Income Inequality and Riots: A Hilariously Unscientific Analysis

Posted on August 13, 2011


Those commenting on the British riots have quickly sorted themselves into two camps, which as usual are roughly arranged to accord with political battle lines. Liberal commenters such as Laurie Penny have insisted that, while the violence of the riots is not to be condoned, it is perhaps understandable as the inevitable lashing out of a generation that has been systematically downtrodden:

Violence is rarely mindless. The politics of a burning building, a smashed-in shop or a young man shot by police may be obscured even to those who lit the rags or fired the gun, but the politics are there. Unquestionably there is far, far more to these riots than the death of Mark Duggan, whose shooting sparked off the unrest on Saturday, when two police cars were set alight after a five-hour vigil at Tottenham police station. A peaceful protest over the death of a man at police hands, in a community where locals have been given every reason to mistrust the forces of law and order, is one sort of political statement. Raiding shops for technology and trainers that cost ten times as much as the benefits you’re no longer entitled to is another. A co-ordinated, viral wave of civil unrest across the poorest boroughs of Britain, with young people coming from across the capital and the country to battle the police, is another.

Conservatives, meanwhile, are insisting that the riots reflect a failure of the British society to instill proper moral values into the teenage rioters. David Cameron’s recent speech in the house of commons is a good example of this attitude:

I have said before that there is a major problem in our society with children growing up not knowing the difference between right and wrong.

This is not about poverty, it’s about culture. A culture that glorifies violence, shows disrespect to authority, and says everything about rights but nothing about responsibilities.

In too many cases, the parents of these children – if they are still around – don’t care where their children are or who they are with, let alone what they are doing.

The potential consequences of neglect and immorality on this scale have been clear for too long, without enough action being taken.

The two sides go back and forth and it is all in a total vacuum of evidence. While I, predictably, fall more on Laurie Penny’s side of the battle, I am unwilling to simply be the umpteenth person regurgitating her point without any further support. So I did a little bit of half-assed digging. I just finished reading The Spirit Level, and have become fascinated by the relationships its authors are able to show between income inequality and social ills. Yesterday on the bus ride home, while reading the spirit level and racking up my wireless bill by watching riot videos on my cell phone, I found myself wondering whether the incidence of rioting, like so many other social ills, might have a relationship with income inequality. Today, I decided to test that relationship:

This was what I was able to come up with in the short time available to me. I used Wikipedia’s list of riots and counted all the riots in wealthy populous countries which occurred since September 11, 2001- a date which I chose because it is an arbitrary start point about ten years ago which arguably constituted the dawn of a new political era. The gini coefficient refers to the distribution of wealth in a society. A gini coefficient of 0 means total equality, or a society in which everybody has the exact same income. A gini coefficient of 100 means perfect inequality, or a society in which one person holds all the wealth. I’ve drawn a line of best fit through the plot, which shows an apparent correlation between income inequality and tendency to riot.

It is plausible that this correlation could reflect a causal relationship. Societies with high levels of income inequality must necessarily have a disenfranchised lower class, and this lower class is unlikely to feel any particular obligation to the society it serves. A large demographic with no sense of connection to larger society is far more likely to take advantage of an opportunity to get free stuff. This analysis, if true, does not justify the actions of the rioters, but it might provide some insight into what kind of social policies can best prevent future riots.

I’ll be the first to admit that this is a horrendously imperfect analysis. Wikipedia, a notoriously bad source, probably did not list all of the riots that have happened in the countries I investigated. I have also not taken the relative severity of the riots into account. My explanation of the correlation is also pretty speculative. But for something I threw together after work on a Friday while drinking a beer, I think this is a starting point for some further inquiry. I might see if I can do something more detailed in this vein sometime in the next few weeks. Ultimately, we need more talented people than me to adopt an evidence-based approach to avoiding future recurrences of these riots, rather than the rhetoric-based approach so favoured by pundits and politicians.

Posted in: Uncategorized