Does the Death Penalty Deter Crime?

1
Does the Death Penalty Deter Crime?

• Ben Tyree
• PLSC 374
• Dr. Wang
• 7 December 2005

2
The Theory of Deterrence

• This theory rests upon the belief that nothing
  is so sacred to an individual as his life. The
  prospect of impending death is a threat too
  ominous to be ignored. While one might be
  willing to run the risk of a lesser penalty for
  the sake of achieving his object, he would not be
  willing to risk anything so highly cherished as
  life itself. The death penalty, therefore, is
  upheld as the most powerful of all deterrents,
  and a potent aid in the repression of crime.1

1Bye, Raymond T. Capital Punishment in the United
States. Menasha, WI George Banta Co., 1919.
31-40.
3
Hypotheses

• 1) States with a death penalty statute will have
  lower rates of crimes punishable by death than
  states without death penalty statutes.
• 2) States that have the most executions will
  experience fewer crimes punishable by death than
  states that do not use their death penalty often
  and those without a statute at all.

4
To Execute or Not?

• Death penalty statutes exist in 38 states.
• These states do NOT have the death penalty
• Alaska, Hawaii, Iowa, Maine, Massachusetts,
  Michigan, Minnesota, North Dakota, Rhode Island,
  Vermont, West Virginia, and Wisconsin

5
Murders in the U.S., 1970-2004
6
U.S. Executions Per Year 1976-2004
7
What Crimes are Punishable by Death?

• All states with the death penalty hold that
  first-degree murder is a crime punishable by
  death
• Some states consider other crimes also punishable
  by execution
• Treason, train wrecking, perjury causing
  execution, capital drug trafficking, capital
  sexual battery, kidnapping with bodily injury or
  ransom when the victim dies, hijacking an
  aircraft, aggravated rape

8
Testing Hypothesis 1Case Study One

• States with a death penalty statute will have
  lower rates of crimes punishable by death than
  states without death penalty statutes.
• Texas has executed far more people than any other
  state (336/944), so according to hypothesis 1,
  Texas should have a lower crime rate than a state
  without a death penalty statute, such as
  Michigan.

9
Crime Rates 1970-2000 TX vs. MI(murders per
100,000 inhabitants)
10
Significance of TX vs. MI
    Crime Rates
Death Penalty? Pearson Correlation .391
Death Penalty? Sig. (2-tailed) .166 NOT STATISTICALLY SIGNIFICANT!
Death Penalty? N 14
11
Testing Hypothesis 1Case Study Two

• States with a death penalty statute will have
  lower rates of crimes punishable by death than
  states without death penalty statutes.
• Virginia was second behind Texas in number of
  executions from 1976-2004 (94/944). Again
  following hypothesis 1, Virginia should have a
  lower crime rate than Massachusetts, which does
  not use the death penalty.

12
Crime Rates 1970-2000 VA vs. MA (murders per
100,000 inhabitants)
13
Significance of VA vs. MA
    Crime Rates
Death Penalty? Pearson Correlation .874
Death Penalty? Sig. (2-tailed) .000 STATISTICALLY SIGNIFICANT!
Death Penalty? N 14
Correlation is significant at the 0.01 level
(2-tailed).
The correlation is significant, but not in the
direction hypothesized.
14
Comparing Crime Rate Means of States currently
with and without the Death Penalty in 1970 and
2000

• 1970 (still moratorium on death penalty)
• States without current death penalty
• Mean murders per 100,000 inhabitants 3.9417
• States with current death penalty
• Mean murders per 100,000 inhabitants 7.8395
• In 1970, states that currently have a death
  penalty statute had nearly twice the murder rate
  as those who do not currently have the death
  penalty
• Suggests justification for those states to
  reinstate the penalty after moratorium was lifted
  in 1976
• 2000 (moratorium lifted 24 years before)
• States without current death penalty
• Mean murders per 100,000 inhabitants 2.8250
• States with current death penalty
• Mean murders per 100,000 inhabitants 5.3789

15
Comparing Crime Rate Means of States currently
with and without the Death Penalty in 1970 and
2000 (continued)

• Differences of 30 years
• The 2000 mean murder rate of states that do not
  have the death penalty is 71.7 of the 1970 rate
• The 2000 mean murder rate of states that do have
  the death penalty is 68.4 of the 1970 rate
• States that instituted the death penalty
  experienced a 3.3 larger drop in murder rates
  than states that did not institute the death
  penalty
• Sig. (2-tailed).192
• Not significant

16
Testing Hypothesis 2Case Study One

• States that have the most executions will
  experience fewer crimes punishable by death than
  states that do not use their death penalty often
  and those without a statute at all.
• Missouri executed 61 convicts between 1976 and
  2004. Montana executed 2. Therefore, although
  both have death penalty statutes, Missouri should
  have a lower rate of murders.

17
Crime Rates 1976-2000 MO vs. MT(murders per
100,000 inhabitants)
18
Significance of MO vs. MT
    Crime Rates
Death Penalty? Pearson Correlation .903
Death Penalty? Sig. (2-tailed) .000 STATISTICALLY SIGNIFICANT!
Death Penalty? N 14
Correlation is significant at the 0.01 level
(2-tailed).
The correlation is significant, but not in the
direction hypothesized.
19
Testing Hypothesis 2Case Study Two

• States that have the most executions will
  experience fewer crimes punishable by death than
  states that do not use their death penalty often
  and those without a statute at all.
• Oklahoma executed 75 convicts between 1976-2004.
  Pennsylvania executed 3. Hypothesis 2 contends
  that Oklahoma will have a lower crime rate than
  Pennsylvania.

20
Crime Rates 1976-2004 OK vs. PA(murders per
100,000 inhabitants)
21
Significance of OK vs. PA
    Crime Rates
Death Penalty? Pearson Correlation .667
Death Penalty? Sig. (2-tailed) .000 STATISTICALLY SIGNIFICANT!
Death Penalty? N 14
Correlation is significant at the 0.01 level
(2-tailed).
The correlation is significant, but not in the
direction hypothesized.
22
Comparing 2004 Crime Rate Means of States that
Often Execute Convicts and States that Do Not

• Frequent usage defined as at least one execution
  per year (29)
• Nine states fall into this category
• Mean 2004 Crime Rates for states that frequently
  use the death penalty 6.32 murders per 100,000
  inhabitants
• Mean 2004 Crime Rates for states that do not
  frequently use the death penalty 4.08 murders
  per 100,000 inhabitants
• Sig. (2-tailed).135 (NOT significant)

23
Conclusion

• Both the Texas/Michigan test and the
  Virginia/Massachusetts test REJECTED hypothesis
  1.
• Comparing the crime rate means of all states
  currently with and without the death penalty in
  1970 (under moratorium) and 2000 confirmed the
  results of the first two tests.
• Both the Missouri/Michigan test and the
  Oklahoma/Pennsylvania test REJECTED hypothesis 2.
• States that use the death penalty, on average,
  once per year, do not have lower murder rates
  than those who do not use the death penalty often
  or not at all.

24
Future Research

• David P. Phillips introduces the idea that
  capital punishment may be a deterrent to crime,
  but only in the very short term. Rather than
  analyzing years, Phillips looked at crime rates
  in England on a weekly basis for a 53-year
  period. He found that the week after an
  execution usually had a lower crime rate. This
  would be an interesting, but tedious, study to
  conduct with a wider geographical sample and with
  more recent statistics.