"Every year from 1990 until today, states without the death penalty have had lower murder rates, with an average reduction of about 25%. This appears to show that the death penalty is not an effective deterrent, but correlation does not imply causation. It’s possible that states with higher murder rates are the slowest to ban the death penalty.

A more accurate statistic would be one that looks at the crime rate of a state before and after it abolished the death penalty, i.e. did the crime rate go up?

Not a whole lot of states have abolished the death penalty. Some of them abolished it too long ago for there to be easily-accessible statistics, and others abolished it too recently for the data to be worth anything. The only really prime states are Massachusetts, North Dakota, and the District of Columbia.

This table http://mtgap.bilfo.com/Death%20Penalty%20Stats.pdf contains useful statistics. Data is from http://www.disastercenter.com/crime/ . The table gives the average murder rate from 1960 to 2008, the average murder rate before abolition of the death penalty, and the average murder rate after abolition."

Source: http://mtgap.wordpress.com/2010/07/25/death-penalty-a-look-at-the-numbers/