The recent riots in London and the rest of the UK have seen people in respectable jobs prosecuted. This is surprising to most people but we can use game theory to see how people who would not normally riot might end up getting involved and how riots start in the first place.
Someone will only get involved in a riot if they think the gain they get from rioting is more than the penalty (or cost) they might suffer if they are caught.
The benefit of rioting might be a direct material gain from looting, or it could be a social gain if they get more respect from their peers.
The cost to each potential rioter from choosing to riot is the risk of arrest and punishment, or the risk of getting hurt.
The key factor in understanding people’s behavior in riots is to recognize that the cost to each rioter falls as there are more of them. This is because there is less risk of being caught and punished.
Because the cost falls as the number of rioters increases there will be more people join the riot as it gets bigger. A riot will reach a tipping point where more and more people will join rather than the initial trouble fading away quickly. Although the people who join the riot later wouldn’t have started a riot, as the riot gets larger and the cost to them of getting involved falls below their potential gain they will join in. This is why there are people being arrested for being involved who you would not normally expect to get involved. They have been swept along by the emotion of the situation and believed that the chance of them being caught was very small.
Why riots start
It is very unusual for a riot to start, even in a big crowd. When no-one is rioting then the cost to an individual of starting a riot is very high as they will almost certainly be caught. There needs to be a trigger for a number of people to riot at the same time to bring down the cost of rioting to a level where a riot can start.
Usually a riot is started after a significant event pushes a number of people into rioting. In the case of London recently this was the fatal shooting by police of Mark Duggan, in Los Angeles in 1992 it was the acquittal of police officers in the Rodney King case.
These triggers lower the cost of rioting because people want to express their anger regardless of the potential cost to them. Once the riot has started then others will join in because they perceive the risk of being caught to have fallen. Once this starts to happen then the original reason for the riots becomes irrelevant, people are rioting because they feel safe to do so rather than because of the original cause. This is why the London riots spread from Tottenham, where the shooting took place, to other parts of London and then other cities.
The riots have been stopped, at least for the moment, by a huge show of force by the police. This raises the cost of rioting for potential rioters and so puts the vast majority of them off. Once most are put off then the cost to other rioters gets higher and higher as they no longer have safety in numbers so they also don’t riot.
Tomorrow’s post will look at whether the riots might re-start and what conditions will be needed in order for that to happen.
Today’s takeaway: Riots require a trigger to start but once they reach a tipping point they will grow bigger until something raises the cost of getting involved. This applies to lots of group situations, good and bad.