Thursday, January 29, 2009

Economist's Notebook: More Guns, More Crime

Taking a break from the stimulus debate...

The tragic events that occurred in downtown Portland last Saturday night made me think about a very controversial debate to which economists have made major contributions: does gun ownership promote or deter crime?

For sure, someone like Erik Ayala who chooses to take his own life and others' has many means at his disposal.  But easy access to guns (excuse the economics parlance) lowers the cost of doing so.  So, does it matter?  Does a mentally disturbed individual respond to economic incentives?  And does the high incidence of gun ownership increase the cost of crime enough to counter the lower cost of procuring a gun?  

The answers to the particular questions about Ayala are never going to be clear, but what can we say about guns and crime on average?  For sure it is a complicated question to answer because crime and gun ownership are highly correlated: guns are bought to commit crimes, and guns are also bought because crimes are being committed (and guns are bought for many other reasons as well).  How do we say something about the causality of the effect of gun ownership on crime?   

One thing we can't do is look at correlations: the fact that gun ownership and crime are positively correlated tells us nothing about the causal link.  So what economist Mark Duggan did, in his groundbreaking paper "More Guns, More Crime" was to find something that is correlated with gun ownership, but uncorrelated with unexplained variation in crime.  In his case he looks at subscriptions to gun magazines.  This is plausibly correlated with gun ownership and unrelated to unexplained variation in homicide rates.  Using these data to instrument for gun ownership he finds that gun ownership is significantly positively related to the homicide rate - almost exclusively related to homicides committed with a gun.  Ayres and Donohue have also examined the evidence on concealed weapons laws and found that the evidence is mixed, but the bulk of the evidence suggests that, if anything, concealed carry laws increase the incidence of crimes. 

Is the debate over?  No, absolutely not.  But the received empirical evidence gives one pause and suggests that tougher gun ownership laws may indeed reduce gun homicides. 


Jeff Alworth said...

How would you account for data from other countries where gun ownership and murder rates were quite a bit different from the US's? Canada's, for example.

dave said...

think the economics package should include programs to help distribute guns? check it out