Is life without parole an effective way to reduce violent crime? An empirical assessment

• Author: Jeremiah Coldsmith,Ross Kleinstuber
• Published date: 01 May 2020
• Date: 01 May 2020
• DOI: http://doi.org/10.1111/1745-9133.12496

DOI: 10.1111/1745-9133.12496

ORIGINAL ARTICLE

CRIME CONTROL AND RECIDIVISM

Is life without parole an effective way to reduce

violent crime? An empirical assessment

Ross Kleinstuber Jeremiah Coldsmith

University of Pittsburgh at Johnstown

Correspondence

RossKleinstuber, Department of Social Sci-

ences,University of Pittsburgh at Johnstown,

450Schoolhouse Road, Johnstown PA 15904.

Email:rkleins@pitt.edu.

Research Summary: By taking advantage of data pub-

lished by the Sentencing Project to analyze whether states

that use life without parole (LWOP) more often experience

lower violent crime rates or greater reductions in violent

crime, this study is the first to empirically assess the crime-

reducing potential of LWOP sentences. The results suggest

that LWOP might produce a small absolute reduction in

violent crime but that it is no more effective than life with

parole.

Policy Implications: Despite reductions in the use of the

death penalty, LWOP has expanded dramatically—and at a

much faster rate—overthe last quarter centur y. This expan-

sion has come at great financial and human costs and has

not been distributed equally throughout the population. As

such, the public policy debate over the use of LWOP is

likely to intensify. Yet, to date, there have been no empiri-

cal assessments of LWOP’s efficacy to inform this debate.

This study begins to fill this gap in our knowledge, and the

results, if replicated, suggest that the use of LWOP should

be either scaled back or eliminated.

KEYWORDS

deterrence, incapacitation, life without parole, sentencing policy, violent

crime

Over the last several decades, life-without-parole (LWOP) sentences have proliferated in the United

States. In 1970, only seven states even authorized LWOP (Nellis, 2013), but by 2016, only Alaska did

not have an explicitLWOP statute (Nellis, 2017). Even more recently,while t he use of the formaldeat h

penalty has been in decline for the past two decades (Death Penalty Information Center,2018), t he size

of the LWOP population has exploded at a rate that farsur passes the decline in death sentences so that

many who were never at risk of an execution are now being sentenced to die in prison (Henry, 2012).

Since 1992, the number of inmates serving LWOP has more than quadrupled (Nellis, 2013, 2017)

Criminology & Public Policy. 2020;19:617–651. wileyonlinelibrary.com/journal/capp © 2020 American Society of Criminology 617

618 KLEINSTUBER AND COLDSMITH

while the overall prison population has only increased by 71% (Carson & Mulako-Wangota, n.d.). In

2016, nearly 1 in every 28 American prisoners was serving LWOP, compared with just 1 in every 71

prisoners in 1992 (Carson & Mulako-Wangota,n.d.; Nellis, 2013, 2017). More people are now serving

LWOP in the United States than there are people serving life sentences of any kind in Europe, which

has more than double the population, and it has been estimated that the U.S. imprisons more than half

of the worldwide LWOPpopulation (van Zyl Smit & Appleton, 2019). There are, obviously, numerous

reasons for this rapid expansion in the use of LWOP, including as an alternative to the death penalty

(Barkow, 2012; Seeds, 2018), but one of the formal, explicitly stated goals of this sanction in at least

some states was to fight crime (Broun & Allison, 2016; Gottschalk, 2014; Robinson, 2012; Spohn,

2014; Tonry, 2016; van Zyl Smit & Appleton, 2019; Yates, 2015). Nonetheless, as the use of LWOP

has grown, several criticisms of LWOP have started to emerge. These sentences are extraordinarily

expensive, especially considering the fact that the costs of incarceration increase dramatically as

inmates age (American Civil Liberties Union, 2012; Human Rights Watch, 2012). They also have

a racially disproportionate impact as the majority of those sentenced to LWOP are people of color

(American Civil Liberties Union, 2014; Capers, 2012; Nellis, 2013, 2017), and because of the amount

of suffering they cause and their threat to human dignity, LWOP sentences have been denounced

by many as a violation of human rights (Capers, 2012; Henry, 2012; Leigey, 2015; van Zyl Smit &

Appleton, 2019; Villaume, 2005).

Given this state of affairs, it would be prudent to ask whether LWOP is even effective at reducing

violent crime. In fact, more than three decades ago, prior to the explosion in the use of LWOP,Cheat-

wood (1988) called for research into LWOP’s efficacy. Yet, this question has never been examined

empirically, and the evidence that does exist is equivocalon this point. On t he one hand, violent crime

rates have plungedsince t he early1990s (Federal Bureau of Investigation, 2017), so it is entirely plausi-

ble that the growth of LWOPhas contr ibuted to this decline. On the other hand, existing criminological

theory and research both provide contradictory expectations as to whether LWOP is likely to reduce

violent crime. Whereas both incapacitation and deterrence theories suggest that increasingly severe

sentences (like LWOP) will reduce criminal offending, both theories also point out that as sentences

get overly harsh, there are likely to be diminishing returns and that at some point, the crime-reducing

power of a punishment will max out. Empirically, there is a robust and consistent finding that the cer-

tainty of punishment matters more than its severity and that increasing already harsh punishments does

not reduce crime (Travis, Western, & Redburn, 2014), but there is considerable debate in the academic

literature over the deterrent effect of the death penalty (Nagin & Pepper, 2012). In many ways, LWOP

is more similar to a death sentence than it is to a lengthy term of years (Berry, 2010; Henry, 2012;

