shows the ‘effect of crime threat’ (on support for redistribution) decreases as GDP per capita increases).
They go on to argue:
“Not only do those who perceive crime to be a problem more strongly support a state role in reducing inequality, but those who believe government is effectively increasing their security by combating crime are likewise more supportive of redistribution…Pursuing anticrime policies should encourage public preferences for additional efforts at reducing inequality as well.”
The article is great – it asks an important question and made me think of a load of new ideas about how people decide whether to support redistribution (redistribution here defined as ‘strong policies to reduce inequality between the rich andthe poor’) . They also cite other people’s research that inequality and crime genuinely seem to be related to each other: people’s connections between these ideas have some basis in reality. But I don’t completely agree with their argument….
Confusingly, they find a relationship between crime anxiety and wanting redistribution, but not a relationship between actual crime victimization and redistribution. Rather than being about people’s day-to-day experience of crime, it’s therefore possible that their results are about an entire way of seeing the world – for example:
- Their measure of ‘worry about crime’ is the question “how much do you think that the level of crime that we have now represents a threat to our future well-being?” This is not just a question as to how far people worry about crime, but an ideological statement about the extent to which this is a national threat.
- Their study is cross-sectional, so they can’t look at change over time – do people who are victims of crime become more supportive of redistribution?
- The apparent effect of crime anxiety is enormous – it dwarfs everything else that they look at. At the same time, their only control for ‘ideology’ – a simple self-reported left-right political belief – is non-significant.
Morgan and Kelly describe their study as ‘strong evidence’ for the link between crime worries and support for redistribution. I disagree – and I’m slightly surprised their paper doesn’t mention any limitations at all. But it’s a worthwhile study nonetheless, and reminds us to pay close attention to the determinants of attitudes to inequality – something I’ll return to in my next post.
[A free pre-print version of the article is available here]