EXECUTION MORATORIUM IS NO HOLIDAY FOR HOMICIDES
by Dale O. Cloninger and Roberto Marchesini
While the Governor of Illinois has declared a moratorium on executions to
study guilt and innocent issues of condemned murderers, reports circulate that
the Governor of Nebraska and the President of the United States are
considering similar moratoriums. Opponents of capital punishment are pushing
for moratoriums as at least temporary reprieves for the death penalty. In
light of these efforts, our current research suggests that the number of
homicides rose as a result of a recent unofficial moratorium on executions in
Texas.
On January 2, 1996 the Texas Court of Criminal Appeals granted a stay of
execution in Davis, ex parte (947 S.W.2d 216; 1996). The effect of this order
was the delay of all but three executions in Texas during 1996. In the prior
three years Texas executed an average of 17 death row inmates per year. The
appeals court lifted its stay on December 18, 1996 and executions resumed on
February 10, 1997. Practicing catch-up, Texas executed 37 inmates in the
remaining months of 1997. Thus, in back-to-back years, the change in actual
executions was twelve fold. This dramatic double reversal in the number of
executions provides an opportunity to compare, in the same population, the
impact of a temporary execution moratorium on the incidence of homicides.
In our study, we developed a statistical model that links changes in the
incidence of Texas homicides with corresponding changes in national homicides
over the immediate preceding "normal" period, 1990-1995. Using this
model, we estimated the number of homicides for each month of 1996 and 1997.
Treating the sudden cessation of executions and the resumption of executions
at double the average rate as "events", we compared the estimated
monthly homicides in the two post-events periods with the actual number of
homicides (after allowing for a 30 day lag).
During the first post-event period, we found that the actual number of
homicides exceeded the predicted number of homicides in nine of the thirteen
months and significantly so in as many as seven of those months. The second
eight month post-event period evinced no pattern save for the first and sixth
months when actual homicides were significantly less than the predicted number
of homicides. Most interestingly, the differences between actual and expected
homicides during the two months immediately prior to the resumption of
executions were significantly positive while in the first month after the
resumption the difference was significantly negative, results that are
consistent with the deterrence hypothesis.
Using the most conservative variant of our model, the actual number of
homicides exceeded the predicted number by 90 over the 21 months of the two
post-event periods. During the first post-event period the actual number of
homicides exceeded the predicted number by 150 while the actual number of
homicides is 60 fewer than predicted in the second post-event period.
Our evidence suggests that as a result of the unofficial moratorium on
executions during most of 1996 and early 1997, Texas experienced a net
increase in the number of homicides over what would have been expected had no
such moratorium been in place. During the interim, there were 40 executions or
three more per year than during the "normal" 1993-1995 period. The
execution hiatus, therefore, appears to have spared few, if any, condemned
prisoners while the citizens of Texas experienced a net 90 additional innocent
lives lost to homicide. Politicians contemplating moratoriums may wish to
consider the possibility that a seemingly innocuous moratorium on executions
could very well come at a heavy cost.
Roberto Marchesini, Professor of Finance
contact marchesini@cl.uh.edu
Dale Cloninger, Professor, Finance and Economics
contact cloninger@cl.uh.edu
The School of Business and Public Administration
University of Houston-Clear Lake
Houston, Texas
The entire study is forthcoming in Applied Economics (probably) later this
year.