Kleinstuber, Joy, & Mansley, 2016; Leigey, 2015; Villaume, 2005), so it is not clear whether the find-

ings of the wider literature on deterrence or the findings of the death penalty literature are more appli-

cable to LWOP. Therefore, it is crucial to study the specific impact of LWOP to understand what role,

if any, it plays in lowering violent crime rates. This study attempts to do just this by taking advantage

of data collected and published by The Sentencing Project on LWOP sentences to examine whether

states that use LWOP more prolifically experience lower violent crime rates or greater reductions in

violent crime when compared with states that use LWOP less often.

1INCARCERATION AND CRIME: THEORETICAL AND

EMPIRICAL OVERVIEW

The idea that harsher punishment may reduce crime is premised on two different theories: deterrence

and incapacitation. Simply put, deterrence theory argues that the threat of punishment will convince

KLEINSTUBER AND COLDSMITH 619

potential lawbreakers to obey the law in order to avoid the displeasure associated with the punishment

(Beccaria, 1767/1963; Becker, 1968). The concept of incapacitation, meanwhile, makes no claim that

punishment will alter the decision-making of potential offenders. Rather, by locking offendersin pr is-

ons, it sees punishment as making it physically impossible (or at least extremely difficult) for them to

re-offend (Canelo-Cacho, 2002; Levitt, 2004).1According to both approaches, therefore, LWOP sen-

tences should be highly effective at reducing crime. LWOP is one of the most severe sanctions that

can be imposed, which means that it should invoke the most fearin potential offenders, and by making

offenders ineligible for release, it should make their incapacitation permanent.

A few caveats in these approaches, however, cast doubt on the assertion that LWOP should reduce

crime. First, deterrence theory suggests that the certainty of punishment is more important than its

severity (Beccaria, 1767; 1963), which means that using LWOP without increasing the risk of appre-

hension is unlikely to havemuch of an effect on crime. Furthermore, t he concept of marginaldeter rence

suggests that each additional increment of a punishment’sseverity will have less effect than the previous

increment because the additional deterrent value added by making a sentence more severe is necessar-

ily limited by the deterrent value of less severe sanctions (Stigler,1970). In other words, increasing the

severity of punishment will only be able to deter those who havenot already been deter red byless ser i-

ous penalties; thus, as punishment severity increases, there will likely be diminishing returns as there

are fewer and fewer people left to deter (Donohue, 2009; Roeder, Eisen, & Bowling,2015; Travis et al.,

2014). Perhaps even more problematically, by increasing the severity of a punishment, the sanctions

for less serious crimes become so similar to those of more serious crimes that a rationally calculating

potential offender may see no reason not to commit the more serious offense (Beccaria, 1767; 1963;

Shavell, 1992). Given this proposition, it is difficult to imagine that LWOP could be any more effective

than life with parole or even a lengthy term of years at deterring potential offenders.

Similarly, because criminal activity tends to diminish quickly after one’s mid-20s (Hirschi &

Gottfredson, 1983; Siegel, 2015), there are also likely to be diminishing returns on incapacitation as

punishment gets more severe. Once offenders have been incarcerated beyond the ages in which they

are likely to be criminally active, further imprisonment is no longer accomplishing any incapacitation.

Thus, it would again seem unlikely that LWOP could be superior to life with parole or to a decades-

long prison term at incapacitation. Furthermore, because society tends to incarcerate the most severe

offenders first, expanding LWOP to include less and less serious offenders is also likely to have dimin-

ishing returns. As Levitt (2004, p. 179) explained, “the two-millionth criminal imprisoned is likely to

impose a much smaller crime burden on society than the first prisoner.” Therefore, even if LWOP does

achieve some incapacitating effect on the most severe offenders who may have continued to offend

until their deaths, the expansion of LWOP to other offenders is not likely to achieve any additional

crime reduction.

The existing empirical research supports the idea that punishment can reduce crime but that the

certainty of punishment matters more than its severity and that the efficacy of punishment dissipates

rapidly.As Nagin (2013, p. 201) has summarized, “t he perceivedcer tainty of punishment is associated

with reduced self-reported or intended offending.” For example, increasing police resources to increase

the odds of apprehension has been shown to lower crime rates (Evans & Owens, 2007; Levitt, 1997,

2004; Marvell & Moody, 1994) and imposing short jail sentences has been shown to be a more effective

deterrent than probation or fines (Hawken, 2010; Hawken& Kleiman, 2009; Kleiman, 2010; Weisburd,

Einat, & Kowalski,2008). Yet, “there is little evidence that increases in the length of already long prison

sentences” produce any additional deterrent value (Nagin, 2013, p. 201)—a conclusion that has been

reached by “[n]early everyleading sur veyof the deterrence literature” (Travis et al., 2014, p. 90). These

findings would suggest that LWOP is unlikely to impose any additional deterrent value that other harsh,

but slightly less severe, punishments have not already achieved.