Does Capital Punishment Work?

Does capital punishment deter potential murderers? Or are people simply not thinking about the consequences when they commit homicide?

New evidence came to my mailbox last week in the form of the lead article (by Franklin Zimring (Berkeley), Jeffrey Fagan (Columbia) and David Johnson (Hawaii)) in this month's Journal of Empirical Legal Studies.

(You can download a working-paper version of the article here. )

The paper looks at differences in the homicide rates in Hong Kong and Singapore between 1973 and 2008. Notably, Hong Kong had very few executions over this period (and made capital punishment illegal in 1993). Singapore, on the other hand, went from 2-3 executions per year in the 1980s to 21 in 1992 and 76 in 1994, before dropping back down to 17 in 2003 and just 2 in 2007.

This setting is a good one (but of course not perfect) for testing the deterrence effect of capital punishment. Hong Kong and Singapore are similar (though not identical) in terms of demographics and ethnicity. Both are small, urban enclaves with a British colonial history. Both had rapidly growing capitalist economies over this period.

If capital punishment deters homicide, then we might expect rates of homicide in Singapore to fall (relative to those in Hong Kong) around the time that Singapore became so execution-happy. So what do the data say? Homicide rates were slowly falling in both Hong Kong and Singapore throughout the entire sample period. But there's no noticeable reduction in Singapore homicide rates (relative to Hong Kong) at the time that Singapore starting executing a lot of people in the early 90s. Nor is there an increase in the Singapore homicide rate (again relative to Hong Kong) around when Singapore stopped executing folks so much around 2005.

So what does it all mean? It's pretty hard (though not impossible) to reconcile this evidence with the view that capital punishment deters homicide. Proponents of the deterrent effect will need a story for why the Singapore murder rate didn't follow a very different pattern from the Hong Kong rate, given the very different patterns in capital punishment rates.