If capital punishment is, as its supporters claim, a unique deterrent to murder, then a declining use of it reflected in a decreasing rate of executions (which necessarily means an improved chance of escaping it) should be accompanied or followed by an increase in the murder rate. Yet statistics available from many countries, and particularly from the United States, tend to show that this is not the case. These statistics indicate, in fact, that murder rates have either remained constant or declined despite trends away from the use of capital punishment.

Chambliss¹² compared the number of prisoners executed under civil authority and murder rate from 1951 to 1966 in the U.S.A. and found that the substantial decline in executions has not been accompanied by any significant change in the murder rate.

Another study carried out in Ohio⁴⁹ tested the relationship between execution rates and homicide rates. Both rates for the entire State for a half-century (1909-1959) were computed. The statistical correlation did not indicate that homicides have increased as executions have decreased, or the reverse. Any correlation between the two rates seemed to be direct rather than inverse, indicating only that homicide rates and execution rates have risen and fallen together. The researchers concluded that the statistical analysis of Ohio execution and homicide rates over the fifty year period revealed no evidence that executions have any discernible effect on homicide rates.

A third study conducted in Australia by Barber and Wilson⁵ similarly revealed that the relationship between execution and murder rates tend to be a positive rather than a negative one. It was found that the State of Queensland has had a higher execution rate than the other Australian States over a longer period of time (1860-1915) and that the murder rate in that State during the preabeyance period (1901-1914) was also considerably greater than in New South Wales and South Australia. Barber and Wilson concluded that:

"The apparently disproportionately high frequency of executions in Queensland during this period would not, then, seem to have had a very great deterrent effect on potential murderers in Queensland. Indeed the evidence is more supportive of Sir Samuel Romilly's contention that brutal punishments accustom people to brutality, and tend to create attitudes conducive to the commission of violent crimes".