Bear with me here. Matt Bruenig has a post on Gawker earnestly making the case that looting and rioting raises the costs of illegal, violent behavior by police and therefore is the economically efficient way to reduce that behavior. He notes that economist Gary Becker’s seminal paper on economics and crime suggests we discourage crime by raising its costs:
https://twitter.com/MattBruenig/status/537351575732518912
However, relying on a colloquy with Cooke (2014), I’ve come to conclude that Bruenig is almost definitely wrong about what rioting does to the costs of aggressive policing.
Here’s Bruenig’s logic:
According to Becker, punishing bad behavior increases the costs of engaging in such behavior and thereby reduces the amount of it. This is the underlying theory of most criminal justice schemes. Rioting that occurs in response to gross police misconduct and criminal system abuses imposes costs on doing those things. It signals to police authorities that they risk this sort of destructive mayhem if they continue on like this. All else equal, this should reduce the amount of police misconduct as criminal justice authorities take precautions to prevent the next Ferguson.
To be sure, burning down AutoZones is not an optimal way to impose costs on state authorities. It would be, as some interviewed Ferguson residents noted, far more effective to target police equipment or other property nearer to criminal justice authorities. But these targets are often difficult and risky to reach, unlike local business interests. Since state authorities are always and everywhere most concerned about capital and business interests, threatening to impose costs on them via rioting should have a similar impact on police incentives.
But a big problem with his case is that net costs of policing do not change, in our situation, the way Bruenig expects. It’s a fairly widely accepted view that “law and order” became a key priority in American politics in the 1960s through the 1990s because of high levels of crime and some specific dramatic instances of urban rioting. Over the long term, citizens get to decide what the government does in response to violence, rather than having to resort to violence themselves, because we live in a democracy. And citizens have chosen more aggressive policing, not less, in response to riots and high crime.
Matt’s theory might have some validity in authoritarian or exceptionally corrupt states (or, not dissimilarly, the Jim Crow South). But the most powerful determinant, in the long run, of the costs of police violence in America is the employers and funders of the police — voting citizens, who choose more policing, not less, when they get rioting.
There are actually lots of other reasons rioting is a terrible deal even if you think police are racist killers (see Collins and Margo 2004a, Collins and Margo 2004b, e.g.), but for now, QED.